MOR A L A I M S

MOR A L A I M S
Essays on the Importance of Getting It R ight
and Practicing Morality with Others

Cheshire Calhoun

1
Oxford University Press is a department of the University of Oxford. It furthers the
Universitys 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
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America

Oxford University Press 2016

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 license,
or under terms agreed with the appropriate reproduction rights organization. Inquiries
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
Library of Congress Cataloging-in-Publication Data
Calhoun, Cheshire.
[Essays. Selections]
Moral aims : essays on the importance of getting it right and practicing morality
with others / Cheshire Calhoun.
p. cm.
Includes bibliographical references and index.
ISBN 9780199328796 (cloth : alk. paper) 1. Ethics. I. Title.
BJ354.C35 2015
170dc23
2015007233

7 9

8 6

Typeset in Arno
Printed on Sheridan 45# Natural 400
Printed by Sheridan

CON TEN TS

Preface
Acknowledgments

vii
xi

Introduction

PA RT I

CR IT IC A L MOR A L IT Y A N D SOCI A L NOR M S

1. Moral Failure
2. An Apology for Moral Shame

27
47

PA RT I I

R E AC H I N G , R E LY I N G O N , A N D C O N T E S T I N G
SOCI A L CONSENSUS ON MOR A L NOR M S

3. The Virtue of Civility

75
V

CONTENTS

4. Common Decency
5. Standing for Something

103
123

PA RT I I I

CON V EN T IONA L I Z E D W RONGDOI NG

6. Kant and Compliance with Conventionalized

Injustice
7. Responsibility and Reproach

157
186

PA RT I V

T E L LI NG MOR A L STOR I E S FOR OT H ER S

8. Emotional Work
9. Changing Ones Heart

213
221

Bibliography
Index

247
255

VI

PR EFACE

Because this is a collection of previously published essays, the reader

might be curious about the point of taking them from their various publishing homes and giving them a single group home. So here is what
moved me to do so.
One reason is the self-interested motive that I suspect is often at work
in single-author collections of essays: the author has some essays that
she thinks are quite good but that were published in venues that resulted
in their being almost never read or read by a narrower audience than
she hoped for. So she takes another stab at getting people to look at the
work by packaging it with pieces that people have read and might want
to read again, or at least have on their bookshelf. There are two essays
in particular that I wanted to recover from obscurity. One is Kant and
Compliance with Conventionalized Injustice, which I assume hardly
anyone has seen. The essay is more timely now than when it was published in 1994, because it is a little exercise in doing non-ideal theory,
and non-ideal theory is hot now in a way it wasnt then. The other essay
that I think is good and would like more people to read is Moral Failure, which Ive now positioned as the lead chapter so it cant be missed.
There I am preoccupied with making a claim that I think bears thinking
more about, namely that the social practice of morality really is morality
and not just a sociological phenomenon.

VII

PR EFAC E

A second motive is more philosophical. I have an ongoing interest,

acquired from my father, in seeing what happens conceptually when
essays that were written independently are brought into one volume and
read together. This was the (or a) motive that led me put together Setting
the Moral Compass: Essays by Women Philosophers. I wanted to see what
happens to our conception of feminist philosophy and to the idea that
philosophy has no gender if we read, side by side, essays by moral philosophers who are all women but who arent all setting out to do feminist
philosophy. Something similar is going on in this volume. I am an essayist by nature, and all of the chapters in this volume were written as independent, topical essays. The few essays that do have a steady readership,
such as Standing for Something and Changing Ones Heart, are, I
suspect, virtually always read as topical essays on integrity and forgiveness. As a result, the themes that I found myself continually returning
to in the essays gathered in this book arent visible. Those themes, about
which Ill have more to say in the introduction, include the significance
of the social practice of morality for what we say as moral theorists, the
plurality of moral aims that agents are trying to realize and that sometimes come into tension, and the special difficulties that conventionalized wrongdoing poses. The essays that bookend this volume explore
another theme: the opacity of what individuals are up to when they
engage in what clearly appears to be wrongdoing. This is a book about
those themes. The introduction, the organization of the essays, and the
introductions to the four parts of the book are all designed to disrupt
efforts to read these essays as independent topics.
I could not have written these essays without the emergence of
feminist philosophy in the 1980s and 90s and without the pioneering
work of people like Annette Baier, Claudia Card, and especially Margaret Walker. Reading feminist philosophy taught me to pay attention to
social norms and the difficulty of thinking clearly and acting well under
conditions of dominance and subordination. It taught me to attend and
be sympathetic to the vulnerabilities of moral persons and the moral
binds they find themselves in. And above all, it taught me to do philosophy that is both deeply personal and begins from the question: What are
moral philosophers not talking about? I suspect these essays also could
not have been written without the work of Tom Hill, who taught me the
VIII

PR EFAC E

importance of having and displaying moral attitudes, and by extension,

of paying closer attention to the interior lives of moral agents. The work
of Bernard Williams figures prominently in chapters 2 and 5. My critical
tone in those chapters does not adequately convey my deep admiration
for his unique abilities to hone in on important, but neglected, moral
phenomena and to think outside the philosophical box.
My thanks to Peter Ohlin, senior editor for philosophy at Oxford
University Press. His continued supportive interest in my work and his
patience with my extraordinarily slow pace have been much appreciated.

IX

ACK NOW LEDGM EN TS

The chapters in this book were published in various venues between

1989 and 2004. Minor editorial changes have been made to the original published versions to ensure consistency of style throughout this
book. There are a few new footnotes, and introductory thank-you notes
to various audiences, institutions, and individuals have been removed,
although footnote credits for specific points have been retained. I thank
the following for permission to reprint the essays in this collection:
Chapter 1: Moral Failure originally appeared in On Feminist
Ethics and Politics, edited by Claudia Card (University of Kansas
Press, 1999), pp. 8199. 1999 by The University Press of Kansas.
Chapter 2: An Apology for Moral Shame appeared in Journal of
Political Philosophy 12, no. 2 (2004): 127146. 2004 John Wiley
& Sons Ltd.
Chapter 3: The Virtue of Civility originally appeared in Philosophy and Public Affairs 29, no. 3 (2000): 251275. 2000 John
Wiley & Sons Ltd.
Chapter 4: Common Decency was originally written for a collection that I edited, Setting the Moral Compass: Essays by Women
Moral Philosophers (Oxford University Press, 2004), pp. 128142.

XI

ACK NOW LEDGM EN TS

Chapter 5: Standing for Something originally appeared in The

Journal of Philosophy 92, no. 5 (1995): 235260. 1995 The Journal of Philosophy.
Chapter 6: Kant and Compliance with Conventionalized Injustice originally appeared in The Southern Journal of Philosophy 32
(1994): 135159. 1994 John Wiley & Sons Ltd.
Chapter 7: Responsibility & Reproach originally appeared in
Ethics 99 (January 1989): 389406. 1989 by The University of
Chicago. All rights reserved.
Chapter 8: Emotional Work was written for Explorations in
Feminist Ethics: Theory and Practice, edited by Eve Browning Cole
and Susan Coultrap-McQuin (Indiana University Press, 1992),
pp. 117122. Reprinted courtesy of Indiana University Press. All
rights reserved.
Chapter 9: Changing Ones Heart originally appeared in Ethics
103 (October 1992): 7696. 1992 by The University of Chicago.
All rights reserved.

XII

Introduction
The aim of normative theorizing is to establish principled ways of thinking about and reaching answers to normative questions. The most important normative question is, of course: What ought I to do? But
there are other important normative questions as well. When is blame
warranted? requires some account of the criteria for moral responsibility and for excuse from responsibility. Related to such questions about
responsibility are normative questions about the reactive attitudes:
When is it appropriate to feel each of the self-directed reactive attitudes
of shame, guilt, pride, and the like? and When are other-directed attitudes such as resentment and forgiveness warranted? Normative
theorizing should also help us to determine which character traits are
virtues, which are vices, and in what those virtues and vices consist.
In short, the task of normative theorizing is to give moral agents guidance, at the most general level, about what to do, what kind of people to
become, how to feel about their own and others moral performance,
and when to let themselves and others off the moral hook.
In addition to providing guidance, normative theorizing has a
second, equally if not more important function: to enable moral agents
to engage in critical moral reflection and thus to become autonomous
from others opinions and shared social views on normative questions.
One of the central moral tasks of agents is to get it rightto latch on to
the correct normative principles and apply them correctly, to grasp what
traits are really virtuous or vicious and not merely thought so, to figure
out what they really should feel ashamed or guilty for rather than just
what they are made to feel ashamed or guilty for. Normative theorizing,
and the critical reflective point of view it supports, supply the necessary
wedge between what is commonly taken to be so and what really is so.

INTRODUCTION

In going about this work, how, at a most basic level, is the normative moral theorist thinking about morality? To answer that, suppose
we begin by reflecting on the phrase morality requires... , which appears often enough in moral theorizing. Thinking about what morality
requires is like thinking about what justice requires. We are tempted
in both cases to capitalize: it is Morality and Justice that require. The
capitalization draws attention to the fact that we are not thinking about
what people think morality or justice requires, nor are we thinking about
culturally local standards. We are thinking about real or genuine, as opposed to supposed, moral requirements. The merely supposed moral requirements that provide the contrast might be those proposed by some
moral theorist. Kant thought that morality requires exceptionless truthtelling, but (we think) he was wrong; morality does not require that. Or
the merely supposed moral requirements might be those that a social
group accepts as moral requirements. At one time in our not so distant
past, women were thought to be under the moral obligation to take their
husbands last name; this was a duty of love, loyalty, respect for husbandly authority. But people were wrong; morality does not require this
of women who marry.
As these remarks suggest, one central conception of morality is
brought into sharpest focus by contrasting moral requirements with
what people have mistakenly supposed those requirements to be. That
contrast might then naturally lead to two additional thoughts. First, the
work of the normative moral theorist, as I suggested above, is to help us
get it rightto latch onto what morality actually requiresas well as to
help us see why morality requires these things, or to latch onto the correct justification of moral requirements. Successfully doing this work
will establish when the various reactive attitudesfor example, of resentment, gratitude, and forgivenessare appropriate and justified. It
will also help to establish what actions are blameworthy, and thus when
actually blaming other people is justified.
The second thought that the contrast between real and supposed
moral requirements quickly leads to is that normative moral theorists
can conduct their work only by going hypothetical. As the contrast between what morality actually requires and what people think it requires
reminds us, there is no guarantee that the set of moral requirements the
2

INTRODUCTION

normative moral theorist helps us to latch onto will match what real
people, operating within a social practice of morality that they accept,
will see as moral requirements. So the system of capital-M moral requirements will be, to varying degrees, hypothetical in the sense that we
are imagining the moral norms that would effectively operate as requirements in a hypothetical social world that endorsed those norms.
By effectively operate as requirements, I mean that it would generally be the case that individuals within that social world would do
such things as address moral demands and complaints to each other;
feel and express resentment, blame, guilt, and pressures of conscience;
and offer excuses and justifications predicated on the acceptance of the
legitimacy of those (correct) moral norms. Although we say morality requires, morality is not a person who can issue and enforce commands. Whatever moral requiring gets done will be done by people
who require things of each other and of themselves, and whose requiring activities stand a good chance of uptake because the targets of those
activities believe the demands are legitimate, or they at least find them
intelligible. Thus talk about moral requirements makes most sense if
we have in mind some social world where the requiring activity is to
take place and receive uptake. If we think that what actual people are
morally requiring of each other is misguided and want to talk about
genuine moral requirements, then we will have to understand those
requirements by thinking about the requiring activity and uptake of
that activity that would hypothetically take place in a counterfactual
social world.
The contrast between real and supposed moral requirements also
encourages the moral theorist to go hypothetical in a different way. Suppose we think that morality is a system of rules that is endorsable by
all because each has good reason to accept those rules. What we mean
by endorsable and having reason cannot be what particular, socially
located individuals find endorsable, nor can the good reasons be limited
to the reasons that those particular, located individuals are presently capable of recognizing as reasons. Genuine moral requirements will have
to be conceived as ones that are endorsable by all within a hypothetical
social world populated by people who are capable of accessing the good
reasons there are to endorse those requirements.
3

INTRODUCTION

Going hypothetical is not the same as going ideal. One of the things
we might be concerned with as moral theorists is what morality requires
under nonideal conditions where people have misguided moral conceptions and wrongdoing is conventionalized. To figure that out, we will
have to pay attention to facts about the actual social practice of morality
and the larger social world in which it takes placefacts that will include the systematic failure to do what morality requires and systematic
misconceptions that morality requires or permits something that in fact
it forbids.
Despite the need to pay attention to social facts, answers to questions about what morality requires under nonideal conditions may still
involve going hypothetical in both of the senses I just mentioned.1 Suppose, for example, you think that the best way to produce racial equality
is through integration and that morality requires taking steps to bring
about that integration. 2 Given that racial segregation in social life is
thoroughly normalized, and that a point of pursuing integration is to
effect conditions under which people will be less able to sustain their
prejudices, to think about integration as being morally required will involve thinking about a hypothetical social world where the requiring
activities of agents was guided by just that imperative. And in thinking
that this integration rule is justified, one would have to think about its
endorsability in a hypothetical social world of people who are capable
of accessing the good reasons for integrationreasons that many will
have in the real social world only after integration is achieved.
So far, Ive presented this conception of moralityas genuine rather
than supposed moralityas the conception employed by normative
moral theorists in order to do their work. But, of course, it isnt just employed by moral theorists. Its the conception that people use whenever
they stand back from their own social practice of morality and look with
a critical, reflective eye at what is actually being morally required (or
permitted, or forbidden, or recommended) within that social practice.
1. Chapter6, this volume, is an instance of nonideal theorizing that goes hypothetical in
both senses.
2. Elizabeth Anderson, The Imperative of Integration (Princeton, NJ: Princeton University
Press, 2010).

INTRODUCTION

A gap may then appear between what ones fellows generally endorse
as a correct moral norm and what one thinks the moral norm ought to
be. Or at least a potential gap appears. One wonders, Is this generally accepted moral norm really correct? Appreciating the gap, or potential gap,
between socially accepted moral norms and correct moral norms may
then lead one to think about the whole of social morality in a particular
way. Social morality isnt really morality. It is a set of social norms that
individuals in a social world generally regard and treat as moral requirements. Social norms, however, even when they coincide with the (genuine) requirements of morality, are still just thatsocial norms. Morality
consists in the set of genuine moral requirements (permissions, prohibitions, recommendations), which may or may not also be embodied in
the requirements of social norms.
To clarify this point: the features that make a norm a social moral
norm differ from the features that make a norm a genuine moral prescription. Most significant, answers to normative questions within a social
practice of morality get settled by achieving social agreement. And
reaching that agreement will occur through social processes of moral
dispute in which individuals stand up for their own views on moral matters without being entitled to claim authority to have settled the question, no matter how good their arguments are. In a social practice of
morality, it is collectives of people that settle moral disputes, because
what is to be established are shared understandings about what can morally be expected of fellow participants. Among the shared understandings are broad agreements on which moral matters are still reasonably
under dispute and which may reasonably be taken to have been resolved.
At one time in U.S. history, for example, the equality of the races was a
matter of social dispute. Today it is not, even if some individuals continue to believe in racial inequality.
By contrast, moral philosophers and participants in a social practice who take up a critically reflective point of view settle disputes by
making up their individual minds about what views are best supported
by reasons. Of course it may be critical to consider what views would be
endorsable by all (not just by oneself) in a hypothetical world of people
who have access to the good reasons for endorsement. Even so, it is the
goodness of the reasons, not the fact of actual agreement, that is doing
5

INTRODUCTION

the work. From a critical point of view, what settles a normative question
is not social agreement but the correctness of the justificatory argument.
It is this fundamental difference in how moral questions are settled that
makes social moral norms distinctively different from genuine moral
prescriptions, even when the two have the same content.
To summarize this first conception of morality: we reach it by taking
up a critical, reflective point of view where our concern is with getting it
rightthat is, with latching on to those moral rules that would be the
focus of the requiring activities of persons in a hypothetical social world
populated by hypothetical agents who are capable of accessing the reasons there are for everyones endorsing just this set of rules, where those
rules differ in kind from social norms.
None of the remarks so far have been meant to be controversial.
Indeed, they might have the air of the obvious. The questions that interests me are: What is the relation of the actual social practice of morality to the capital-M conception of morality just sketched? and To
what extent is reflection on the content of the social practice of morality
a proper part of the work of a normative moral theorist?
By social practice of morality, I have in mind a number of things.
Most obviously, a collection of people who practice morality together
will develop what Margaret Walker calls shared moral understandings
about what the moral norms are, how they are to be applied, what excuses are acceptable, who counts as a responsible agent, and the like. 3 A
social practice of morality settles the same range of questions that moral
philosophers aim to settle when they engage in normative theorizing or
that individuals aim to settle when they take up a critical, reflective point
of view. A social practice of morality will also involve its participants in
various requiring activities, such as demanding, blaming, shaming, and
exhorting.
In thinking about the relation of the actual social practice of morality to capital-M morality, the first thing one might note is that there is a
danger of exaggerating the extent to which the critical, reflective point
of view, which aims to latch onto genuine morality, escapes the point of
3. Margaret Urban Walker, Moral Understandings: A Feminist Study in Ethics 2nd ed. (New
York: Oxford University Press, 2007).

INTRODUCTION

view of the social practitioner of morality and thus escapes dependence

on the available shared moral understandings. Our efforts to get it right
are always limited by, and draw on, the resources provided by socially
available concepts and methods of moral reasoning. The concepts of
moral obligation and the moral equality of persons havent always been
around, nor have consequentialist and Kantian methods of moral reasoning. If we think that our predecessors were limited in their critical,
reflective thought by the unavailability of these concepts and methods,
we should also think that we, too, may be similarly limited in our efforts to get it right by the unavailability of concepts and methods that
our successors may have. In addition, moral philosophers appeal to
intuitionssomething that it is virtually impossible to imagine doing
entirely withoutis at bottom an appeal to shared moral understandings that have emerged from a social practice of morality.4 So here,
again, critical, reflective thinking will not be entirely freed from ties to
the social practice of morality. All of this is just to say that knowledge
is always socially situated, even when we are trying our best to escape
that social situatedness and gain critical purchase on what we think we
know, and that social consensus on some elements of morality will be
important to the conduct of critical reflection.
There is an accompanying danger, too, of exaggerating the extent to
which critique is not part of the social practice of morality. While there
would not be a social practice of morality were there not broad consensus on what the moral norms are and the reasons for endorsing them,
social practices of morality are also sites of contest and negotiation over
the acceptability of particular norms, conceptions of the virtues, standards of accountability, and so on. This will be particularly so in complex societies where individuals move among a variety of social practices
of moralitywork, family, religion, and so onand in societies that acknowledge diversity of moral view and value the open exchange of ideas.
A critical moral point of view is relevant not only to practitioners decisions about when to contest elements of social morality in an effort to
4. Margaret Urban Walker pointed this out in Moral Understandings: A Feminist Study in
Ethics (New York: Routledge, 1998), 6668. The advent of experimental philosophy has
brought the point to prominence among philosophers.

INTRODUCTION

transform them; it is also relevant to their decisions to comply with and

enforce social moral requirements. Social morality has the stability it
does in part because most people think that the moral norms on which
there is social consensus are in fact the correct moral norms and are justifiable from a critical, reflective point of view.
So the distinction between supposed and genuine morality, which
seems central to understanding the work of the normative moral theorist, is also central to the position that participants in a social practice
of morality can find themselves in. On the one hand, participants know
what the social moral norms are; what they will be subjected to social
shame and reproach for; how they can (and cannot) successfully communicate important moral attitudes like respect and toleration; which
moral positions others will find disputable and which they will not; who
counts as a competent, well-formed, responsible moral agent; and the
like. On the other hand, they are well aware that social moral norms,
shaming and reproaching practices, some socially prominent arguments, and so on may be misguided. That is, participants in a social practice of morality are (or know that they ought to be) also critical moral
thinkers about that very practice. As critical moral thinkers, they are
positioned both to describe what the social practice of morality is and to
engage in normative reflection on what it ought to be. To this extent, the
familiar distinction between a social practice to be described and a critical normative theory to be applied captures something about the social
practice of morality itself: its practitioners are outsiders withinboth
insiders to a shared moral practice and critical outsiders to those social
norms they reject. In short, within the social practice of morality itself,
we occupy the double role of being participants in the practice of morality and of being autonomous individuals who aim to latch onto the
correct moral principles even if these are socially unacknowledged or
rejected.
What about our question, To what extent is reflection on the content of the social practice of morality a proper part of the work of the
normative moral theorist? Given the distinction between supposed
(social) morality, on the one hand, and genuine morality, on the other,
it would seem that reflection on the social practice of morality will, at
most, make an indirect contribution to the proper work of the normative
8

INTRODUCTION

moral theorist by suggesting which questions need to be addressed.

There will, for example, likely be lacuna in the set of shared moral norms.
New social arrangements, technologies, and biomedical advances may
all raise moral questions that shared moral understandings are illequipped to address. For example, who is responsible for rectifying climate change? Who has an ownership claim to tissue samples acquired
during medical procedures? There will also be periods where shared
moral understandings are shifting under the pressure of widespread
contestation that focuses attention on the question of what our shared
moral understandings should be. The charge leveled by some social conservatives against gays and lesbians, for example, that their demands
for coverage by antidiscrimination laws amounts to claims for special
rights, invited us to get clearer about the conditions under which a social
group has a legitimate claim to legal protection against discrimination
and when it does not. There are also likely to be at least some ways in
which injustice and wrongdoing are socially conventionalized and thus
rendered both invisible from within the social practice of morality and
difficult to change. Since the normative moral theorist is also embedded in social practices of morality, the conventionalization of wrongdoing and injustice raises questions about how normative theorists can
avoid their own blind spots; and the normalization of wrongdoing raises
questions about how best to effect changes in social understandings and
practices. The work of the normative theorist in these cases is thus to
recommend supplementary norms to cover new circumstances, to adjudicate social dispute over what the correct understandings are, and to
reveal and critique widely shared moral understandings.
In short, it would seem that the social practice of morality can
appear in normative theory only as kind of datum to be described and as
that for which we need to develop correct normative principles. Features
of the social practice of morality may suggest questions for our normative theories to address, including questions about what conceptual
and methodological tools are needed to carry off these tasks of supplementing, adjudicating, and critiquing. But neither the fact that we actually share moral understandings in a social practice of morality nor
the content of those understandings has any importance independent
of the correctness of those understandings. What matters are the shared
9

INTRODUCTION

understandings that people would reach were they able to access the reasons there are. In short, what has significance is the hypothetical social
practice of morality that is governed entirely by correct moral norms.
Thus it is social practice conceived as ideal and goal that has significance.
The actual sharing of moral understandings (regardless of the correctness of their content), the requiring activities and reactive attitudes that
shared understandings make possible, and the social processes involved
in collective settling upon and contestation of shared understandings
have no moral importance in themselves.
It might seem that that is exactly the right conclusion to reach. If
morality concerns what we ought to do, require, and feel, then getting it right is what matters. Actually sharing moral understandings in
a real-world practice of morality has no independent importance. But
notice an implication of this view. That we actually do share moral understandings is, as I suggested earlier, what makes effective requiring
activities, and thus moral requirements, possible. Those requiring activities include the addressing of moral demands, expressing resentment,
reproaching and blaming, and relatedly, offering excuses and justifications in response to the demands, resentments, reproachings, and blamings. The success of these requiring activities depends on the existence
of shared understandings about their bases. I cannot effectively make a
moral demand on you that is not intelligible to you as a moral demand.
This does not mean you must agree with me. So long as the demand is
intelligible as the sort of demand people do make, I put you in the position of needing to respond to that demand with either an excuse, or a
justification, or a defense of an alternative view. There is a real sense in
which the social practice of morality, whatever its imperfections may be,
is the only moral game in town.
Given this, one might wonder whether the conception of capital-M
morality that led us to assign no independent importance to the social
practice of morality is the right one. One of the effects of focusing on the
task of getting it right, and thereby latching onto genuine as opposed
to merely supposed morality, is to focus our attention on what moral
philosophers do: develop the tools for correct moral deliberation (for
example, versions of Kantian or consequentialist theories), specify what
particular virtues consist in, work out accounts of what responsibility
10

INTRODUCTION

consists in and under what conditions individuals can be held accountable, argue for particular normative principles, and the like. Morality
appears to be centrally a matter of moral knowledge, as Margaret Walker
stresses in her extraordinary and oddly neglected (outside feminist circles) book Moral Understandings. Walker dubs this knowledge-oriented
conception of morality the theoretical-juridical model of ethics. In her
words, it is a conception on which Moral philosophy has as its central
aim the discovery/construction, testing, comparison, and refinement of
moral theories... which exhibit the essential core of a pure or proper
moral knowledge, in distinction from merely collateral practical, psychological, sociological, anthropological, historical, etc. (i.e., merely
factual, nonmoral) information.5
Walker pursues a number of important critiques of this model both
of morality and of moral philosophy. But what interests me here is the
alternative conception of morality she proposes, which she calls the
expressive-collaborative view. That view
supplies instead the picture of morality as social negotiation in
real time, where members of a community of roughly or largely
shared moral belief try to refine understandings, extend consensus, and eliminate conflict among themselves. We are the members of some actual moral community, motivated by the aim of
going on together, preserving or building self-and mutual understanding in moral terms. 6

This is, in her view, what morality really is. Morality on this conception, we might say, is not only about the shared moral understandings
and the social processes that produce, contest, and refine them. It is also
an actual, rather than hypothetical, scheme of social cooperation that
is enacted in real time. And morality is, perhaps most importantly, embodied not in what people know but in what they do and feel in light of
that knowledge. It is a matter of how we actually treat each other, the
things we do to hold each other to account, and the moral sentiments
5. Walker, Moral Understandings, 37.
6. Ibid., 6465.

11

INTRODUCTION

we feel and express. It is also a matter of the moral identities we actually

come to have for others, not merely our private self-conception of our
character traits. If we ask ourselves why we care about acquiring moral
knowledgegetting it rightit is not for the sake of simply having that
knowledge (of having an action guide in our back pocket, as it were). It
is because we live and interact with other people and we need to find a
collective way of doing this well.
Like Walker, Barbara Herman emphasizes that A reasonable morality is well-integrated into ordinary living, not something we are
endlessly at war with (like a diet), nor a distant goal toward which we
direct substantial amounts of energy. 7 On the contrary, in her view, a
central function of morality is to secure routine action in everyday life.
That function is possible only if there is an existent local practice of morality into which agents are morally educated, resulting in moral values
being seamlessly integrated into the structure of desire and perception being attuned to morally relevant facts. Routine moral performance
depends on individuals acquiring moral literacy in a social practicea
moral intelligence that can read and respond to moral facts, incorporating their evaluative import into a shared way of life.8 One expects
the normal moral adult to be morally confident about acting in everyday
life, a confidence that depends on her experiencing no profound rupture between her own moral sensibility and the moral norms that govern
her social world.9
The conception of morality I sketched earlier is not entirely divorced
from the idea that morality is centrally a shared cooperative scheme.
Nor is it entirely divorced from the idea that participants in that scheme
must be able to see reason to endorse it. But as I said earlier, the aim
of getting it right forces us to go hypothetical about that cooperative
scheme, the participants in it, and their process of negotiating an agreement. That move to the hypothetical, however necessary it may be to
7. Barbara Herman, Morality and Everyday Life, Proceedings and Addresses of the American Philosophical Association 74, no. 2 (2000): 2945, 31.
8. Barbara Herman, Moral Literacy. The Tanner Lectures on Human Values. Delivered at
Stanford University April 2324, 1997, 372. <http://tannerlectures.utah.edu/lecturelibrary.php#g>
9. Herman, Morality and Everyday Life, 37.

12

INTRODUCTION

our envisioning an ideal practice of morality, has the unfortunate side

effect of obscuring an important point: there are no shared cooperative schemes, no treatments of others, no moral identities, no reactive
attitudes that effectively hold people responsibleother than in actual
social practices of morality. The social practice of morality is, to repeat,
the only game in town. Absent a social practice, there is no morality, although there might be moral knowledge. This is why I have been pressing the oddity of thinking that actual shared understandings, actual
processes of negotiating agreements, and actual requiring activities are
not significant in their own right.
What Walker and Herman propose, and what is central to this book,
is a second conception of what genuine, capital-M morality is. It is a conception shaped not by the aim of getting it right or the contrast between
correct and supposed moral requirements. It is a conception shaped by
the moral aim of engaging with each other on shared terms. It is also
shaped by a different contrastnot that between correct and supposed
but, rather, the contrast between actual and hypothetical social practices. If practices based on misguided moral norms seem not to be genuine morality under the first conception, merely hypothetical practices
seem not to be the genuine article under the second conception.
To summarize this second conception of morality as I see it: we rely
on this conception of morality insofar as we are practitioners in an actual
social practice of morality where our concern is with making what we
are morally up to and who we morally are intelligible to others, reaching
shared moral understandings, and communicating moral attitudes and
demands to co-participants.
Return now to the questions I posed earlier: What is the relation
of the actual social practice of morality to capital-M morality? and To
what extent is reflection on the content of the social practice of morality
a proper part of the work of a normative moral theorist? The response
suggested by this second conception of morality is that the social practice of morality just is capital-M morality. It is what has claim to the
title of being genuine morality. The hypothetical practice of morality
governed by hypothetical agreements is not morality at all. This is, as
I understand her view, Walkers position. Morality needs to be seen as
something existing, however imperfectly, in real human social spaces in
13

INTRODUCTION

real time, not something ideal or noumenal in character.10 Reflection

on the content of the social practice of morality is what normative moral
theorizing should be about. The theorist is not to begin by ignoring
actual social practices of morality, including those that shape the theorists own thought, in order to construct an ideal normative standard to
then be applied in evaluating actual practices. The task, on Walkers view,
is to start with a rich, empirically informed description of some actual
social practice and then test whether moral understandings really are
intelligible and coherent to those who enact them, whether they are similarly so from diverse points of view within them, and whether they are
the kinds of understandings that can be so.11
How are we to choose between these two conceptions of capital-M
morality? The difficulty is that neither by itself seems entirely satisfactory. Consider, first, the unhappy implications of the first conception.
If we insist that what matters is determination of and action on the correct moral standards, then our advice to all individual agents must be
to focus on getting it right (in thought and action). Individuals are not
barred from caring about what others think; indeed, if they care about
getting it right, then they ought to care that others do so as well and so
ought to be interested in persuading others to the correct views. They
are also not barred from factoring in others moral misconceptions in
deciding how to proceed; indeed, getting it right extends to getting it
right under nonideal conditions. But there will be limits to how far one
can accommodate others different moral views and still claim to aim
at getting it right. At some point, commitment to getting it right will
entail being willing to incur others incomprehension, contempt, resentment, unwillingness to stay in dialogue, and the like. In particular, one
must be prepared to buy out of the social practice of morality and do
and think about things differently from others. Because the injunction
to get it right applies to all agents, one must also be prepared for others to
buy out of the social practice of morality at whatever junctures it seems
to them that capital-M morality requires this. At the limit, one must be
prepared for there to be no social practice of morality.
10. Ibid., 18.
11. Walker, Moral Understandings, 11.

14

INTRODUCTION

Being prepared to buy out of the social practice of morality when

one thinks it wrong may seem admirably high-minded. It will seem
especially so if one focuses on cases where a largely well-functioning
social practice of morality seems obviously misguided on some limited
dimension. You might, for example, think that our social practice, while
getting many features of morality right, is misguided on the question
of whether it is permissible to purchase meat in grocery stores. Buying
out of the social practice of morality at this particular juncture may,
indeed, be admirably high-minded. But notice how focusing critically
on isolated imperfections inand opportunities for high-mindedness
in response toa social practice of morality allows us to continue tacitly
taking for granted the importance of there being a larger, background
social practice of morality and to evade thinking about how much we
depend on it in order to do much of anything connected with morality
at all: make demands, offer counsel, present excuses, justify our choices,
express moral attitudes to others, have an identity as a kind or honest
person, get uptake on our resentments, and the like. In addition, because
such criticisms typically proceed on the assumption that (some) others
share our objections to this feature of our social practice, we are able to
evade thinking about what a loss to ones moral life it would be were one
unable to find others who agree with, or at least find intelligible, ones
objections and alternative ways of living.
What I am suggesting here is that socially critical normative theorizing that assumes genuine morality consists in the correct set of norms
is guilty of taking out an unacknowledged loan on the social practice
of morality. It simply takes for granted the importance of there being
an actual social practice of moralitycomplete with shared moral
understandings that enable moral theorists to make their criticisms
intelligiblewhile at the same time working with a conception of
capital-M morality that rejects the importance of the (imperfect) actual
in favor of the (correct) hypothetical.
Annette Baier was the first to observe how normative theorizing
takes out unacknowledged loans.12 She argued that contractarian moral
12. I take the idea that moral theory has to take out a loan from Annette Baier, What Do
Women Want in a Moral Theory? Nous 19, no. 1 (1985): 5363.

15

INTRODUCTION

theory has to take out a loan on the loving parental care and socialization of children that produce people who are willing to trust other
people, to take morality seriously, and thus to contract with each other
and abide by the terms of the contract. Contractarian moral theory has
to assume a natural duty of parental care that it cannot afford to acknowledge. My point here is a similar one: what is unsatisfying about
using the first conception of morality is that we then cannot afford to
acknowledge that the actuality of a social practice of morality, independently of whether the practice gets it right, matters.
Equally unsatisfyingand surely more obviously so for moral
philosophers, who are concerned with identifying correct moral
principlesis the second conception of capital-M morality as nothing
but social morality. One shouldnt, of course, exaggerate the lack of resources for critical reflection within a social practice of morality. There
might be, for example, widespread familiarity with a diversity of moral
practices (family, work, political), as well as shared higher order moral
concepts and methods of moral reasoning that can usefully be drawn
on to engage in critique both of and from within a social practice.13 But
there will be limits to how divergent from generally shared understandings one can be in ones moral thinking and action and still claim to be
operating within the social practice of morality. At some point, taking
the social practice of morality seriously will entail being unwilling to
incur others incomprehension, contempt, resentment, unwillingness to
stay in dialogue, and the like. If you think that capital-M morality is like
a legal systemit exists only so long as there is in fact a shared scheme
of cooperationthere will be a point where deviance from the scheme
is not just a violation of it but also is inconsistent with commitment to
having such a scheme at all.
Here again there is a danger of focusing on misleading cases that
make it difficult to see the unsatisfactoriness of the particular conception of morality. When there has been substantial progress in a social
practice of moralityfor example, in shared understandings concerning
13. Michelle Moody-Adams, The Idea of Moral Progress, Metaphilosophy 30, no. 3
(1999): 168185, argues that such resources will indeed be available given the semantic depth of fundamental moral notions.

16

INTRODUCTION

same-sex sexuality and marriageretrospective, historical stories

about how we were able to get conceptually to the enlightened here-now
from the unenlightened there-then invite us to think about morally progressive historical actors as latching onto what are obviously more defensible moral views. From our present perspective, the improvement
in moral thinking is indeed obvious. And because what enlightened historical actors thought and did fits comfortably with our moral practice
now, we are able to evade thinking about the lack of fit between what
those historical actors did and thought and their own social practice of
morality. Because the story is told from the perspective of greater knowledge, it is difficult to grasp what it was like for historical actors in their
own, historical practice within which these morally progressive views
were not obvious and conflicted with what everyone knew at the time.
Retrospective stories thus obscure how little connection there may be at
any particular historical moment between the right views and the social
practice of morality. In doing so, they allow us to evade thinking about
the reasonableness of co-practitioners rejecting individual efforts to
get it right because the conceptual resources for recognizing that individuals have indeed gotten it right are generally unavailable. In short,
the unsatisfactoriness of (nonevasively) using this second conception of
morality is that it is overly deferential to the content of the social practice of morality.
It is easy, both as moral philosophers and as social practitioners, to
overlook the fact that we rely on two very different conceptions of morality, the commitment to the importance of which has the potential to
pull us rather seriously in different, and equally unsatisfying, directions.
That we overlook the bifurcation between morality as correct action
guide and morality as effective (because actual) scheme of social cooperation is due to an entirely contingent circumstance. In point of fact,
questions about what our moral views should be arise on the backdrop
of a social practice of morality whose general correctness we assume; so
we never have to confront the implications that the conception of morality appropriate to the aim of getting it right has for the importance we
are entitled to attach to actually being able to do morality with others.
We never have to go fully hypothetical in our conception of morality
as a shared scheme of social cooperation. In addition, in point of fact,
17

INTRODUCTION

our social negotiations with others over moral norms typically have the
character of entering an already ongoing fray where there are established
opposing viewpoints that are familiar and intelligible to everyone; so we
never have to confront the implications that the conception of morality
appropriate to the aim of sharing a practice of morality with others has
for the importance we are entitled to attach to the aim of getting it right.
We never have to go fully social in our conception of morality, acknowledging that not all attempts to get it right can be intelligible within the
current social practice of morality.
In short, contingent features about the context in which moral questions are raised create an illusion. The illusion is that there is no problem
requiring a solution in being committed to the importance of getting it
right, with the commitment to going hypothetical that this entails, and
at the same time being committed to really enacting morality, with the
commitment to going actual that that entails. Suppose, now, that we want
to get beyond these illusions. What then? Picking just one or the other
conception of morality is, for the reasons Ive suggested unsatisfying.
The third option is simply to acknowledge that there are two legitimate and important conceptions of capital-M morality, neither one of
which is dispensable. That is because we have two moral aims. On the
one hand, we want to get it right. Both as normative moral theorists and
as critically reflective practitioners, the distinction between what morality really requires and what it is supposed, by some or many, to require is an important one. And thinking in terms of a hypothetical social
practice is valuable for that purpose. On the other hand, we want to be
practitioners of morality, which requires our being located and literate in,
as well as taking seriously, the shared moral understandings of a social
practice. If both are important conceptions of genuine morality, then
both should be the proper business of the normative moral theorist. But
exactly how is that to be accomplished? How is the moral theorist to go
about working simultaneously with two such different conceptions of
capital-M morality?
The place to begin is, I think, to focus on cases where the two conceptions are in sharpest and most painful tension. Those are not cases
where wecritically reflective individualsare spectators on some
moral practice that isnt our own (e.g., the moral practice of another
18

INTRODUCTION

culture, or a different religion, or a profession that isnt our own). Rather,

the painful tensions arise when the individual participant in her own
social practice of moralitya practice in which she is held to account
by others, has to make what she is morally up to intelligible to others,
and bears a (good or bad) moral identity for othershas come, in virtue
of her critical moral reflection, to disagree with some part of that practice and to act contrary to it. If we pay careful attention to what it is like
for those who occupy this uncomfortable position of beingin some
respect, outsiders within the only moral game in their townwe can
begin to see how a normative moral theorist might go about building
into her theorizing an acknowledgment of the importance of the social
practice of morality.
This is what I do in the two chapters in part I. The project is not to
construct some overarching theory or methodology for handling the
two conceptions of morality. The project instead is to provide case studies of that work being done. Both chapters focus on developing new
moral concepts or refining old ones so that it is possible to articulate
what is going on in the moments of tension between the critical reflective point of view and the point of view of the participant who takes her
moral practice seriously. The central new concepts developed in these
chapters are those of moral failure in virtue of getting it right within a
social practice that gets it wrong (chapter 1) and the nonepistemic
notion of the practical weight that co-participants moral assessments
can have for each other even when they regard those assessments as misguided (chapter2). I also argue that we need to revise our conception of
a morally mature and autonomous agent so that it can accommodate the
importance that both getting it right and taking the social practice of
morality seriously have for agents (chapter2).
Second, we can look for places where our account of moral requirements or of moral virtues would be improved if, instead of thinking that
the social practice of morality is irrelevant to specifying the content of
those moral requirements or virtues, we instead tested the opposite assumption: the content of shared moral understandings directly determines what those norms or virtues are. The chapters in part II are all case
studies in taking this approach. First, I argue that we get an improved
account of the virtue of civility and the way civility norms operate if
19

INTRODUCTION

we think about how civility functions in our moral practice with others
(chapter3). I begin with what is likely to seem an uncontroversial way
that normative moral theorizing needs to accommodate shared moral
understandings: sometimes what we need to do morally is communicate moral attitudesin this case, attitudes of respect, consideration,
and tolerance. We can do that only by relying on a shared behavioral
language for doing so, namely the respectful, considerate, and tolerant behaviors required by social norms of civility. Furthermore, we
get an improved account of the bounds of civility if we give up the idea
that the viewpoints and behaviors that are not owed a civil response are
to be determined by reference to some critical moral viewpoint. Philosophical work on civility typically sets the bounds of civility by arguing that some views and behaviors have gotten things so morally wrong
that they do not deserve a civil response. A critical function of civility
norms, however, is to keep dialogue afloat long enough to resolve moral
disputes. Civility norms cannot serve that function if individuals are left
free to set the bounds of civility by appealing to their own conceptions
of what it would be correct to dignify with a civil response and what it
would not. Instead, civility norms can serve their dialogue-supporting
and dispute-resolving function only if the bounds of civility are set by
appeal to social consensus on what falls within and what falls outside the
bounds of civility. Thus the contents of both normative prescriptions for
civil speech and action, as well as normative permissions not to dignify
with a civil response, are directly determined by the social practice of
morality.
We can also get an improved account of the puzzling features of
common decencies if we work from the assumption that shared moral understandings play a role in determining which actions count as common
decencies (chapter 4). The category of common decencies is puzzling
because common decencies dont seem to fit either the category of the
obligatory or the category of the supererogatory. Holding doors open
for others, making pleasant conversation, and forgiving minor moral
errors, for instance, do not fit the model for obligatory actions such as
keeping promises or telling the truth. But nor do these common decencies fit the model of supererogatory actions whose nonperformance is
not criticizable. The puzzle can be solved, I suggest, by noting that any
20

INTRODUCTION

social practice of morality will have to have some conception of what a

minimally well-formed moral agent is like. That conception will include
a conception of which moral gifts a minimally well-formed agent can
be normatively expected to elect and thus be criticized for not electing.
Social practices of morality determine which elective acts qualify as
common decencies. They do so in part because having shared social expectations about what a minimally well-formed agent will elect to do for
others contributes to those expected actions being obviously good and
motivationally nontaxing to do. So here, too, as in the case of civility, the
social practice of morality directly determines the content of normative
claims.
Finally, we can get an improved account of integrity if we give up
the exclusive concentration on the connection between having integrity and being committed to living according to what one regards as the
correct evaluative views (chapter 5). The aim of the chapter is not to
sever the connection between having integrity and getting it right but,
instead, to draw attention to why it is important within a social practice
of morality to have people who think carefully about how to get it right
and who are willing to stand before others for their own best judgment.
Integrity is not just a personal virtue; it is also a social virtue that includes being mindful of the place of affirming ones own best judgment
among co-practitioners who aim collectively to get it right. One of the
disadvantages of linking integrity exclusively to agents efforts to get it
right, I argue, is that the various ways of doing so end up reducing integrity to something with which it is not equivalentto the conditions of
unified agency, to the conditions for continuing as the same self, and to
the conditions for not cooperating with evil. By contrast, we can get an
account of the distinctive nature of the virtue of integrity by focusing on
the social function of that virtue.
Third, we can pay special attention to theorizing nonideal conditions
where misguided moral understandings have been socially conventionalized. Sometimes actions and dispositions that in fact satisfy correct
standards of what morality requires appear within the social practice
of morality to be morally criticizable. The chapters in part I examine
these kinds of cases, focusing on what it is like for individuals aiming to
get it right in social contexts where right action elicits resentment and
21

INTRODUCTION

shaming treatment, and where the individuals who are striving to act
correctly fail to be able to make what they are morally up to intelligible
to co-practitioners. Under nonideal conditions it can also be the case
that actions and dispositions that fail correct standards of what morality requires appear within the social practice of morality to be morally
permissible or obligatory. Wrongdoing is conventionalized. The pair of
chapters in part III take up this latter phenomenon, focusing particularly on what it is like for individuals immersed in the assumptions of
their social practice of morality.
One problem, faced acutely by moral philosophers, is to find some
method for detecting conventionalized wrongdoing and for explaining
what makes it wrong (chapter6). Since getting it right involves going hypothetical and imagining what moral norms would be endorsed within
a hypothetical social practice of morality, we need guidance on how
exactly going hypothetical is to be conducted. Given that many social
practices of morality involve some level of conventionalized wrongdoing that is rationalized by cultural ideologies and culturally available
knowledges, and given that socialization into social practices of morality aim to produce general acceptance of and compliance with those
practices, we cannot go hypothetical merely by imagining individuals
who are better reasoners. As a test case, I explore how Kants universalizability testhis Categorical Imperative procedurewould need to
be revised in order to be able to detect and explain what is wrong with
forms of wrongdoing that are socially conventionalized. The idea is to
try to construct a procedure that does not implicitly assume that we,
moral philosophers, already know what social practices are morally misguided and then simply articulate why similarly enlightened parties to a
hypothetical social practice would agree with us.
Focusing on what it is like for social participants, most of whose
moral knowledge is derived from their social practice and is simply what
everyone knows, raises difficult questions about when individuals can
be held responsible for conventionalized wrongdoing and what the relation is between responsibility assessments and the blaming practices by
which we hold individuals responsible and encourage them to improve
their moral performance (chapter7). In addressing those questions, it
is useful to keep front and center the social function of blaming. If it
22

INTRODUCTION

turns out, as I argue that it does, that individuals can sometimes be exempted for behaving badly because their social practice of morality does
not enable them to discern the wrongfulness of their behavior, then the
warrant for blaming cannot be exclusively agent culpability without undermining the social function of blaming. In emphasizing the point or
function of moral phenomena, such as blaming, chapter7 uses a methodological approach that was central also to chapters3 and 5, which explored the social function of civility norms and the virtue of integrity.
As in some of the books earlier chapters (especially chapters 2
and3), I do not attempt to resolve in chapter7 the tension between the
conception of morality as correct action guide and the conception of
morality as effective scheme of social cooperation. In discussing civility,
for example, I leave unresolved how to handle cases where we cannot
simultaneously correctly treat people with respect and communicate attitudes of respect. In chapter7, I leave unresolved how to handle cases
where we cannot simultaneously exempt from responsibility those who
deserve to be exempted and preserve the point of holding people responsible via our blaming practices.
Fourth, we can shift from an exclusive focus, in moral theorizing,
on developing correct action guides to developing the conceptual resources for describing the full range of moral transactions between
people within social practices of morality. Not all of our transactions,
for example, fit the model of wrongdoing agent-resentful patient and
rightdoing agent-appreciative patient. There are also interesting ways
in which we mediate between agents, on the one hand, and their own
actions, their reactive attitudes, and the patients of their actions, on the
other hand.
The pair of chapters in part IV examine this different sort of transaction. In a very short essay, I explore the phenomenon of emotional
work and the mediator-centered role that we occupy while doing it
(chapter8). In chapter9, I take up the mediating work involved in aspirational forgivenessthat is, forgiveness offered to culpable, unrepentant wrongdoers. We can elect to tell the kind of story for those who
treat us badly that makes possible forgiving those who deserve to suffer
the reactive attitude of resentment. In doing so, we mediate between the
wrongdoer and the patient who happens to be us. This closing chapter
23

INTRODUCTION

on forgiveness also suggests another way we could think more broadly

about the full range of moral transactions that take place within social
practices of morality. Sometimes life is morally unfair. Individuals
may acquire histories of failed moral interactionsfor example, being
repeatedly denied the moral gifts I discuss in chapter 4 and repeatedly receiving morally bad treatment. I briefly introduce the notion of
making biographical sense of ones moral history as a way to capture
considerations other than getting it right that may enter into individuals moral decision making.
As this last comment suggests, this closing chapter puts front and
center a project that I think is important for moral theorists to take up,
and that is implicit in some of the earlier chapters. It is not the standard project of developing the tools for and carrying out assessments,
critiques, and the making of moral prescriptions. It is, instead, the project of developing the conceptual tools for understandingespecially
sympathetically understandingthe difficulties that individuals face
in their efforts to get it right while simultaneously practicing morality
with others.

24

PA R T I

C R I T IC A L MOR A LI T Y
A N DSOC I A L NOR M S
The pair of essays in part I focus on bringing into view two different conceptions of morality. On the one hand, we think of morality as comprising a set
of principles for right action, including principles governing how we should
treat others and how we as self-respecting beings should behave. We also
think of morality as comprising the set of virtues that good persons should
cultivate and vices they should avoid. Finally, we might include under this
first conception of morality (thinking of it more broadly as ethics) conceptions of the good life that are worthy of human pursuit. What distinguishes
this first conception of morality is the requirement that all of its elements
be rationally defensible. Morality comprises the correct set of action guides,
dispositions, and (thinking more broadly) worthy ways of living. This first
conception entails that our moral aim should be getting it rightcoming to
hold the correct moral views and conforming our actions, dispositions, and
goals to those correct views. The capacity to take a critical viewpoint, to distance ourselves from social convention, and to reason autonomously is essential to morality in this sense.
On the other hand, morality is only ever practiced in social worlds. Particular social worlds will have their own shared moral understandings of the
principles that ought to guide co-participants moral behavior and the kinds
of dispositions they ought to cultivate or avoid in order to be morally good
people. In a morally perfect world, the correct moral conceptions that are justifiable from a critical perspective would themselves be embodied in shared

MOR A L A IMS

moral understandings within the social practice of morality. In a morally imperfect world, the two conceptions come apart to a greater or lesser extent.
As a result moral actors may find that in their efforts to get it right, what
they are morally up to is not legible to co-participants in their moral practice.
My central aims in these essays are, first, to make the case that the social
practice of morality really is morality and not merely a set of social norms
that parades as morality; and second, to begin to develop concepts adequate
for capturing the moral experience of persons who aim at getting it right
within imperfect moral worlds where what they are morally up to is not legible to others. To that end, I focus on a phenomenon I called moral failure
and on moral shame construed as a social emotion.

26

C ha pt e r 1

Moral Failure
Moral revolutionaries are people who succeed in thinking from a moral
point of view that both exceeds and improves upon the conventional
moral understandings broadly shared in their social worlds.* They get
it right under social circumstances that make it difficult to do so. And
we admire them for it. In this chapter, I pursue the paradoxical thought
that their getting it right actually produces a particular kind of moral failure of their lives. Thus, such revolutionaries are likely to have reason for
regret about how their lives turn out morally.

I. FA ILUR ES, MOR A L FA ILUR E,

A ND MOR A L LUCK
Failure is not the same as culpable error. For culpable errors one is held
responsible, downgraded, chastised, penalized, punished, disapproved,
resented, held in contempt. One may feel guilty about, repent, make
amends for culpable errors. Failures, by contrast, are not culpableat
least not the failures I am interested in. For want of talent, one might fail
to be a good philosopher; or for want of the inner resources to be cheerful, one might fail to have friends; or for want of natural grace or rhythm,
one might fail to be able to dance. Nor are failures simply excused errors.
A good excuse gets one off the evaluative hook. To be excused is to have
* Two works have particularly influenced the content of this chapter: Claudia Cards The
Unnatural Lottery: Character and Moral Luck (Philadelphia: Temple University Press,
1996) and Margaret Urban Walkers Moral Understandings: A Feminist Study in Ethics
(New York: Routledge, 1998). My particular approach to moral failure grows out of
themes I develop in chapters 2, 5, 6, and 7.

27

MOR A L A IMS

no reason to think badly of oneself or for others to think badly of oneself.

To have failed, by contrast, is to have a reason to think badly of oneself
and to expect others to do the same. However unavoidable turning out
to be a bad philosopher or a friendless person or incapable of dance may
have been, these failures leave their evaluative mark. They are sources of
regret, shame, loss of self-esteem, and the thought that ones character or
life is blemished by falling short of some standard for what lives should
look like.
One might, of course, deny that there are any such things as the
failures I have described. Either one is culpable or one isnt; evaluation
tracks those two conditions. Much moral philosophy, in its focus on
the will, obligation, and responsibility, gives the impression that no one
simply fails. But without a space for the notion of failure, it is hard to
make sense of many of the things that shame us or inspire the thought
that our lives have not turned out as human lives are supposed toour
uncomeliness, lack of talent, gracelessness, competitive poor showings,
and crumbled marriages.
In addition, without a space for the notion of failure, there will be
no way to acknowledge that what we expect from other people and from
ourselves is not in fact confined to what is under voluntary control.
Some of our expectations are tied to thoughts about what is statistically
normal for persons or for persons of a certain sort.1 Normal people have
some modicum of talent, or cheerfulness, or grace; those who dont are
failures. Other expectations are tied to an ideology of the normal that is
disconnected from what real people are typically like. Normal people
are supposed to be self-supporting and capable of sustaining long-term
marriages; those who arent are failures. Other expectations are tied to
ideals rather than to normalcy. To embark on a career is to hold up for
oneself an ideal of excellence or be held to it by others. To fall short is to
fail, sometimes in a minor way and sometimes thoroughly.
1. On the idea that our attributions of responsibility are connected to expectations about
what is statistically normal, see Ferdinand Shoeman, Statistical Norms and Moral
Attributions, in Responsibility, Character, and the Emotions: New Essays in Moral Psychology, ed. Ferdinand Shoeman (Cambridge: Cambridge University Press, 1987); and
Jeffrie G. Murphy, Moral Death: A Kantian Essay on Psychopathology, in Ethics and
Personality, ed. John Deigh (Chicago: University of Chicago Press, 1992).

28

MOR A L FA I LU R E

In moral philosophy, the notion of moral luck captures one sense of

specifically moral failure. As Thomas Nagel developed the notion, our
actions and characters are vulnerable to moral assessment so long as we
have made some contribution to what our actions and characters are,
even if what we actually do or actually turn out to be is mostly a matter
of luck, pure and simple.2 So, for example, we morally assess the accidentally successful rescue attempt and the accidentally botched rescue
attempt differently, even though succeeding or botching was a matter of
luck. We morally assess the character of those who participated in Nazi
Germany differently from those who didnt, even though it was a matter
of luck that some people but not others faced the particular moral tests
posed by life in Nazi Germany. Victims of bad moral luck fail to perform
well and we blame them for it, even though much of what contributed to
their deeds being what they were was not under their control.
In Nagels view, the moral part of moral luck hinges on our having
made some contribution to our deeds or character. It is the fact that we
can be held partly responsible for what we do or are that gives moral assessment a foothold. Underlying this view is a remnant of the Kantian
notion that the domain of morality extends only to what we can control.
Thus moral failures must partially connect to that domain.
A quite different account of the moral part of moral luck seems to be
at work in Martha Nussbaums use of that notion. 3 For her, the ideal of a
morally excellent life is what makes moral failure possible. Oedipus, for
example, fails to live a morally excellent life. Through no fault of his own,
his life becomes blemished by acts of incest and patricide. Although he
made contributions to these deeds, that is not what makes him vulnerable to moral bad luck, as opposed to just plain bad luck, on this account.
Rather, his bad luck and failure are moral, because the ideal in the light
of which he is assessed is a moral ideal of what human lives should be.
Claudia Card also develops an account of moral luck that differs
from Nagels.4 Whereas Nagel emphasized the luck that enters into our
2. Thomas Nagel, Moral Luck, in Mortal Questions (Cambridge: Cambridge University
Press, 1979).
3. Martha C. Nussbaum, The Fragility of Goodness: Luck and Ethics in Greek Tragedy and
Philosophy (Cambridge: Cambridge University Press, 1986), esp. chap. 11.
4. Card, The Unnatural Lottery.

29

MOR A L A IMS

being held responsible, blamed, or praised, Card emphasizes the luck

that enhances or undermines our capacity to take responsibility for ourselves. Taking responsibility for ourselves includes taking responsibility
for the social meaning of our lives and actions. For example, when being
lesbian is socially defined as unnatural and perverse, taking responsibility for being lesbian will involve creating and imposing new meanings so
that one can stand behind ones life. Success, however, depends on how
others receive these new meanings. Thus taking responsibility will be a
matter of luck. The luck is moral because taking responsibility is a basic
form of moral activity.
The notion of moral failure I have in mind is closer to Nussbaums
and Cards than to Nagels. What I will suggest is that among the ideals
of what a human moral life should look like is the ideal of living a moral
life within a shared scheme of social cooperation where ones moral understandings are shared by others. Under these conditions ones moral
activity and ones moral reasons will be intelligible to others. Given sufficient bad luck, our moral lives can fail because they are characterized
by abnormally frequent unintelligibility to others or abnormally frequent inability to defend ones actions in terms that others find meaningful. Our attempts to be self-respecting, to avoid misplaced gratitude,
to generously offer what is not owed may be received by others as arrogance, ingratitude, and mere dutifulness. Under such conditions, our
moral practice is idiosyncratic, not part of a common scheme of social
cooperation. If this is in fact a kind of moral failure, it is a failure from
which impeccable exercises of responsibility cannot protect us.
Obviously, it will take some work to make the case that there is such
an ideal, that falling abnormally short of it is a moral failure, and that
trying to do the right thing can produce this failure. Let me begin, then,
with doing the right thing.

II. DOING THE R IGHT THING A ND FEMINIST

R ESISTA NCE
Trying to do the right thing, to live morally well, is not just one thing but
many. Realizing that moral philosophers disagree among themselves
30

MOR A L FA I LU R E

about what these moral tasks are, let me propose the following four commitments as relatively uncontroversial and basic to (if not exhaustive of)
any attempt to do the right thing:
1. The principle of self-respect. I am a being with self-respect; and
as a being with self-respect, I will affirm my place in the moral
world.
2. The principle of mutually agreeable rules. I am a reasonable
being; and as a reasonable being, I will act according to principles that could be mutually agreed to by free, equal, reasonable, and rational beings.
3. The principle of pursuing the good. I am a rational being with the
powers to frame a conception of the good; and as a rational
being, I will act on my conception of the good.
4. The principle of character. I am a being with moral character;
and as a being with moral character, I will cultivate and express
the virtues.

These principles, if correct, express the moral commitments any

agent, in any social context, must have and act on if she is to do the
right thing. In this sense, doing the right thing is always the same thing.
However, these moral commitments must be enacted in the agents
own social world, where a moral practice is already under way and
where there are established and broadly shared social understandings
of what counts as doing the right thing. In morally well-formed social
worlds, doing the right thing will be a matter of compliance with shared
moral understandings. But in morally ill-formed social worlds, doing
the right thing will require resistance to the existing practice of morality. In this sense, doing the right thing is not the same thing across all
possible social contexts.
Feminist moral philosophers, unlike more conventional moral philosophers, have been interested in describing the shared moral understandings that operate in sexist, heterosexist, classist, and racist social
worlds. They have also been interested in what it means, particularly for
members of subordinate groups, to try to do the right thing in these social
contexts. In particular, feminists have drawn attention to the facts that, in
31

MOR A L A IMS

our social world: (1) some groups are socially constructed as moral inferiors to be treated as second-class citizens in the moral world; (2) unjust
practices to which members of subordinate groups could not possibly
agree absent coercion are socially institutionalized; (3) some healthy
conceptions of the good are deemed inappropriate for some social groups
(for example, fulfilling same-sex erotic relationships, marriage, and
family for gays and lesbians), whereas damaging conceptions of the good
are deemed appropriate (for example, for women, the pursuit of excessive slimness and use of plastic surgery); and (4) the images of virtue or
of what it takes to avoid vice that are offered to women, blacks, gays and
lesbians, and the poor are deformed and demeaning ones (for example,
avoiding arrogance means deferring to male and white authority, being
civilly respectful of others feelings means concealing ones lesbian identity, and having a work ethic means accepting poverty as ones own fault).
The four principles for doing the right thing, when put into play in
ill-formed social worldsparticularly when put into play by members
of subordinate groupswill be principles of resistance. From the standpoint of the subordinated, for example, the principle of self-respect is
primarily a principle of intolerance: I am a being with self-respect,
and as a being with self-respect I will not tolerate____. To be selfrespecting is to refuse to put up with humiliation, abuse, unfair denial
of opportunities, objectification, demeaning or defaming stereotypes, 5
silencing, and domination. It is to refuse to offer misplaced gratitude for
treatment that is simply ones due.6 And it is to resist the idea that members of subordinate groups are not entitled to morally judge members
of dominant groups and, thus, are not entitled to express anger at moral
mistreatment.7 Because ones own mistreatment is connected to that of
fellow subordinates, the resistance required by a principle of self-respect
5. On the construction of gay and lesbian identities as demeaning and defaming ones, see
Claudia Card, Lesbian Choices (New York: Columbia University Press, 1995), 151168.
6. Thomas E. Hill Jr., Servility and Self-Respect, in Autonomy and Self-Respect (Cambridge: Cambridge University Press, 1991), 418. It is from this piece that the principle
of self-respect I suggest is derived.
7. Elizabeth V. Spelman, Anger and Insubordination, in Women, Knowledge, and Reality:
Explorations in Feminist Philosophy, ed. Ann Garry and Marilyn Pearsall (Boston: Unwin
Hyman, 1989).

32

MOR A L FA I LU R E

is likely to be not just resistance to ones own mistreatment but also a

general resistance to a system of domination.
Similarly, from the standpoint of the subordinated, the principle of
accepting only mutually agreeable rules is primarily a principle of resistance. Since we are not now in Rawlss original position (a position
of ideal freedom and equality, from which the principles to govern our
choices would be chosen), but find ourselves immersed in a practice of
morality already under way, and since much of that practice supports
systems of domination, to accept only mutually agreeable rules will inevitably mean refusing to abide by existing social norms to which women,
blacks, gays and lesbians, and the poor would not have consented had
they occupied positions as free and equal participants in the social
scheme.8 This principle may also require resisting decision-making arrangements that exclude participation by those whose lives will be significantly affected by those decisions (for example, the policy of having
experts within welfare bureaucracies make unilateral decisions for their
clients).9 At a theoretical level, it may require resisting philosophical
constructions of impartial decision making that exclude the very dialogue with real others that might secure the genuine impartiality necessary for locating rules that in fact could be mutually agreed to by all.10
Acting on the principle of pursuing ones own conception of the
good will also largely be a matter of resisting those conceptions socially
prescribed as appropriate for ones social groupas women have historically tried to resist patriarchal marriage by refusing to marry, by
constructing Boston marriages with other women, by cross-dressing
and marrying women, and by divorcing out of inegalitarian marriages.
As these examples suggest, it may also require pursuing conceptions

8. On the original position, see John Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press, 1971), 1722.
9. See Iris Marion Young, Justice and the Politics of Difference (Princeton, NJ: Princeton
University Press, 1990); and Kathryn Pyne Addelson Moral Passages: Toward a Collectivist Moral Theory (New York: Routledge, 1994).
10. See Seyla Behabib, Situating the Self: Gender, Community and Postmodernism in Contemporary Ethics (New York: Routledge, 1992); Young, Justice and the Politics of Difference;
and Marilyn Friedman, What Are Friends For? Feminist Perspectives on Personal Relationships and Moral Theory (Ithaca, NY: Cornell University Press, 1993).

33

MOR A L A IMS

of the good that are socially deemed unwise, unnatural, or irrationally

riskyconceptions that are inconceivable within the dominant view as
possible conceptions of the good. In addition, it will require both resistance to the culturally normalized but unfair distributions of resources
to the subordinated (distributions that constrain their pursuit of the
good) and resistance to their lack of credibility as judges of the good (a
lack that undermines social negotiation for conditions more conducive
to their flourishing).11
Finally, the principle of moral character will be a principle of refusing to comply with social definitions of the virtues appropriate to ones
station that in fact crush or cramp genuine expressions of virtue. Central to the application of this principle of moral character will be resistance to ideologies and social practices that naturalize and normalize the
idea that there are different, and differently valued, virtues for different
social groups. In particular, it will be necessary to resist the maddening
idea that there is a set of virtues appropriate to generic, mature humans
and a different, incompatible set of virtues appropriate to women or
other social groups.12 Sometimes it will be necessary to resist ideologies
and practices that construct the absence of virtue as a natural, unalterable feature of some social groups.

III. THE MOR A L IDEA L OF DOING THE R IGHT

THING
Although resistance is often personally costly, it is also morally attractive. These four principles, which under unjust conditions become
principles of resistance, are connected to a particular moral ideal. That
11. On epistemic credibility, see Walker, Moral Understandings. I have argued in Family
Outlaws: Rethinking the Connections between Feminism, Lesbianism, and the
Family, in Feminism and Families, ed. Hilde Lindemann Nelson (New York: Routledge, 1996), that gays and lesbians lack definitional authority with respect to the
family and thus do not have the same standing that heterosexuals do to recommend
changes in family law.
12. Kathryn Pauly Morgan, Women and Moral Madness, in Science, Morality, and Feminist Theory, ed. Marsha P. Hanen and Kai Nielsen (Calgary: University of Calgary
Press, 1979).

34

MOR A L FA I LU R E

ideal is the ideal of a life beyond self-reproach. One aim of moral life
is to become sufficiently critically reflective, sufficiently motivated, and
sufficiently alive to ones own moral status, to the importance of a cooperative scheme, to ones options for constructing a good life, and to
ones possibilities for virtue that one need not reproach oneself later for
having been servile or unfair or thoughtless about the good or vicious.
It is an ideal fit for self-determining beings who are custodians of their
own lives and who are capable of deciding for themselves what shape
those lives should take. It is, I think, a correct ideal. This is, in part, what
we are trying to do when we participate in the enterprise of morality.13
To say that it is an ideal is to say that real human lives are not in fact
going to be beyond reproach. Negligence, narrow-mindedness, a desire
to retain privileges, cowardice, and the like will make for culpable fallings short of the ideal. In addition, when dominance and subordination
are conventionalized and rendered natural, normal, and unproblematic,
when necessary knowledges are suppressed (for example, knowledge of
the history of oppression), or when critical moral concepts are not socially available (for example, the concept of date or marital rape), then
there is a live possibility that a person will just not be able to see how
morally badly her or his life is going. Loving devotion turns out to have
been servility. Living up to ones station and its duties turns out to have
been complicity with injustice. Being a good X turns out to have meant
the cultivation of vice rather than virtue. These are moral failures. They
are failures of ones life to embody the ideal of doing the right thing in
spite of ones best efforts. One kind of moral failure, then, that is an especially live possibility when injustice is conventionalized, so that agents
themselves are not well positioned to determine what the right thing is,
is the possibility that trying to do the right thing might end in failure.14
13. Feminists have been highly critical of conventional accounts of the autonomous
person. But those criticisms are, I think, less critiques of the ideal of living beyond reproach and more critiques of the lack of realism, often characteristic of moral philosophies that ignore our actual social context, about how possible it is to live such a life.
14. For a discussion of the difficulties involved in getting it right, as well as a critical evaluation of some now standard attempts to get it right in academia (by, for example,
promoting affirmative action), see Marilyn Fryes Getting It Right, in Willful Virgin:
Essays in Feminism 19761992 (Berkeley, CA: Crossing Press, 1992).

35

MOR A L A IMS

The more paradoxical possibility, which I pursue here, is the possibility

that resistantly trying to do the right thing might produce moral failure.
How could that be? I begin by describing the kind of failure that I think
resistantly trying to do the right thing produces. I then turn to reasons
for thinking this is a specifically moral form of failure.

I V. ILLEGIBILIT Y A ND U NR E ASONA BLENESS

One of the most important effects of liberation movements is that they
produce critiques of conventional moral norms. Such critiques show
why compliance with conventional moral norms is not in fact a way of
doing the right thing, but is instead a way of participating in and sustaining systems of domination. The feminist movement, for example, challenged a conventional assumption that wives who take on the principal
burden of unpaid domestic labor are simply doing their fair sharefair
because this is what wives owe to their families. It also challenged the
idea that a good life for women must include childrearing and personal
attachment to a man. The lesbian and gay movement challenged a conventional assumption that making ones lesbianism or homosexuality
known is, among other things, rude and shameless. Some of the moral
critiques produced by liberation movements have now been conventionalized. They have become part of our common stock of moral understandings. This is not to say that everyone endorses those critiques. It
is to say that everyone finds them familiar and comprehensible. So, for
instance, a black mans angry response at being called boy or a womans filing sexual harassment charges are now legible as affirmations of
self-respect. What both are morally up to doesnt need explaining.
However, when large portions of dominance systems continue to be
conventionalized, formulating moral critiques will produce what I have
elsewhere called abnormal moral contexts.15 Abnormal moral contexts occur when some segment of a society produces advances in moral
knowledge that outrun the social mechanisms for disseminating and
normalizing that knowledge in the society as a whole. In that case, a gap
15. Chapter 7, this volume.

36

MOR A L FA I LU R E

opens between what everyone knows is the right thing to do and what
from a (presumably) advantaged epistemic position is viewed as the
right thing to do. The gap, of course, will be obvious only to those who
take themselves to be reasoning from a more advanced, socially critical
point of viewas feminists, for example, generally take themselves to
be doing. It is that gap that makes doing the right thing, as determined
from this socially critical point of view, necessarily a form of resistance.
To do the right thing under circumstances where dominance systems are conventionalized requires rejecting broadly shared social
assumptions about the moral place persons are entitled to claim for
themselves; about which practices are morally legitimate; about what
counts as courage, generosity, proper pride, and so forth; and about
which forms of life count as good ones. Of course, from the point of view
of those who dont have access to these critiques and the evidence that
supports them, these acts of resistance will not be legible as either acts of
resistance or as attempts to do the right thing. They will simply look like
doing the wrong thing. Refusing to be grateful for help with the housework will appear to be ingratitude. Refusing custody of ones children
upon divorce will appear coldly unloving rather than a resistance to
compulsory motherhood. Kissing ones domestic partner in public will
appear confrontationally obscene rather than affectionate.
I have chosen the terms legible and illegible to underscore the
fact that the social practice of morality depends heavily on our being
able to read the meaning of others actions. To take a simple example,
were expressions of gratitude, such as saying thank you, not interpretable by recipients as an expression of gratitude, this particular moral
exchange would break down. Under these conditions, a person might
privately intend to express gratitude, but if the expression is illegible,
there is a real sense in which no gratitude is actually expressed.
When moral resisters have the opportunity to explain what they
are doing, and thus make their actions legible, they may still be unable
to make themselves seem reasonably justified. Their justifications may
be received as wildly implausible, irrational, based on patently false assumptions, and thus not really justifications at all. The difficulty of justifying oneself is often further complicated by the fact that subordinate
groups typically are also socially constructed as defective reasoners.
37

MOR A L A IMS

Moral resisters commitment to doing the right thing thus risks producing two forms of failure: a failure to make what one is morally up to
legible to others and a failure to provide justifications that are recognizable to others as justifications.

V. MOR A L FA ILUR E A ND THE IDEA L

OF A SH A R ED SCHEM E OF SOCI A L
COOPER ATION
But why think that these failures are moral? Why think that a life characterized by abnormally frequent illegibility or by abnormally frequent
inability to defend ones actions in terms others find intelligible is morally defective, lacks the moral excellence one expects of a moral life, and
is an occasion for moral regret and possibly also moral shame? Quite
the contrary, living a genuinely self-respecting life, refusing to comply
with unjust practices, correctly conceiving and enacting the virtues,
and living out a genuinely estimable life plan all seem reasons for moral
self-congratulation. If failure is to attach anywhere, it seems more reasonable to attach it to those whose lack of critical distance from social
moral norms prevents them from seeing and finding intelligible what
moral resisters are up to. If another cannot see being out of the closet
as an affirmation of self-respect, isnt the failure theirs rather than the
uncloseted persons?
In addition, moral philosophers standardly distinguish between morality as a system of social normsa cultures moral codeand morality as a set of prescriptions that are justifiable from a critical, reflective,
theoretical point of view. Because social moralities may not survive critical review, they are better thought of simply as social norms rather than
as constitutive of morality. It is from the point of view of social norms
that moral resisters actions are illegible or without minimal justification. From a genuinely moral point of view, what they are up to morally
is perfectly legible and intelligible. To view failures of social legibility
and justifiability as moral failures thus seems to confuse social norms
with genuine morality. If there is any failure here, it is merely social, not
moral. Indeed, moral resisters may well be social failures, regarded as
38

MOR A L FA I LU R E

deviant, outlaw, perverse, crazy, or extremist. Being so regarded does

not reflect on their moral excellence.
These are compelling objections. However, I think they rest on three
interconnected, mistaken assumptions: (1) that the moral ideal of doing
the right thing is the only relevant ideal for assessing the moral excellence of lives; (2) that the successful social enactment of morality is not
itself a moral ideal; and (3) that if there are multiple moral ideals, they
cannot be in such fundamental tension with each other that it is impossible to orient ones life toward all of them simultaneously.
I take the second assumption first, since it is the heart of the matter.
Morality is fundamentally social, and one common way of stating this
idea is to say that morality is a scheme of social cooperation. The fact
that morality is a scheme of social cooperation suggests that the distinction between social norms and genuine morality is misleading.
Indeed, any attempt to cleanly distinguish social norms from genuine morality is like the attempt to imagine an unperceived world. As
Bishop Berkeley pointed out, in the very process of imagining an unperceived world, we covertly insert a perceiverourselves. So, too, in
conceiving a distinction between genuine morality and social norms,
we do not purify morality of the social. Instead, we covertly insert
a different social world into the picture, one in which what we take
to be genuine moral norms are also socially normative. Kants kingdom of ends is a hypothetical social world. In that world, universal
moral laws are social norms. The ends in this kingdom are social
participants in a practice of morality. They share common moral understandings of what things mean morally (for example, when gratitude is misplaced or what treatments are humiliating). The correct
contrast, then, is not between genuine and merely social morality
but between two different social moralities, one hypothetical and the
other actual, where we take the hypothetical one to be preferable to
the actual one.
The original objection to counting as a moral failure resisters failure
to make legible what they are morally up to might, then, be more accurately put this way: moral resisters, whose actions are illegible according
to actual social norms, have not failed morally because their actions are
legible according to a more nearly correct, although hypothetical, set of
39

MOR A L A IMS

social norms. In other words, the only thing that really matters so far as
moral success or failure is concerned is the ideal of getting it right.
But is it? Is getting it right the only thing we aim to do when we
participate in the practice of morality? Or do we aim at other things as
wellthings that might depend upon our being able to make what we
are morally up to legible and justifiable to others?
Let us return to the idea that morality is a scheme of social cooperation. That morality is a scheme of social cooperation means that even
though individuals are to guide their behavior by moral rules, moral
rules are not designed for individuals. They are designed instead for the
social worlds that individuals inhabit. Similarly, even though individuals are to cultivate virtues, the point of virtue is not just to make our individual lives good but also to make our common lives good. The shared
cultivation of virtue enables us to count on others to do the things that
need doing. Because morality is a scheme of social cooperation, both the
attempts of philosophers to frame justifiable schemes of social cooperation and the efforts of individuals to do the right thing have the same
practical aim: to put into play in our social world a shared set of understandings about how we are to do things morally together. It makes no
sense to engage in critical moral reflection or to attempt to do the right
thing without this practical aim. To do so would require treating morality as a kind of private language whose rules or conceptions of virtue
need not be accessible or meaningful to anyone else.
It is no surprise, then, that moral theories so often articulate justification as a matter of justifying ourselves to others, with the aim of
securing shared moral understandings that can guide our common life
together. Role-reversal tests embody this social conception of justification in a modest way by focusing our attention on what individual
others might think of our proposals. Social-contract theories like those
of Hobbes and Rawls, dialogic models like that of Habermas, and legislative models like that of Kant in the third formulation of the categorical
imperative (that of the kingdom of ends) employ more fundamentally
social conceptions of justification (even if the society is a hypothetical
one). More fully social conceptions of justification (because they are less
hypothetical) are communitarian models where justification appeals
to traditions and understandings that are actually shared, and some
40

MOR A L FA I LU R E

feminist reconstructions of dialogic models that employ real, rather

than ideal, discourse situations. Common to all these approaches is the
assumption that what we are aiming at in the process of justification
is mutual agreement on a common scheme of social cooperation. The
moral ideal operating in theories of justification is the ideal of making
ourselves intelligible to othersso that, for example, any contractor
behind a Rawlsian veil of ignorance can take up our positionand of
actually reaching shared moral understandings.16
Now, whether we are in truth justified in what we do as real moral
actors may depend only on what would happen in a hypothetical social
world, like that of the Rawlsian original position, in which we attempt
to justify our actions to hypothetical others. Thus orienting our lives
toward the ideal of doing the right thing may not require that we be
able to justify ourselves to real others in our actual social world. This is
especially true in social worlds where the real participants suffer from
epistemic defects, such as socialization to accept dominance systems
as natural, normal, and legitimate. But even if being justified is detachable from how others receive us in our actual social, moral world, the
ideal of being able to make ourselves intelligible and to reach shared
moral understandings continues to operate in our actual social world.
Hypothetical social worlds, like those of the Rawlsian original position
and the ideal discourse situation, help us to specify what the ideal is,
what we ultimately want out of our moral livesnamely, shared moral
understandings.
But that ideal does not operate only in hypothetical worlds. As
participants in an actual moral practice, we operate under the ideal of
participating in a shared scheme of social cooperation. Maximally, a
shared scheme is one in which there is full consensus on who has which
moral status, on which principles and practices are legitimate, on what
constitutes particular virtues, and on what falls within the range of possible conceptions of the good. Minimally, a shared moral scheme means
that we share enough moral understandings that we can successfully
16. On the veil of ignorance, which screens out the particulars that enable us to distinguish one individual (such as ourselves) from another, see Rawls, A Theory of Justice,
136142.

41

MOR A L A IMS

interpret what others are morally up to and see their reasons as providing some justification, even if we ultimately disagree. To abandon the
ideal of a shared scheme of cooperation would be to give up hope for the
possibility of a moral practice in our actual social world.

V I. T WO IDE A LS
I suggest, then, that there are two ideals for what moral lives should
look like. One is the familiar ideal of getting it right. The search for
correct principles and adequate justifications is part of realizing this
ideal. The other is the ideal of participating in a shared scheme of social
cooperation. Communicating our moral views to others, offering explanations and justifications and seeking consensus, is part of realizing
this ideal.
What distinguishes the two moral ideals is that orienting our lives
toward the first is up to us in a way that orienting our lives toward the
second is not. It is substantially up to me whether I govern my life by
principles, conceptions of virtue, and a conception of the range of possible good lives that would be shared in a hypothetical social world.17 It
is substantially not up to me whether my life is, at the same time, also
oriented toward reaching common moral understandings. Whether it is
also so oriented largely depends on who my fellow moral practitioners
are and on the possibilities for reception. Bad moral luck may undermine the aim of participating in a shared scheme of social cooperation by which we can make what we are morally up to legible to others.
Self-respect may be persistently received as arrogance, integrity as irrational extremism, generosity as merely fulfilling an obligation, love as
17. I say substantially because the control we have over getting it right is a matter of
degree and is also vulnerable to luck. As actual moral reasoners, we are embedded in
social worlds that may provide better or worse resources for successfully conducting
the sort of inquiry required by hypothetical contract or discourse scenarios. It has for
that reason been a central feminist critique of Rawls that the method of going behind a
veil of ignorance, and of imagining oneself in multiple social positions or as bearers of
multiple conceptions of the good, is not a method that real moral reasoners can employ
in its pure form.

42

MOR A L FA I LU R E

perversity, demanding fairness as demanding special rights, sustaining a family as leaching off the system, and so on.
What the two ideals share is that both provide yardsticks, independent of considerations of praiseworthiness or blameworthiness, for
measuring the excellence and success of our moral lives. I may not be to
blame that my life has been one of servility or arrogance or unfairness.
But to discover after the fact that it has been so because of my moral misconceptions is to discover moral failure. It is a failure that merits moral
regret and also shame, because I am now revealed, particularly in the
eyes of others about whose opinion I care most, not to have measured
up to a standard that applies to me. Similarly, although less obviously, to
find that ones moral life is marked by abnormally frequent occasions of
being morally illegible to others, and of having ones reasons rejected as
not even minimally justifying, is to find that ones moral life has failed.
It has failed in much the way van Gogh might have thought that his life
as a participant in the social practice of art had failed. No matter how
good his work was, his life as an artist was in part a failure because art
also aims at being shared. So, too, no matter how much one gets it morally right, ones life as a moral practitioner may end in failure because the
practice of morality also aims at a common moral life together.
This failure merits moral regret. It is less obvious that it merits
shame. What would seem to bar this sort of failures being a fit subject
for shame is that if one really is getting it right, then one has nothing
to be ashamed of. If others mistake objecting to unfair treatment as arrogance, nagging, shrillness, or demanding special treatment or special
rights, that is simply their mistake. The moral resister is not really any
of these things. She has no reason for shame. But the line between reality and appearance may not be so sharp. Our actions have meanings
in the social world. Individuals cannot change those meanings at will.18
A woman who persistently complains that her husband is not doing
enough domestic labor is a nag. That is what her actions mean in this
social world, even if, from the point of view of the hypothetical social
world that guides her decision making, her actions also mean standing
18. Claudia Card develops this point more elegantly and forcefully that I do here (The Unnatural Lottery, 140162).

43

MOR A L A IMS

up for fair treatment. Because our actions have social meanings, who
we are and thus our sources of shame will be partly determined by who
others take us to be.19
This result may seem unfair. Morality sets before us the task of living
well. Now it seems that one might fail simply because of the collective
backwardness of those with whom we must interact. But this objection
rests on a mistake. Morality sets before us the task of doing the right
thing. Success on this dimension is not a function of how others receive
us. Living a moral life that is successful on all relevant dimensions, however, includes more than successfully executing this task. Consider an
analogy with teaching. Successful teaching only partly depends on successfully executing such tasks as preparing and giving comprehensible
lectures and grading fairly. It also depends on class chemistry, students
willingness to work, their interest in the subject, and the like. These are
not tasks; teaching may fail owing to these non-task factors. In this case,
the only way to avoid the conclusion that one has failed as a teacher is to
reduce teaching to a set of tasks performable by the teacher alone. But
this approach wrongly treats teaching as an individualistic enterprise.
Teaching is a fundamentally social activity. Successful participation in
this social activity depends both on the parts that are up to oneself and
the parts that are up to others. So, too, in the case of morality.
If success depends both on what is up to us (correct task execution in
getting it right) and on what is up to others (reception of what one does
as an intelligible part of a shared scheme of social cooperation), then
there is no guarantee that both moral ideals will be realized simultaneously. Indeed, as I have been suggesting, when getting it right requires
repudiating shared moral understandings, success on this dimension
may produce failure on the other.
Central features of moral philosophizing often work to obscure both
the ideal of participating in a shared scheme of social cooperation and
the possible conflict between this ideal and that of getting it right. The
19. In chapter 2, I discuss and critique the idea that the mature, autonomous agent should
only feel shamed by moral criticisms she regards as correct and not also shamed by who
she is in the eyes of fellow practitioners, even if she regards their moral criticisms as
misguided.

44

MOR A L FA I LU R E

(perfectly appropriate) focus on determining correct moral principles

and adequate justifications can easily lead to the impression that getting
it right is all that matters. Moral philosophizing may also make the reception of ones moral activity seem far less problematic that it actually
is. Using hypothetical worlds peopled by reasonable beings who would,
of course, agree to correct moral principles is one way of doing so. Relying heavily on shared intuitions is another way.20 Eschewing radical
social critique in favor of moral fine-tuning or more theoretical topics
is yet another.
In a different way, the highly discursive nature of philosophical practice works to obscure how problematic real-life moral communication
may be. Making ones choices legible, and ones reasons intelligible as
justifying reasons, is much more likely of success when choices and reasons can be carefully formulated in essays and books. As a philosopher,
for example, Claudia Card can devote an entire chapter to articulating
what she means by lesbian and thus what she is morally up to when
she stands up for her lesbian life. As a participant in the daily practice of
morality, she does not have this luxury. As a result, exercises of integrity,
like hers, that involve rejecting conventional meanings are bound to be
and will remain illegible to others as an exercise of integrity or even as
minimally justified.
In making moral communication with others seem unproblematic,
moral philosophy obscures the possibility that our moral lives will not
in fact be conducted within a minimally shared scheme of social cooperation and that our moral practice will be an idiosyncratic performance. The point here is not that there is something wrong with moral
philosophy. The point is that central features of moral philosophizing
make it difficult to entertain thoughts about what is happening to moral
resisters lives in abnormal moral contexts as they try to get it right. Thus
we arent invited to think about what we would say about such a life,

20. Shared here amounts to one of two things. Moral arguments draw either on the most
conventionalized and socially legitimated moral beliefs (for example, that it is wrong to
inflict gratuitous suffering), or on moral beliefs that are shared by those who also share
the philosophers gender, race, and class location (for example, the belief that contractors in the Rawlsian original position should, of course, be heads of household).

45

MOR A L A IMS

especially if it were our own. Would we think it tragic that a life devoted
to doing the right thing was incomprehensible to others or vilified as
perverse, irrational, or immoral?21 Would we think our lives had turned
out as moral lives are not supposed to turn out? Would our pride in doing
the right thing be spoiled with shame for the other, social meanings of
our actions? And would we find it forgivable, because it is understandable, if someone chose participation in a common moral life over doing
the right thing?
The answer to all these questions seems to me to be yes. Indeed,
there would be something perverse about a person who cares only about
how things would go between herself and others in a hypothetical, morally more perfect social world, and who is morally untroubled by the
fact that in her actual exchanges with others she is received as arrogant,
unfair, ungrateful, selfish, uncivil, and intolerant. 22
In sum, moral revolutionaries are to be admired for their commitment to doing the right thing. Even so, their lives will be, in part,
moral failures. It is an element of the tragedy of morally ill-formed social
worlds that the morally best will have reason to regret how their lives
turn out morally.

21. I owe to my former colleague, Jill Gordon, the idea that such a life is tragic.
22. One might try to capture the idea that it matters how our actual exchanges with others
go by placing moral value on sustaining relationships. Unfortunately, this strategy factors considerations about how others will receive us into decisions about what the right
thing to do would be. Others misguided responses, however, should not be decisive in
decisions about what morally ought to be done.

46

C ha pt e r 2

An Apology for Moral Shame

In daily life, we often use shaming criticisms like Shame on you, I expected more from you, or What kind of a friend are you? to impress
our moral expectations on others. We expect people to feel ashamed of
being petty, of snooping through others mail, of refusing to do a simple
favor, or of pretending not to see people who want a moment of their
time. We think that some people who arent ashamed should belike
the many young men in my classes who shamelessly confess to misdirecting lost people. In short, in our everyday exchanges with other
people, we act as though moral shame over bad behavior is simply to be
expected of any mature moral agent. Occasionally, philosophers have
also thought that shame is important to the life of a moral agent. In a
1939 article published in Ethics, Virgil Aldrich suggested that feelings of
shame provide us with much more reliable and concrete moral guidance
than does any appeal to principle or consequences. He urged the adoption of a new Golden Rule: Never do anything which you would feel
ashamed to do and always do what you would feel ashamed not to do.1
Generally, however, philosophers have not shared this positive outlook on shames moral importance. More often, one finds analyses of
shame that displace it from any significant role in morality. Shame, it is
said, is not a response to moral wrongdoing, but something we feel when
weve had a shock to our self-esteem, or when we discover our shortcomings in relation to our ego ideal, or when our failure to be or act as befits
our station in life is publicly exposed. 2 Or, if shame is a moral emotion,
1. Virgil C. Aldrich, An Ethics of Shame, Ethics 50 (1939): 5777.
2. John Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press, 1971); John
Deigh, Shame and Self-Esteem: A Critique, Ethics 93 (1983): 225245; Gabriel Taylor,
Pride, Shame, and Guilt (Cambridge: Cambridge University Press, 1985); Gerhart

47

MOR A L A IMS

it is a more primitive and less useful moral emotion than guilt, one that
both cultures and individuals would be better off moving past. 3 That is
because shame seems less directed at the wrong done than at how we
appear, or how others will receive us, or what good or bad opinion we are
entitled to have of ourselves. Thus what fuels philosophers suspicions
about the value of feeling ashamed is the way shame seems to shift attention away from what morality requires and toward what other people require us to do or be like. In shame, we see ourselves through others eyes
and measure ourselves by standards that we may not share. We take seriously the prospect of being subjected to ridicule, demeaning treatment,
or social ostracism for falling short of others moral standards. And we
fear being exposed as the less worthy beings they might take us to be.
The problem with shame, then, is that vulnerability to being shamed appears to signal the agents failure to sustain her own autonomous judgment about what morality requires.
Given this, it might well seem morally preferable for agents to be,
or to strive to be, insensitive to the shaming gaze of others and attentive only to the demands of their own practical reason. This might seem
particularly good moral advice for members of socially subordinated
groups. The sorts of shaming criticisms to which racial minorities, the
poor, women, Jews, lesbians and gay men, and so on are subjected often
repeat demeaning cultural stereotypes of group members moral character (for example, as lazy, or untrustworthy, or mendacious). Shaming
criticisms may also repetitively attribute diminished capacity for moral

Piers and Milton B. Singer, eds., Shame and Guilt: A Psychoanalytic and a Cultural Study
(Springfield, IL: Charles C. Thomas, 1953); Self-Conscious Emotions: The Psychology of
Shame, Guilt, Embarrassment and Pride, ed. June Price Tangney and Kurt W. Fischer
(New York: Guilford, 1995).
3. June Price Tangney argues that shame is morally counterproductive because shame
often motivates hostile aggression toward shamers, and shamed people typically fail to
empathize with the victims of their wrongdoing; June Price Tangney, Shame and Guilt
in Interpersonal Relationships, in Tangney and Fischer, Self-Conscious Emotions (113
139). John Kekes claims that shame rivets the agents attention on the defectiveness of
the self, undermining self-confidence and paralyzing action; John Kekes, Shame and
Moral Progress, in Ethical Theory: Character and Virtue, Midwest Studies in Philosophy,
vol. 13, ed. Peter A. French, Theodore E. Uehling Jr., and Howard K. Wettstein (Notre
Dame, IN: University of Notre Dame Press, 1988), 282296.

48

A N A POLOGY FOR MOR A L SH A M E

agency, thereby implicitly rationalizing cultural practices that give subordinate groups lesser moral consideration. Women of all races, for example, are more likely than white men to be criticized for irrationality,
lack of self-control, and inadequate attention to principle. They are thus
given cause to feel ashamed of their agency. Shaming criticisms may also
be leveled against anyone (whether subordinated or privileged) who
does not conform to sexist, racist, or other norms of behavior, or who
protests the demeaning or unequal treatment of social subordinates. In
short, in societies structured by relations of domination and subordination, shame is an especially worrisome moral emotion. Subordinated
people who suffer shame before bigoted criticisms seem to have failed
to achieve (or failed to be able to sustain) a sufficiently critical moral
perspective. They are thus not well positioned to challenge the gender,
racial, sexual, and religious politics that erode their life prospects. The
subordinate would be better off ignoring others shaming criticisms and
shamelessly pursuing more egalitarian ideals.
In what follows, I plan to side with our everyday assumptions about
the importance of feeling morally ashamed. I think that shame over
moral failings is essential to a mature ethical agents psychology. More
controversially, I think that vulnerability to feeling ashamed before
those with whom one shares a moral practice, even when one disagrees
with their moral criticisms, is often a mark of moral maturity. It need not
spring from any failure in autonomous judgment. Thus, I think it is possible to understand the pervasiveness of shame among socially subordinate groups without attributing to them either internalized contempt
for their own social group or a failure to maintain their own critical perspective in the face of others shaming contempt.

I. SH A M E A ND AUTONOM Y: T WO STR ATEGIES

OF R ECONCILI ATION
Given the worry that shame signals heteronomous and excessive concern with others opinions, any good defense of moral shame will have to
show that, despite appearances, moral shame is in fact compatible with
autonomous moral judgment. The most obvious strategy is to argue that
49

MOR A L A IMS

morally mature persons only feel shame over their failure to live up to
their own, autonomously set standards.
There are two fairly different ways of conducting this game plan.
One is represented by such authors as Anthony OHear, John Kekes,
and Virgil C. Aldrich.4 I will call this the shame of the moral pioneer
strategy. This strategy reconciles shame with autonomy by claiming that
mature agents only feel shame in their own eyes, and only for falling short
of their own, autonomously set standards. The second strategy is represented by Bernard Williams in Shame and Necessity. I will call this the
shame of the discriminating social actor strategy. This strategy reconciles shame with autonomy by claiming that mature agents only feel
shame in the eyes of others whose ethical reactions she respects.
Neither strategy, I will argue, satisfactorily reconciles shame and
autonomy. Both make shame suitable for an autonomous agent only by
reducing the other before whom we feel shame to a mirror of ourselves.
Both drop from view the fundamentally social nature of shame. I will
then suggest, in section II, a third strategy that I think works better.
In pursuing it, I will be stressing the fact that morality is something
practiced with others in a social world. Taking others seriously as coparticipants in a moral practice means giving their opinions weight
and thus the power to shame.

The Shame of the Moral Pioneer Strategy

Lets start, then, with the first strategy. Those who adopt the shame of
the moral pioneer strategy claim that shame does not require a real or
imagined audience before whom one might feel shame. It is not the eyes
of others that matter. What matters for the experience of shame is that
there be some moral standards that one cares about living up to. And
the mature ethical agent will only care about standards that she autonomously sets for herself. Those standards could, of course, be those of
a conventional social morality so long as the agent adopts them for her
4. Anthony OHear, Guilt and Shame as Moral Concepts, Proceedings of the Aristotelian
Society 77 (19761977): 7386; Kekes, Shame and Moral Progress; Aldrich, An
Ethics of Shame.

50

A N A POLOGY FOR MOR A L SH A M E

own. Those standards could also, in theory, be entirely idiosyncratic

ones that no one else shares. They could, for example, be the standards
of a lone moral pioneer whose moral vision outstrips that of her social
world. Such a lone pioneer would be shameless before her less enlightened social peers and ashamed only before her private moral court. In a
sense, however, all shame on this view is simply shame before oneself.
That is because, on this view, an agent will experience anothers moral
criticisms as shaming only if he already endorses the others standards
and specific criticism. 5
This way of thinking about the mature ethical agents shame fits
nicely with a standard way of contrasting moral immaturity with moral
maturity. Immature moral agents submit to external moral standards.
Children, for example, begin by uncritically submitting to parental prohibitions and then later submit to what their larger social world prohibits and requires. As individuals morally mature, their submission to a
particular set of moral requirements becomes increasingly independent
of the consideration that others will approve obedience and disapprove
disobedience. Instead, the standards that have authority are increasingly based on reasons that might be shared by rational beings (but that
may not in fact be shared by those in ones social world). Full moral maturity arrives with the development of an autonomous practical reason
that enables the individual to arrive at her or his own moral standards
and specific moral judgments. Those reflectively endorsed standards
give the agent critical purchase on conventional morality.
Shame, on the view just described, traces a similar developmental
path. Childrens earliest experience of moral shame stems from being
observed violating parental prohibitions, and later, conventional moral
prohibitions. In both cases, the immature are shamed simply by others
critical gaze. Moral maturity arrives when agents learn to spurn public
opinion and think for themselves.6 Mature agents are then only shamed
when they fall short of their own standards.
5. OHear, Guilt and Shame as Moral Concepts, 77.
6. John Kekes draws a similar contrast between what he calls honor shame (shame at
violating public, conventional standards) and worth shame (the shame felt by selfdirective people who spurn public opinion in the name of private standards); Kekes,
Shame and Moral Progress, 293.

51

MOR A L A IMS

Such accounts of moral shame appear to solve the original problem.

Moral shame does not necessarily signal heteronomy. The mature ethical agent feels shame before her own autonomous evaluative gaze. Unfortunately the price of reconciling moral shame and autonomy this way
is the loss of a plausible depiction of shame.
To see this, imagine a moral pioneersomeone who has achieved a
level of moral knowledge that her less enlightened social peers have not.
If she transgresses her moral standards, she will be treating her social
peers badly by her lights. They, however, will not realize this; and no one
in her social world will think badly of her. Under these circumstances, a
moral pioneer could surely feel guilty about mistreating unwitting peers.
That they fail, out of ignorance, to protest is no safeguard against guilt.
(Indeed, her guilt at mistreating her peers might then be compounded
with guilt at getting away with it.) Proponents of the shame of the moral
pioneer account add that this moral pioneer should also feel ashamed
of herself. On their view, what others might think of her is utterly irrelevant to mature shame experiences. So she should feel ashamed even
though none of her peers do, or could, morally criticize her.
But this doesnt seem right.7 Like other forms of shame, moral shame
seems intrinsically tied to the thought of social others actual or imagined contempt. 8 Moral shortcomings must first be exposed to public
view before they can be the source of shame; or at the very least, the contempt that others would show us were our shortcoming exposed must
7. We do sometimes feel ashamed even though witnesses of the act do not think it shameful. A parent might feel ashamed of slapping his child even though he is among parents
who believe in corporal punishment. A diner at the Cattlemens Association dinner
might feel ashamed of eating the veal. Shame in these cases is possible, however, because
not everyone in ones social world believes that corporally punishing children or eating
veal are permissible. Some really would find slapping a child or eating veal contemptible.
The parents and diners shame is shame at failing to live up to the standards of these
other people. The case of the ashamed moral pioneer is different. We are supposing that
none of her social peers recognize the wrong she has done; indeed, they are unable to
recognize it.
8. John Deigh makes an especially persuasive case that a satisfactory characterization [of
shame] must include in a central role ones concern for the opinions of others; Deigh,
Shame and Self-Esteem, 238. The others, whose opinion one cares about, however,
need not be living. One can feel ashamed imagining what ones beloved grandmother
would have thought were she still living.

52

A N A POLOGY FOR MOR A L SH A M E

be clearly imaginable. The primary fears attached to shame are fears

of being ridiculed, made the subject of gossip, subjected to demeaning treatment, and of being ostracized or abandoned.9 Thus, shame is
strongly connected with the desire to conceal failings from others view,
with fear of exposure, and with anxiety about how it will be for ones
life with others if one acts shamefully.10 The first strategy for reconciling shame with autonomy cannot explain these connections. On the
contrary, it severs the connection between shame and concern for ones
standing in a social world. It does so because it mistakenly takes the
object of shame to be what the agent alone believes is a moral failing. The
real objects of shame, however, are failures to meet moral standards that
are also held by other people. Shaming moral failures are paradigmatically ones that might, if exposed, reduce ones social standing in some
actual group and might degrade the quality of ones social interactions.
Proponents of the shame of the moral pioneer strategy could try
to defend the plausibility of their picture of shame in the following way.
The moral pioneer, they might argue, imagines a hypothetical social
world where her enlightened standards are publicly shared. She thus
imagines people before whom she could feel shame and among whom
she would lose standing were her failings exposed. In this way, shame is
reconciled with autonomy: the standards she is ashamed of failing are
hers. And autonomy is reconciled with concern about how one appears
to others: people in this imagined social group judge her by the same
standards.
This reconciliation, however, is more apparent than real. The only
opinions of others (real or imagined) that the autonomous person will
be shamed by are ones that she shares. When the moral pioneer says, I
can imagine social peers whose criticisms would shame me, she must
be prepared to add, if she is in fact an autonomous judge, But I would
not feel shame if I did not share their view. That is, in order for the
mature agents shame to remain firmly tied to her own autonomously
chosen moral standards, the others before whom she can imagine
9. Piers and Singer, Shame and Guilt.
10. Bernard Williams, Shame and Necessity (Berkeley, CA: University of California Press,
1993), 102.

53

MOR A L A IMS

feeling shame cannot be thought of as persons with their own minds

that they might make up differently. They cannot be imagined as full
others. Instead, they must be imagined to be people who reach the same
moral conclusions she does because their minds mirror her own. They
are simply stand-ins for her own reactions to herself.11 It is only because
others mirror the agent that feeling shamed by their moral criticisms
does not threaten her autonomy.
In sum, the shame of the moral pioneer account seeks to make
shame and autonomy compatible in order to explain why moral shame
would be part of the mature ethical agents emotional repertoire. But the
particular strategyof claiming that mature agents are only shamed by
criticisms that mirror their own self-criticismsuggests that shame
is no more tied to the social than is guilt. Mature shame, like mature
guilt centers on how one appears in ones own eyes. Concern about how
one appears in others eyes, fear of having discrediting facts socially exposed, anxiety about others contempt and about having ones social relations impaired may accompany shame. But on this view they are not
central to and distinctive of shame experiences (or at any rate, to mature
shame experiences).

The Shame of the Discriminating Social Actor Strategy

In light of the implausibility of detaching shame from an awareness of
how we appear to others, Bernard Williams pursues a different strategy
for countering the objection that shame signals excessive heteronomy.
Central to this strategy is insistence that shame is always shame in the
eyes of real social others. Williams suggests that it seems difficult to
place this social shame in the mature ethical agents psychology because
we have inherited from Kant a tendency to oppose capital-M morality
to conventional standards, individual judgment to social judgment, and
autonomy to heteronomy. Thus we end up thinking that there are only
two options: either we are autonomous, set moral standards for ourselves, are unconcerned about the opinions of others, and are invulnerable to shaming criticisms; or we fall into heteronomy, uncritically take
11. Ibid., 84.

54

A N A POLOGY FOR MOR A L SH A M E

on whatever moral standard our social peers use, care deeply about how
we appear to others, and are vulnerable to every shaming criticism.12
There is, however, a third option. We can reconcile the essentially
social nature of shame with autonomy if we keep in mind that at any
point in time there typically are many ethical standards endorsed
within ones social world. Immature or poorly developed agents do not
discriminate among these standards and thus may be shamed by virtually every moral criticism levied at them. But ethically well-developed
agents care only about the opinions of some social others. They choose
whose standards to respect and thus whose eyes will have the power to
shame. If they are to be worthy of respect, those standards will largely
mirror the agents own. Thus criticisms that the agent finds outlandish
or irrational because they fail in important ways to mirror her own evaluative standards and canons of reasoning will not inspire shame.13 It
does not, however, follow that the criticisms that do inspire shame must
mirror the agents own self-criticism. As I interpret Williams, he leaves
open the possibility that respected others can shame us with their criticisms even when we disagree with their evaluation of us. Mature moral
agents care about how they appear in the eyes of respected others. They
care how they appear because they have a general respect for others
evaluative commitments, skill at moral reasoning, and perceptiveness.
That general respect grounds the power to shame, and thus people may
be shamed by particular criticisms that fail to mirror their own selfcriticism. In this way, Williams allows for the evaluative gaze of others,
fear of exposure, anxiety about others contemptthat is, the social
dimensions of shameto play a central role in his account of shame
independent of our own self-assessments. At the same time, mature
shame experiences are ultimately tethered to the agents own evaluative standards, since she must choose whose evaluative gaze merits her
respect. So while it is true that shame is always shame in the eyes of
real social others who will interact with her differently if her moral failing is exposed, vulnerability to shame does not entail an abdication of
12. Ibid.
13. Both Jeffrey King and Jennifer Vest proposed this way of thinking about who has the
power to shame.

55

MOR A L A IMS

individual judgment. On the contrary, individual judgment is critical to

ethically mature shame.
What distinguishes the discriminating social actor strategy from
the moral pioneer strategy is, first, that the eyes of (some, respected)
others have the power to shame independently of their exactly mirroring
the agents eyes. Second, it matters that the others before whom I might
be shamed are real social others, not merely imagined others. Williams
refuses to disconnect shame from the eyes and standards of people in
our lived social world. In theory, a reflective agent might decide that
none of the standards instantiated in her social world deserves respect.
If no eyes merit her respect, then no gaze within her social world will
have the power to shame her. He thus leaves open the possibility that
the autonomous choice of standards might result in an inability to feel
shame. But Williams argues that agents have good reason not to divorce
themselves altogether from social standards of morality, and thus not
to become shameless. That reason is provided by the indeterminacy of
moral truth:
But if we now think, plausibly enough, that the power of reason
is not enough by itself to distinguish good and bad; if we think
yet more plausibly, that even if it is, it is not very good at making
its effects indubitably obvious, then we should hope that there is
some limit to these peoples autonomy, that there is an internalized other in them that carries some genuine social weight. Without it, the convictions of autonomous self-legislation may become
hard to distinguish from an insensate degree of moral egoism.14

Thus, because moral opinions do not come clearly labeled correct

or incorrect, an agent who comes to moral conclusions that diverge
substantially from any endorsed within her social world has no way of
telling for sure whether she is a moral revolutionary whose views advance the social stock of moral knowledge or a deluded crank. Agents
thus have good reason to give significant weight to the best among the
available standards already flourishing within their social world. In this
14. Williams, Shame and Necessity, 100 (emphasis mine).

56

A N A POLOGY FOR MOR A L SH A M E

way, vulnerability to shame ends up securely anchored in the mature

ethical agents psychology.
I think Williams is right to insist that shame is a social emotion. Neither our own solitary gaze nor the gaze of purely fantasized others can
trigger shame. But this second strategy for reconciling shame and autonomy is ultimately vulnerable to the same criticism that the first strategy
was: the distinctively social character of shame and the power of others
eyes to shame is not adequately accounted for.
On this second strategy, the mature ethical agent reflects on others
standards and decides whose she respects. But in order to evaluate
others standards, she must already have standards of her own. This
means that others criticisms have the power to shame only because
they invoke standards that she already largely endorses. Notice that
others power to shame turns entirely on the question of what their
standards are. Are they ones the agent endorses? Thus the standard
by which the agent is judged is what explains whether she will or will
not be vulnerable to being shamed. That she is exposed to another who
views her with contempt and will now interact with her differently
plays no independent role. It is true that Williams aims to give an account of shame where others opinions have social weight; he also
seems to allow for the possibility that others particular criticisms of us
will have this social weight, and hence the power to shame, even when
we reject those criticisms. But in the end, what it means for anothers
opinion to have a social weight independent of ones endorsement remains unexplained. On the contrary, because this view, like the first
strategy, traces the power to shame to the shamers mirroring to a large
extent the agents own evaluative perspective, it is not clear why moral
criticisms with which one disagrees would have any power to shame
at all. That I share anothers standards may explain why, in general, I
respect her and care how I appear in her eyes. But it does not explain
why I would respect her and care about how I appear in her eyes when
she misjudges me. Since on this view, as on the first strategy, the power
to shame ultimately comes from my endorsement of the others evaluative outlook and not from the fact that an other appraises me, it is hard
to see why particular moral criticisms shame an agent who does not
endorse them.
57

MOR A L A IMS

In short, any strategy that reconciles shame with autonomy by rooting the power to shame in the agents endorsement of the shamers evaluations will have trouble capturing shames distinctively social character.
If shames more social nature than guilt is to be explained, then we
need to trace the power to shame to a more social source. What could it
mean for an opinion to have social weight independent of the agents
endorsement of the shamers standards and particular criticisms? And
how is sensitivity to the social weight of others opinions compatible
with being an autonomous moral judge?
Before turning to these questions, it is worth looking at one final
problematic feature of the two strategies for reconciling shame and autonomy considered so far.

Shame Before the Others Unmirrored Gaze

Grounding the mature agents shame in her own self-criticism not
only fails to explain the social character of shame; it also forces us to
discount as irrational or immature quite common shame experiences.
On the discriminating social actor view, and even more so on the
moral pioneer view, a man who sees nothing wrong with his purchasing sexual services yet suffers shame when the newspaper prints his
name in a list of Johns is not experiencing the predictable shame of
a mature, well-formed ethical agent.15 A mature, well-formed ethical
agent would only feel shamed by moral criticisms that mirror his own,
or that at least invoke ethical standards he respects. More worrisome,
we must discount as irrational or immature much of the shame suffered by socially disesteemed populationsracial minorities, women,
the poor, lesbians and gay men. As Sandra Bartky observes, shame, for
the subordinated, is the pervasive affective taste of a life in which
15. Some advocate using shaming penalties as an alternative to prison time for some
crimes; see Dan M. Kahan, What Do Alternative Sanctions Mean? University of Chicago Law Review 63 (1996): 591653. Shaming penalties seem to presuppose that it is
natural for mature moral agents to be vulnerable to shaming criticisms even when they
see nothing wrong with what they have done.

58

A N A POLOGY FOR MOR A L SH A M E

being subjected to shaming treatment is a routine part of social interaction.16 Some of the shaming criticism is specifically moral, as
when black men are routinely suspected of being shoplifters or muggers, or when the poor are assumed to have brought poverty on themselves through their own laziness or lack of self-control; some of the
shaming criticism is not specifically moral, as when female and black
students are presumed to be less educable. Pervasive shame often coexists, however, with a denial that there is anything to be ashamed of.
The women students that Bartky so poignantly describes are a case in
point. At a discursive level, they believe their work is meritorious. They
would deny that they are unintelligent or unable to compete academically. They would not respect the attitudes of teachers who ridicule or
demean them in class. However, having been regularly demeaned in
the classroom throughout their educational lives, they have come to
be ashamed of their work, ashamed to express their ideas, and fearful of incurring punitive or demeaning treatment in the classroom.
Shame moves them to apologize for the quality of their work, to express themselves without confidence, to bow their heads and hunch
their shoulders.
The two strategies we have considered must both conclude that these
women are not experiencing the predictable shame of a mature, wellformed ethical agent. On both views, shaming classroom experiences
would not faze a rational, mature person convinced of her own academic
talent. She would ignore or scoff at her teachers contempt. The fact that
these women feel shamed while claiming to believe in their academic
talent shows that there is something awry with them. Perhaps they do
not know their own minds; deep down they really do believe they are
unintelligent and unable to compete. Or perhaps they lack strength of
mind; they succumb to others opinions and abandon their own view of
themselves. Or perhaps the problem is what Bartky suggests it is: they
hold inconsistent views of themselvesone at the discursive level of

16. Sandra Lee Bartky, Shame and Gender, in Femininity and Domination: Studies in the
Phenomenology of Oppression (New York: Routledge, 1990), 96.

59

MOR A L A IMS

belief, one at a nondiscursive level of feeling.17 Whatever the diagnosis,

the conclusion is the same. No rational, mature person who firmly rejects her subordinate social status would feel shame in the face of sexist,
racist, homophobic, or classist expressions of contempt.18 The two views
we have considered so far thus encourage us, at best, to seek out psychological explanations for these irrational shame responses; and at worst,
to chastise the subordinated for feeling ashamed and to exhort them to
buck up, think for themselves, be more thick-skinned, and spurn public
opinion. That is, they encourage us to find fault with ashamed people.19
This strikes me as both uncharitable and the wrong conclusion. It
may be true that Bartkys women students feel shame because they fail to
sustain their own positive self-evaluation. Their shame and apologetic,
self-effacing behavior are consistent with a low self-evaluation. Feeling
ashamed, however, does not entail that they agree with their shamers
contempt. That people who wholeheartedly condemn sexist or racist insults are still vulnerable to feeling shamed by those insults, and that this
is a perfectly natural response for a mature, well-formed agent to have, is
made painfully clear by Adrian Piper in her narratives of being shamed
in academia.20 She gives us every reason to believe (as Bartkys students
do not) that she is perfectly capable of sustaining her confidence in her

17. Ullaliina Lehtinen, How Does One Know What Shame Is? Epistemology, Emotions,
and Forms of Life in Juxtaposition, Hypatia 13 (1998): 5677, proposes a similar analysis of the shame of the oppressed. Following Bartky, she argues that women are less
likely than men to be able to autonomously defy shamers judgments because women
have internalized a low self-evaluation. Thus for women and members of other subordinate groups, experiences of shame function as confirmations of what the agent knew all
alongthat she or he was a person of lesser worth (62). Notice here how the pervasive
shame experienced by members of subordinated groups is explained by attributing low
self-esteem to them.
18. Someone who insists that rational, mature people would not feel shamed by criticisms
they reject might nevertheless think that a rational, mature person could experience
some other unpleasant feeling, such as discomfort at the awkwardness of having to interact with openly sexist or racist people. One need not agree with a would-be shamers
contemptuous views to be made uncomfortable by them.
19. The fault here need not be culpable.
20. Adrian Piper, Passing for White, Passing for Black, in Passing and the Fictions of
Identity, ed. Elaine Ginsberg (Durham, NC: Duke University Press, 1996), 234270.
Thanks to Gary Watson for drawing my attention to this piece.

60

A N A POLOGY FOR MOR A L SH A M E

own worth, no matter how insultingly she is treated. By her own account, Piper was raised by parents who tried to provide her with an invincible armor of self-worth with which to fight [racism]. 21 As a result
her instinctive reaction to racist insults is the disbelief, outrage, sense
of injustice, and impulse to fight back actively that white males often
exhibit at unexpected affronts to their dignity.22 Pipers firm belief in
her own worth and in the unacceptability of racism does not, however,
protect her against shame. Here is one of her stories:
Refusing to pass as white, although she could, Piper identifies herself
on graduate school applications as black. When she shows up at the reception for new graduate students, she is approached by a professor, one
of her intellectual heroes, who remarks with a triumphant smirk, Miss
Piper, youre about as black as I am.23 This is one of a series of occasions
on which Piper feels what she calls groundless shame.24 Her shame is
groundless because she does not share her shamers view that she is manipulative or deceitful. The moral pioneer strategy would thus have to
conclude that Pipers shame reaction does not befit a mature moral agent.
One might also imagine that she does not share with him more general normative views about what the moral point of affirmative-action
policies is and who is permitted to present themselves as an affirmativeaction candidate; about what integrity requires from persons who could
pass as the bearer of a less discrediting identity; about what general sorts
of actions a commitment to resisting racial oppression requires; and
about the proper application of evaluative concepts like honesty, integrity, manipulativeness, and racism. Of course, Piper and the professor
share some evaluative views in commonfor example, the wrongness
21. Ibid., 239.
22. Ibid., 260.
23. Ibid., 234.
24. On other occasions, she feels shamed by demeaning assumptions about what can be
expected of her as a black person. She is assigned to remedial classes in anticipation of
poor performance, complimented by a fellow grad student on her good English, and
interviewed for four and a half hours by a colleague to ascertain whether she was smart
enough to hold the position she already had. In response, she experiences the groundless shame caused by people who, having discovered my racial identity, let me know
that I was not comporting myself as befitted their conception of a black person; Ibid.,
258 (emphasis mine).

61

MOR A L A IMS

of deceit and manipulation, understood very abstractly. But absent more

robust commonalities, especially ones connected to Pipers antiracist
commitments, agreement that deceit and manipulation, abstractly described, are wrong looks insufficient to ground respect for the professors standards that would make her care how she appears in his eyes.
Thus the discriminating social actor view must also conclude that Pipers shame reaction does not befit a mature moral agent.
Yet shame, not just social discomfort, seems the reasonable response
to being treated with contempt by people whose evaluations partly define
who these women are within a shared educational practice. Rather than
signaling a failure to sustain their own positive views of themselves, their
shame instead signals their capacity to take seriously fellow participants
in their social world. I turn now to the question of what it might mean to
take others seriouslyto give their opinions weight.

II. THE W EIGHT OF OTHER S JUDGM ENTS

One might be skeptical that Bartkys and Pipers examples are really examples of mature shame experiences. These shame experiences might
be psychologically understandable. But we expect mature agents to be
autonomousthat is, to rely on their own judgment. It is thus tempting
to say to Bartkys students and to Piper, You should not be shamed by
the wrongheaded opinions of others. After all, in your own view, you
have nothing to be ashamed of. Spelling out this skepticism more formally, we have the following familiar and compelling argument:
1. Moral shame is made possible by the fact that we take seriously
others appraisals of our moral shortcomings; we give them
weight.
2. The mature ethical agent is an autonomous self-legislator; this
means that mature agents submit only to the demands of their
own practical reason.
3. Mature ethical agents thus have reason to take others appraisals of their shortcomings seriouslyto give them weight
only if those appraisals mirror the agents reasoning.
62

A N A POLOGY FOR MOR A L SH A M E

4. Thus, mature ethical agents will only feel shamed by the appraisals of others when those appraisals mirror the agents own
reasoning about her shortcomings.

The shame of the moral pioneer account draws quite clearly on this
line of reasoning. The shame of the discriminating social actor account
draws on a kindred line of reasoning, because it invokes the importance
of agents determining for themselves whose practical reasoning they respect enough to accept even when it might diverge from their own.
What could be wrong with the argument? Premise one simply affirms a principal feature of shamenamely that it is a fundamentally
social emotion. Premise two affirms an equally uncontroversial connection between moral maturity and autonomy. That mature agents
submit only to the demands of their own practical reason does not mean
adopting a policy of dismissing others views whenever they disagree
with ones own. It does mean that ultimately agents must make up their
own minds; and doing so may mean reaching the sincere conviction that
others are wrong and one is going to stick to ones guns. Bartkys students, and even more obviously Piper, are autonomous self-legislators
in this sense. Exposed to repeated critical messages, they make up their
own minds not to believe those messages.
This leaves premise three: mature ethical agents have reason to
take others appraisals of their shortcomings seriouslyto give them
weightonly if those appraisals mirror the agents reasoning. At first
glance this premise seems to follow from the thought that mature agents
are self-legislators who make up their own minds. After all, if Ive made
up my mind that others criticisms are misguided, how could I give those
criticisms weight? Wouldnt giving them enough weight to shame me
amount to giving in to others views of me?
I do not think so. That a person can only be shamed by a view of
herself that she accepts as true, at some level, is not supported by everyday experience. It is instead a lacuna in our moral theories that makes it
seem that, in order for an opinion to have shaming weight, we must at
some level accept it as true. Moral theories are typically slanted toward
moral epistemology, and this induces us to think that weight must be
an epistemic notion. It must have something to do with the weight of
63

MOR A L A IMS

reasons, a weighty argument, or the compelling force of truth. The assumption that weight is an epistemic notion then drives us toward the
idea that others opinions having weight amounts to those opinions
having weight in our reasoning process. But if they have weight in our reasoning process, we must have accepted their truth.
Moral agents, however, are not just knowers. They are also participants in various social practices of morality with other people. What I
want to suggest is that the weight central to shame is not an epistemic
notion. It is instead the weight that other people have for us when we acknowledge them as fellow social participants. That anothers view of us
has weight in this latter sense is compatible with denying its truth.
Moral criticism that shames has what I will call practical weight.
Moral criticism has practical weight when we see it as issuing from those
who are to be taken seriously because they are co-participants with us in
some shared social practice of morality.
Co-participants stand in a different relation to us than do agents in
general. Agents in general are responsible beings, open to reason and
capable of exhibiting goodwill. We take their moral interpretations seriously by listening to what they have to say, engaging in moral dialogue,
taking care not to give offense needlessly, and so on. Co-participants
are more than this, and we take their moral interpretations seriously in
a thicker, more substantive way.25 Co-participants are part of a moral
we that shares a social practice of morality. 26 That social practice generates shared understandings about exactly what is obligatory and what

25. Taking agents seriously by listening to what they have to say, engaging in moral dialogue, and taking care not to give offense is compatible with denying that others have
the standing to criticize us. A pro-choice woman might, for example, take seriously a
religious conservatives condemnation of abortion by civilly listening and responding to that view. In the thicker more substantive sense, the pro-choice woman does
not take seriously the religious conservatives moral appraisal. Because his views on
abortion derive from different social practices of morality, the pro-choice woman has
no reason to acknowledge the religious conservatives standing to criticize her reproductive choices and call her to account. The accusation, You murdered your unborn
child, does not define who she is for others within the social worlds she claims as her
own. As a result, it lacks the power to shame.
26. The following discussion owes a good deal to Margaret Urban Walkers Moral Understandings: A Feminist Study in Ethics (New York: Routledge, 1998).

64

A N A POLOGY FOR MOR A L SH A M E

is supererogatory, as well as shared understandings about how to interpret when basic moral obligations, like the duty of truth-telling, have
been fulfilled. Physicians, for example, generate shared understandings
about the required level of disclosure to patients and about the moral
status of hastening death. Co-participants are thus able to engage in a
shared enterprise of evaluating each others behavior and character, determining who has lived up to and who has fallen short of shared moral
ideals, and calling each other to moral account for transgressions.
A social practice of morality comes about because there is something else that we want to do togetherwork in a profession, engage in
religious worship, play sports, live in a neighborhood, have a marriage.
These various activities are sites of particular moral problems that produce the need to generate shared moral norms. The practice of education, for example, produces a need for norms governing student-teacher
relations, including sexual relations. The practice of medicine generates a need for norms governing the response to terminal illness. Those
moral norms then get hammered out among people who already share
a social world.
In everyday life, we move between a plurality of moral practices.
Each has its own shared understandings about how we do things
ones family of origin, the family one creates as an adult, ones workplace, ones profession, ones neighborhood, ones political association,
ones religion, and so on. All these groups engage in the business of negotiating and articulating moral norms. This is how we do things in our
family: we spend major holidays together; this is how we do things in
our profession: we do not have sexual relations with students (or clients
or patients); this is how we do things in our religion: we do not divorce;
this is how we do things in a participatory democracy: we engage in civil
dialogue.
Shaming criticisms work by impressing upon the person that she has
disappointed not just one individuals expectations but also the expectations of some we. In effect they say, You claim to be one of us, but just
look how youre behaving! The power to shame is a function of our sharing a moral practice with the shamer and of recognizing that the shamers opinion expresses a representative viewpoint within that practice. The
shamers opinion tells us who we are for any number of co-participants
65

MOR A L A IMS

within a social practice of morality that we take ourselves to be a part of.

Shaming criticisms have, in this sense, practical weight.
An extended example may help clarify this notion of practical
weight. Imagine an agent who finds herself seen critically by a faculty
colleague. Perhaps he lets her know that he thinks she is manipulative
of colleagues or insensitive to student needs. Suppose, now, that on
reflection she rejects this assessment; he has misinterpreted the facts,
or perhaps his conception of these vices is flawed. Given the opportunity, she may try to remove the criticism by trying to change his mind.
She argues with him. She lays out the facts. She invites him to rethink
what he imagines manipulativeness or insensitivity to be. But this may
not be effective. He disagrees. Perhaps he thinks her arguments confirm his very point (look how she tries to manipulate him in this argument!). She is now left with the fact that she is for him this morally
flawed person.
Faced with his unchangeable critical gaze, she has two options.
First, she could shift out of the co-participant attitude, refusing to take
him seriously. He is paranoid, she might tell herself, and thus is unable
to be a competent judge. Or he is new to this profession or this school,
and thus is not yet a competent judge. In short, he is only someone to
be humored, or resignedly suffered, or avoided, or written off. His criticism is deactivatedit now lacks practical weightbecause he is not
someone to be taken seriously as a competent participant in this social
practice of morality. Dismissed as pathological or an outsider, his critical gaze cannot represent a general viewpoint that any number of colleagues might take. Thus what he thinks of her cannot define one of the
(shameful) ways she is for others in her social world. His gaze lacks the
power to shame because it is nonrepresentative.
Alternatively, she could sustain the co-participant attitude even
though she thinks he is wrong. She continues to take him seriously as a
person who has the standing to criticize her. She does not write him off
by attributing his misjudgment to some flaw or inexperience that undermines his standing as a co-participant in this moral practice. Instead,
his is the sort of misjudgment co-participants just do make. One of the
permanent hazards of engaging in a social practice of morality is that
one ends up being criticized, and sometimes ridiculed, by people whose
66

A N A POLOGY FOR MOR A L SH A M E

moral appraisals one does not agree with. The hazard is a serious one
because their appraisals may be representative ones. Who they take us
to be represents what any number of fellow co-participants would take
us to be. Their eyes define one of the many public identities that we have
for others within our shared social practice of morality. Some of those
public identities are shameful ones.
In sum, I am proposing that vulnerability to shame has more to do
with our sharing a moral practice with others than it does with our accepting anothers criticism. Of course, when we share a moral practice
with others we typically share certain basic moral values, principles,
and styles of reasoning with them. Thus there usually is something in
our shamers evaluative thinking to which we attach epistemic weight.
Bartkys shamed students, for example, surely shared their teachers
standards of academic excellence. We need to be careful, however, not
to infer that others power to shame derives entirely from the fact that
we endorse their evaluative framework. Sharing basic values and reasoning styles with other people does not explain why particular criticisms are felt to be shaming, since we may reject particular criticisms
while endorsing the underlying evaluative framework. To explain how a
person could be shamed by a criticism that itself has no epistemic weight
because she thinks it is plain wrong, we will need to appeal to something other than a shared evaluative worldview. I have suggested that
this something else is the fact that, in sharing a moral practice with us,
others views come to have practical weight in the sense that they articulate moral interpretations of our character and actions that any number
of others within the practice might share.
At this point, defenders of the first two strategies might object that
my view does not seem very distant from their own. For them, vulnerability to feeling shamed hinges on which evaluative framework the agent
endorses. They might go on to observe that endorsing the same evaluative framework that others do is just what makes us co-participants in a
shared practice of morality. And while we may need to bring in something else to explain why particular criticisms shame, the fact that we
endorse one evaluative framework rather than another sufficiently explains why some people have the power to shame us and other people do
not. We give practical weight to some criticismsincluding ones whose
67

MOR A L A IMS

truth we denyprecisely because they issue from people whose evaluative commitments, skill at moral reasoning, and perceptive judgment
command our respect. In short, defenders of the first two strategies
might give up the idea that particular criticisms must mirror the agents
self-appraisals if they are to evoke shame. But they might still insist that
the power to shame depends on the shamers evaluative and epistemic
commitments mirroring the agents own. 27
I agree that, typically, co-participants in a practice will endorse the
same basic values and style of reasoning. But those who share a moral
practice do not necessarily endorse the same evaluative standards. First,
people do not usually choose to enter a practice of morality because they
endorse its evaluative commitments and reasoning style. What people
choose is to do something else with othersfor example, to work in academia. That choice moves them into an ongoing moral practice. 28 We
thus come to share multiple practices of morality with multiple groups
of others for reasons often having little to do with our individual evaluative judgments.
Second, as Margaret Walker has argued in her work on shared moral
understandings, [t]o share terms in this sense need not mean that the
terms in force are endorsed by all, much less that they exist by the consent of all who are required to recognize and respond to them. 29 Members of subordinate groups quite often reject substantial chunks of the
evaluative commitments, styles of reasoning, and assumptions about
group difference embedded in the dominant social practice of morality.
Even so, that dominant practice of morality generally continues to be
one of the moral practices that members of subordinate groups share.
To share a social practice means that one finds its moral understandings
intelligible, even if not endorsable. One understands how people could
come to think this way about moral matters. One understands what

27. This objection is a variant of one proposed to me by Jeffrey King. His particular concern was that what I call practical weight might ultimately reduce to epistemic weight.
28. Physicians, for example, do not choose to enter the social practice of morality that
dominates the medical world. Rather, they choose to practice medicine with others
and thereby find themselves (re)located in an already ongoing moral practice.
29. Walker, Moral Understandings, 63.

68

A N A POLOGY FOR MOR A L SH A M E

counts for others as acting responsibly, being truthful, being honorable,

giving good moral advice, and so on. 30
In short, the moral pioneer and discriminating social actor
strategies place explanatory weight on the fact that the shamed person
endorses something about the shamers evaluative perspective. I think
the explanatory weight more properly belongs on the representativeness
of the shamers viewpoint. What inspires shame is recognition of who
we are for those with whom we share a moral practice. Within a shared
practice of morality, those whose criticisms express a representative
viewpoint have the power to shame. As a result, even if in ones own view
one has nothing to be ashamed of, one may nevertheless have reason to
feel ashamed. 31
This means, unfortunately, that the power to shame is likely to be
concentrated in the hands of those whose interpretations are socially authoritative. It is no accident that Adrian Piper finds herself shamed by a
senior faculty member. His seniority, prominence within the profession,
maleness, and whiteness work together to authorize his moral interpretation of her. In other cases, interpretations are socially authoritative because of their sheer conventionality; they express what generally goes
without saying in a particular moral practice. Among doctors, for example, it goes without saying that physicians should never deliberately
harm patients. Moral criticism of doctors who advocate active euthanasia thus has significant shaming power. 32 By contrast, those who lack
social status or who voice controversial or idiosyncratic moral criticisms
often lack the power to shame. This means that the power to shame will
typically be differentially distributed, tracking social status and what a
group finds intuitively obvious. 33
30. The intelligibility of these moral interpretations is not a result of abstract understanding but of the fact that one does or has identified oneself with the social world from
which those views emerge.
31. Rob Cummins suggested this distinction between You have nothing to be ashamed
of and You ought not to feel ashamed.
32. Thanks to Gerald Dworkin for this example.
33. Margaret Walker rightly points out that the moral intuitions appealed to by philosophers are just thatmoral intuitions of the social group philosophers. Other
practices of morality might find different claims intuitively obvious; Authority and
Transparency in Walker, Moral Understandings.

69

MOR A L A IMS

This is not good news for members of subordinate social groups.

For example, in sexist societies, the power to shame will be disproportionately concentrated among men; and since ideologies about womens
moral deficiencies and the unreasonableness of protest typically underwrite sexist systems, vulnerability to being shamed will be disproportionately concentrated among women. Given this, one might naturally object
that any apology for this often emotionally debilitating and demoralizing emotion is misplaced. Weparticularly the we who are socially
subordinatemight be better off to train ourselves not to feel shame.
This is not the simple objection that shame is an unpleasant feeling
so who needs it? Guilt, too, is an unpleasant feeling, yet it seems central to the full-bodied appreciation of moral error, to holding ourselves
responsible, 34 and to differentiating moral agents from psychopathological individuals. 35 The objection here is that shame does not clearly serve
any important moral function. Moreover, the burden of shame seems
unfairly distributed in inegalitarian societies, serving only to further
burden those who are already unfairly burdened. It is thus unclear what
apology could be made for moral shame. 36
Let me begin with the question of moral shames social function,
and then turn to the concern that the power to shame and vulnerability to shame track social stratifications. Morality is, in part, a critical,
normative enterprise conducted by individuals who use their own best
judgment to arrive at moral standards and practical conclusions, who
seek the rationally best justifications for their judgments, and who critically assess the standards and practical conclusions of both particular
others and social practices of morality. Shame, as I have characterized it,
does not serve this dimension of the moral enterprise. Moral criticisms
that we judge to be rationally indefensible may, I have argued, provoke
shame. Shame thus does not second the critical normative judgments
34. Strawson argues that guilt, indignation, and resentment are constitutive of what it
means to hold ourselves and others morally accountable; P. F. Strawson, Freedom and
Resentment, in Freedom and Resentment and Other Essays (London: Methuen, 1974).
35. Jeffrie G. Murphy, Moral Death: A Kantian Essay on Psychopathy, in Ethics and Personality: Essays in Moral Psychology, ed. John Deigh (Chicago: University of Chicago
Press, 1992), 207222.
36. I owe this criticism to Nancy Potter.

70

A N A POLOGY FOR MOR A L SH A M E

that we reach as autonomous, reflective individuals. The enterprise of

morality, however, is not just this reflective, normative business of exercising ones own critical, autonomous judgment. Morality is also fundamentally a social enterprise. 37 While hypothetical moral worlds of
ideally rational agents are heuristically useful in evaluating the justifiability of moral principles and norms, morality is only practiced in real
social worlds. Even if particular social practices of morality seem flawed
from the individuals critical, normative perspective, the social practice
of morality is the only moral game in town. It is only in real social worlds
that I have a moral identity. Who I am, morally, is who I am interpretable and identifiable by others as being. That I fancy myself (even with
what I take to be the best reasons) to be one kind of person rather than
another does not give me an identity as that kind of person. Instead, the
set of ones possible moral identities is delimited by the available moral
interpretations within an ongoing moral practice.
In insisting on the moral importance of remaining vulnerable to
being shamed before others with whom we disagree, I have meant to
second a basic feminist claimnamely that our selves are, in part, socially constructed and we need to take seriously that fact. So long as we
participate in any moral practices at all, we will have some inescapable
moral identities. They are inescapable, first, because ones own selfconception does not decisively determine who one is. For any number of
others within the moral practice of higher education, Piper was underhanded in refusing to pass as white. (Of course, in other practicesfor
example, the moral practice of her familyshe was a person of integrity
for refusing to pass as white.) Second, the identities that we have within
particular moral practices are inescapable because we typically do not
choose moral practices. What we choose are social practices of higher
education, or medicine, or family life; we then find ourselves located, for
better or worse, in particular ongoing moral practices.
In sum, shame is not the emotion of a critical, normatively reflective, autonomous agent. Shame is the emotion of the practitioner of
morality. To attempt to make oneself invulnerable to all shaming criticisms except those that mirror ones own autonomous judgment, or that
37. I defend versions of this point in chapters 1, 3, and 5, this volume.

71

MOR A L A IMS

invoke ethical standards one respects, is to refuse to take seriously the

social practice of morality. And that, I have suggested, amounts to refusing to take morality seriously, because the various social practices of
morality are the only moral game in town.
But what are we to do with the fact that dominant social groups
monopolize the power to shame and subordinate social groups are excessively vulnerable to being shamed? As moral philosophers who have
been trained to think of moral agents primarily as critically reflective,
autonomous persons, it is tempting to conclude that the subordinated
would be best served by becoming more thick-skinned and refusing to
give others shaming criticisms practical weight. It is tempting, that is, to
focus on altering the emotional responsiveness of the socially subordinate moral agent. This is a mistake, though. From both moral and political points of view, the social practice of morality needs to be taken more,
rather than less, seriously. From a moral point of view, taking the social
practice of morality seriously is central to taking morality seriously. Thus
it is no error on the part of the subordinated that they feel the practical weight of their fellow participants moral criticisms. From a political
point of view, taking the social practice of morality seriously is central
to the pursuit of social justice. Political discussion and public policy
often repeat the same shaming criticisms that are part of everyday moral
practice. The U.S. military policy of dont ask, dont tell, for example, clearly rested on the idea that same-sex desire is shameful and thus
tolerable only if hidden. Political critiques of welfare and advocacy of
workfare also often repeat widespread shaming criticisms of the poor as
people who lazily live off the welfare system. Because shaming criticisms
that articulate representative viewpoints are not things that people can
just steel themselves against, we need to take very seriously the sexism,
heterosexism, racism, and the like that are embedded in ongoing moral
and political practices. We need to take seriously the deformed identities
that the subordinate inhabit and the practical importance of contesting
defective moral understandings and struggling to achieve their reform. 38
38. Claudia Card defends this claim with respect to the defaming and demeaning identities to which gay men and lesbians are subjected in her Lesbian Choices (New York:
Columbia University Press, 1995), chap. 9.

72

PA R T I I

R E AC H I NG, R E LY I NG
ON, A N D CON T E ST I NG
SOC I A L CONSE NS US
ONMOR A L NOR M S
Chapters 1 and 2 in part I brought into view two different conceptions of
morality and two different fundamental aims that moral agents should have.
First, there is what we might call a critical conception of morality in which
the moral aim is to latch onto and conform to the right set of moral principles
and the right conception of the moral virtues. Second, there are the shared
moral understandings that define a particular practice of morality and that
enable us to make what we are up to legible and intelligible to others and
to have a moral identity for others. In order to bring out both the difference
between these two conceptions of morality and the moral importance of not
only getting it right but also practicing morality with others, I focused in those
essays on cases where the two conceptions of morality are in strong tension.
In part II, I examine places where I think philosophers critical conception of morality cannot be fully action-guiding if it fails to keep in mind that
we are always practicing morality with others. That general theme will be continued in part III. In chapter 3, I argue that an adequate critical conception
of what civility is and of where the bounds of civility are set must be anchored
in the social functions of civil behavior and civility norms. In particular, civil

MOR A L A IMS

behaviors enable us to display and thus communicate important moral attitudes, and normative prescriptions of civility serve the primary purpose of
enabling dialogue to stay afloat long enough to produce shared moral understandings. These functions are possible only because shared moral norms
enable us to interpret others behavior as expressing respect, tolerance, and
considerateness, and because social consensus rather than individual judgment sets the bounds of civility.
In chapter 4, I take up an ordinary moral phenomenon: elective but
strongly morally expected acts of common moral decency. I argue that moral
theories of obligation and supererogation cannot explain the odd features of
common decenciesthey are not obligatory, but at the same time they are
not merely commendable elective acts. As in the case of civility, I think we get
a better account of what is going on in common decencies by keeping in mind
that moral agents act within a social world. Particular moral practices give
substance to the idea that minimally well-formed moral agents will offer at
least some moral gifts to others. Social norms that institutionalize particular moral gift-giving practices render obvious, unambiguous, and motivationally nontaxing some supererogatory acts, converting them to acts that
can be expected of any minimally well-formed moral agent.
My interest in chapters 3 and 4 was in seeing how social norms are centrally involved in the very essence of some moral phenomena: civility and
common decency. I shift gears in chapter 5 to thinking about a distinctive
virtue of those who adopt a critical moral perspective and aim to get it right:
integrity. I argue that we get a more adequate account of this virtue if we do
not think of integrity as primarily a personal virtue but, rather, as a social
virtue. For anyone who aims to both get it right and to participate in the
progress of social moral knowledge, being able to stand before others for ones
own best judgment will be an important social virtue.

74

C ha pt e r 3

The Virtue of Civility

The decline of civility has increasingly become the subject of lament,
both in popular media and in daily conversation. Civility forestalls the
potential unpleasantness of a life with other people. Without it, daily
social exchanges can turn nasty and sometimes hazardous. Civility thus
seems to be a basic virtue of social life. Moral philosophers, however, do
not typically mention civility in their catalogues or examples of virtue.
In what follows, I suggest that civility is a particularly interesting virtue
for moral philosophers, because giving an adequate account of the virtue
of civility requires us to rethink the relationship between moral virtue
and compliance with social norms.

I. THR EE STR IK ES AGA INST CI V ILIT Y

At least three quite different reasons might be offered for why philosophers so often do not count civility among the moral virtues nor count it
among the philosophically interesting moral virtues.
First, more so than other virtues, civility has intimate associations
with etiquette or good manners. If one takes the elaborate Victorian
fork rules as a paradigm for rules of etiquette, the primary function
of etiquette rules would seem to be neither maximizing utility nor
respecting persons as ends. Instead, proper etiquette distinguishes
the civilized from the barbaric, the upper from the lower classes, and
members of polite society from the rabble. Insofar as being civil is
identified with complying with class-distinguishing etiquette rules,
civility appears not to be a moral virtue but, rather, a badge of class
distinction.

75

MOR A L A IMS

However, even though civility has been a prime subject for etiquette
experts (one of the Miss Manners books is devoted to what she considers
lapses in civility), it is not obviously true that good manners are either
primarily or exclusively class demarcators. Good manners include the
distinctly moral: considering others feelings, expressing gratitude, engaging in tolerant restraint, and respecting others personal privacy. Nor
is it obviously true that civility is limited to good manners. One prominent early meaning of civility was fitness for a civil, post-feudal society.
Defining marks of that fitness included obeying authoritative law, refraining from violence, and having the literacy and education necessary
for public service.1 Contemporary political philosophers similarly take
civility to be a mark of the good citizen.
Even so, there is a second reason for thinking that civility is at best
a minor virtue, or perhaps not a moral virtue at all. More so than other
virtues, civility has intimate associations with following socially established rules, whether those be rules of etiquette or civil law. For example,
the civil debater complies with the written rules of debate; civil neighbors comply with local norms for neighborly behavior; and civil drivers
comply with conventional expectations about courteous driving. If civility is a virtue, it appears to be more like law-abidingness than justice.
Like being a law-abiding citizen, being civil appears to require conforming to whatever the social rules are. Unlike justice, it does not require
adopting a socially critical moral point of view. When one adopts a socially critical moral point of view, one does not try to determine how
people ought to be treated by investigating how they are treated, or how
social conventions recommend they should be treated, or any other
social fact of this kind. Instead, adopting a socially critical moral point
of view means adopting a standpoint that enables one to evaluate the
moral merit of established social norms and to recommend the moral

1. Marvin Becker argues that this early concept of civility originated with the demise of
feudal social organization and depended on a new concept of the individual as someone
not primarily defined by his loyalty to local lordships. It also depended on the emergence
of a conception of the public good distinct from that of local societies and on a shift away
from the pursuit of glory to more peaceable practices; Marvin Becker, Civility and Society
in Western Europe, 13001600 (Bloomington: Indiana University Press, 1988).

76

THE V IRTUE OF CI V ILIT Y

principles that, ideally, would be embodied in our social norms. Adopting a socially critical moral point of view means being prepared to violate existing social conventions. Kantianism and classical utilitarianism
are socially critical moral points of view. Although many socially established rules might also be recommended from a utilitarian or Kantian
point of view, they are not guaranteed to be. Thus, if civility is a matter
of complying with socially established rules, what it is civil to do and
what a critical moral point of view recommends that we do may not be
equivalent. Because of this, even if like Kant one finds merit in obedience even to unjust laws, neither civil obedience nor civility is likely to
appear among the top-ranked moral virtues. Moreover, they may not
seem like moral virtues at all, as what one must do in order to be civil or
civilly obedient may conflict with what a socially critical moral point of
view would prescribe.
Someone might object, however, that absent further argument, it
is not obviously true that genuine civilityas opposed to the social interpretation of what counts as civilitycannot be detached from social
rules. Even if, as a society, we tend to codify civility in etiquette manuals and civil law, perhaps genuine civility is not a matter of mindlessly
complying with those codifications. Instead, it might be thought that
genuine civility requires adopting a critical moral point of view and attempting to determine what really counts as kindness, respect for privacy, tolerance, reasonable concern for others feelings, and so on.
Although detaching civility from mere conformity to established
social rules moves civility more clearly into the realm of moral virtue,
one might still think that there is a third and quite different reason for
not ranking civility among the philosophically interesting moral virtues. Because the scope of application of the terms civil and uncivil
is so huge, it will be very tempting to reduce civility either to something
like Kantian respect or to a set of virtues in order to explain the broad
scope of these terms. To get some sense of just how large the scope of
application of uncivil is, consider these lapses of civility mentioned
by Miss Manners.2 In addition to shoving, shouting, giving the finger,
2. Judith Martin, Miss Manners Rescues Civilization from Sexual Harassment, Frivolous Lawsuits, Dissing and Other Lapses in Civility (New York: Crown, 1996).

77

MOR A L A IMS

making insulting remarks, and not waiting ones turn in line, there
are the incivilities of nosiness, bossiness, snobbishness, breaking appointments, overstaying visits, failing to offer thanks or apologies or
responses to invitations, not reciprocating hospitality, hogging the
road, littering, proselytizing, and offering unsolicited advice. Colonial
American manuals on manners remind us of earthier incivilities, such
as returning half-eaten food to communal dishes, scratching or revealing private parts, spraying spit while talking, and farting. 3 And political
philosophers include in their various lists of incivilities coercion, intimidation, harassment, violence, unrestrained pursuit of self-interest,
the arbitrary exercise of power, disrespect for others rights and dignity,
stating deep but unshared moral convictions, inattentively listening to
anothers argument, intolerance, indifference to offense, voting on the
basis of private preferences, expressing contempt for others life plans,
engaging in vigilante justice, rioting, and not obeying the law.4
In short, incivility differs from other vice labels in tending to be
applicable to virtually any example of moral or mannerly misbehavior.
Thus civility does not seem to name a distinct virtue. Instead, civility seems either to pick out a fundamental attitude that lies at the core
of all the more particular virtues (much the way Kants concept of respect does) or to designate a collection of virtues such as tolerance, considerateness, law abidance, and the like. In either case, the philosophical
analysis of civility will be parasitic on the analysis of the more basic
virtue(s) to which civility is reducible.
Each of these reasons for discounting civility as a moral virtue or
demoting it to a derivative moral virtue depends on a different understanding of what civility is: (1) a set of class-demarcating behaviors;
3. George Washington, George Washingtons Rules of Civility and Decent Behaviour in Company and Conversation, ed. Charles Moore (Boston: Houghton Mifflin, 1926).
4. Michael Walzer, Civility and Civic Virtue, Social Research 41 (1974): 593611; Burton
Zwiebach, Civility and Disobedience (New York: Cambridge University Press, 1975);
Clifford Orwin, Civility, American Scholar 60 (1991): 553564; Mark Kingwell, A
Civil Tongue: Justice, Dialogue and the Politics of Pluralism (University Park: Pennsylvania State University Press, 1995); John Rawls, A Theory of Justice (Cambridge, MA:
Harvard University Press, 1971) and Political Liberalism (New York: Columbia University Press, 1993); Richard C. Sinopoli, Thick-Skinned Liberalism: Redefining Civility,
American Political Science Review 89 (1995): 612620.

78

THE V IRTUE OF CI V ILIT Y

(2) a morally uncritical conformity to socially established rules of respect, tolerance, and the like; or (3) an equivalent to one or more items
on the familiar philosophical list of moral virtues. Anyone who wants
to argue, as I in fact do, that civility is a distinct and important moral
virtue will need to do at least two things. The first and primary task is
to provide an account of civility that does not reduce civility to some
other virtue(s). Why isnt civility just another name for being respectful,
tolerant, and considerate of ones fellows? Second, given civilitys close
association with following socially established rules, it will be necessary
to explain why civility should be considered a moral virtue. It would also
be a bonus if the account captured some of the basic intuitions underlying the three understandings of civility just mentioned. Those intuitions
are that civility is intimately connected to socially established rules, including rules of etiquette (view 2); that it is also intimately connected to
other moral virtues like tolerance and respect (view 3); and that there is
special reason to worry about class bias in our judgments about who is
and who is not civil (view 1).
In what follows, I will be adopting a variant of view 2that civility
involves conformity to socially established rules of respect, tolerance,
and considerateness. I do not, however, take the social conformism
built into civility to be a reason for discounting civilitys moral importance. On the contrary, I will argue that this conformity is critical to
civilitys moral function. The function of civility, I will suggest, is to
communicate basic moral attitudes of respect, tolerance, and considerateness. 5 We can successfully communicate these basic moral attitudes
to others only by following socially conventional rules for the expression of respect, tolerance, and considerateness. Thus, I take civility to
be tied to social rules in a way that, for example, honesty, justice, kindness, and respect are not. Although civilitys tie to social rules sometimes occasions a conflict between what it would be uncivil to do and
what, from a critical moral point of view, is morally correct, I will argue
that there is, nevertheless, good reason to count civility among the
moral virtues.
5. For a similar view of the expressive function of manners, see Sarah Buss, Appearing
Respectful: The Moral Significance of Manners, Ethics 109 (1999): 795826.

79

MOR A L A IMS

II. POLITICA L A ND POLITE CI V ILITIES

By taking a look at what political philosophers and etiquette experts
have had to say about civility, we can get a more detailed picture of the
terrain of civility. More important, we will be able to see how both political philosophers and etiquette experts have tended to favor the third
understanding of civility mentioned abovethat being civil is nothing
but a matter of being respectful, considerate, and tolerant. Thus civility
does not name a distinct virtue.
While early political conceptions of civility linked civility to the
formation and stability of any civil society, for contemporary political
theorists, civility is a virtue specific to liberal democratic societies. Civility fits citizens for life in a pluralistic society and is closely connected
to tolerance. The civil citizen exercises tolerance in the face of deep disagreement about the good. She respects the rights of others; refrains
from violence, intimidation, harassment and coercion; does not show
contempt for others life plans; and has a healthy respect for others privacy. As Clifford Orwin puts it, civility is a bond uniting honest men
busy minding their own affairs who are neighborly but who recognize
that good fences do make good neighbors.6
Tolerant self-restraint, however, is only part of what fits citizens for
life in a liberal democracy. In addition, citizens must seek accommodation and compromise through reasoned dialogue.7 As the virtue that fits
citizens for life in a participatory democracy, civility thus gets equated
with respectful dialogue: keeping a civil tongue.
Political theorists differ on what speech constraints civility requires. For Rawls, because the civil citizen respects others capacity to

6. Orwin, Civility, 560.

7. Although some authors continue to include law-abidingness within the scope of civility,
the rationale for doing so is not clear. Some law-abidingness is, of course, directly connected to liberal tolerance, such as respecting rights to speech, association, and privacy.
But paying taxes, obeying the speed limit, and not evading the draft are not similarly
connected to the distinctive requirements of liberal democracies. Instead, the temptation to equate civility with law-abidingness generally seems to reflect the continuing
cultural influence of an earlier conception of civility. See Walzer, Civility and Civic
Virtue, and Zwiebach, Civility and Disobedience.

80

THE V IRTUE OF CI V ILIT Y

be reasonable in setting the terms of fair social cooperation, civility requires that people be willing to explain the grounds of their actions,
especially when the claims of others are overruled.8 On matters of basic
justice, civility additionally requires that individuals refrain from appealing to comprehensive doctrines and instead appeal only to basic
principles of justice that all can be expected to share.
Amy Gutmann and Dennis Thompson (whose work on mutual respect influenced Rawlss remarks about civility in Political Liberalism)
reject such severe speech constraints. Instead, the civil citizen simply
seeks for points of moral agreement, offers rationales that minimize
the risk of her position being rejected, and refrains from presenting her
views as unalterable convictions.9 Mark Kingwell takes a similar approach. Civility, in his view, requires a willingness not to say all the
true, or morally excellent things one could say, especially when expressing ones deeper moral convictions is likely to be offensive, hurtful,
or a conversation stopper.10
All agree that civility is, importantly, a matter of restraining speech.
They also agree that civility has an active side as well. For Rawls, the
civil citizen also exhibits an active willingness to listen to others, to try
to see things from the point of view of their conceptions of the good;
she is neither contemptuous of nor indifferent to others life plans and
makes fair-minded accommodations to their views.11 For Gutmann and
Thompson, the civil citizen magnanimously acknowledges that his opponents view is a genuine moral position about which reasonable people
may disagree. For Kingwell, the civil citizen listens attentively, actively
coaxing out the interests of others through sensitivity and tact.12
As a polite (rather than political) virtue, civility has been understood as the mark of the competent participant in the social settings of
8. Rawls, A Theory of Justice, 179.
9. Amy Gutmann and Dennis Thompson, Moral Conflict and Political Consensus, in
Liberalism and the Good, ed. R. Bruce Douglass, Gerald M. Mara, and Henry S. Richardson (New York: Routledge, 1990). Their analysis is of mutual respect, not civility.
Following Rawls, I interpret them as in fact describing civility.
10. Kingwell, A Civil Tongue, 44.
11. Rawls, A Theory of Justice, 337338; Political Liberalism, 217218.
12. Kingwell, A Civil Tongue, 211.

81

MOR A L A IMS

everyday lifeat work and at parties, in restaurants and private dining

rooms, in churches and synagogues, on public transportation and on
urban streets, in hospital rooms and doctors offices, at family gatherings, weddings and funerals, in courtrooms, board rooms, and on the
floor of Congress.
Like political civility, polite civility has varied historically. Young
George Washingtons Rules of Civility and Decent Behaviour in Company
and Conversation focuses in roughly equal measure on basic bodily control and hygiene, and on considerate and respectful interaction befitting
ones own and others social stations.13 Like the early political civility
that enables the citizen to escape from barbarism to civil society, Washingtons polite civility enables the social participant to avoid barbaric
and potentially disgusting bodily displays. The civil person refrains
from humming, finger drumming, nail biting, bedewing others with
spittle, eye rolling, lolling out the tongue, gaping, killing fleas and lice in
others sight, wearing foul clothes, and falling asleep while others speak.
The civil person also shows both some sign of respect to everyone in his
or her company and special respect for persons of quality by, for instance, careful attention to the order in which persons speak, walk, and
are seated.
Such manuals that focused on deferential displays of respect for
rank were not ultimately suitable to American egalitarian ideals. The
nineteenth century saw a flood of etiquette manuals, one of whose aims
was to adapt the more rank-conscious French and British conception of
civility to an egalitarian democracy.14 In the twentieth century, the Miss
Manners etiquette manuals provide a particularly clear example of the
attempt to work out a conception of polite civility that is tightly connected to the ideals of toleration, egalitarianism, reasoned dialogue, and
positive respect for others life plans.
Like many political theorists, she regards a sizable portion of the
norms of civility as supporting toleration of differences by creating
13. George Washingtons Rules of Civility and Decent Behaviour in Company and Conversation was adapted via a 1640 English manual from a 1595 French Jesuit manual.
14. John F. Kasson, Rudeness and Civility: Manners in Nineteenth Century Urban America
(New York: Hill and Wang, 1990).

82

THE V IRTUE OF CI V ILIT Y

social distance. Civility requires respect for others privacy, particularly

by not intrusively probing into how others are conducting their lives and
by not expressing ones assessment of those lives or advice on how to improve them. Thus, incivilities include nosiness, attempting to improve
others by offering unsolicited advice, proselytizing, self-righteously insisting that others adhere to particular moral standards, and correcting
others manners.
Like political theorists, she too recommends speech constraints for
civil dialogue. For her this means that in public fora where discussion
of controversial issues is appropriate, civility simply precludes insults,
invectives, displays of contempt, or attempts to humiliate, embarrass,
demonize, or demean ones opponent; it also means waiting ones turn
to speak and refraining from airing personal prejudices. In more social
and private fora such as at the dinner table, civility, in her view, places an
almost exceptionless bar on raising controversial and potentially offensive moral, political, and religious issues that would disrupt the social
events that others care about.15
Polite civility also requires considerately respecting others life plans
by, for instance, waiting ones turn in line, keeping appointments, not
treating others time as though it were less important than ones own,
not hogging the road, replying to invitations, not overstaying visits, and
graciously accepting gifts rather than asking if they might be exchanged.
In little ways, all of these actions acknowledge the value of others lives.
As descriptions of what in fact we expect of civil people, I take both
the political and polite narratives of civility to be relatively uncontroversial, but the account of civility implicit in both is problematic. The lists
of political and polite civil behavior do not appear to depend on a prior
understanding of civility as a distinct virtue. Instead, they appear to be
entirely derived from a prior understanding of tolerance, considerateness, mutual respect, and a sense of justice. The question, What should
a civil person do? appears to be interchangeable with the questions
How should mutually respectful citizens treat each other? or How
should considerate social participants treat each other? or What does
being tolerant of others differences involve? But if civility is just the
15. Martin, Miss Manners Rescues Civilization, 366ff.

83

MOR A L A IMS

exercise of tolerance, respect, and considerateness toward fellow social

participants or fellow citizens, then civility does not name a distinct
virtue, and there is no reason for moral philosophers to mention civility
in a catalogue of moral virtues.
This may be the right conclusion. But it is sufficiently counterintuitive to make it worthwhile asking whether there is some way of analyzing
civility that preserves much of the above description of civil behaviors
but also establishes a distinction between civility and other virtues.16

III. CI V ILIT Y AS A DISTINCT V IRTUE

What might separate civility from other kinds of moral behavior? Let
me suggest this: civility always involves a display of respect, tolerance,
or considerateness. By displays of respect, tolerance, and considerateness, I have in mind acts that the target of civility might reasonably
interpret as making clear that the civil actor recognizes some morally
considerable fact about her that makes her worth treating with respect,
considerateness, and tolerance. That morally considerable fact might be
that she is a person, or that she has feelings, or that she has views, tastes,
or interests of her own, or that she has earned an authority position, or
that she is my neighbor. The civil person regards such morally considerable facts as placing restrictions not just on how she treats others but also
on the messages about their worth that she conveys to them.17 Those
messages sometimes get sent through positive tokens of respect and
consideratenesslistening carefully, saying thank you, replying to
an invitation. On other occasions, conveying ones willingness to consider others feelings or the fact that they have tastes and views of their
own depends primarily on acts of concealment. In social life, there are
16. The account of civility I offer may not distinguish civility from all the virtues that civility threatens to collapse into, particularly not from law-abidingness and civil obedience. My aim is to distinguish civility from respect, tolerance, and considerateness in
particular.
17. Jean Hampton develops the idea that moralities and immoralities convey messages
about worth in Forgiveness, Resentment and Hatred, in Jeffrie G. Murphy and Jean
Hampton, Forgiveness and Mercy (Cambridge: Cambridge University Press, 1988).

84

THE V IRTUE OF CI V ILIT Y

unending opportunities to find other people boring, disagreeable, repulsive, stupid, sleazy, inept, bigoted, lousy at selecting gifts, bad cooks,
infuriatingly slow drivers, disappointing dates, bad philosophers, and so
on. The civil person typically conceals these unflattering appraisals, because conveying them may easily suggest that one does not take others
feelings, or the fact that they may have different standards, to be worth
taking into consideration or tolerating.18
In short, what makes being civil different from being respectful,
considerate, or tolerant is that civility always involves a display of respect, tolerance, or considerateness. Thus civility is an essentially communicative form of moral conduct. In addition, because communicating
our moral attitudes is central to civility, being genuinely civilunlike,
say, being genuinely considerate or genuinely tolerantrequires that
we follow whatever the socially established norms are for showing
people considerateness, tolerance, or respect. Only because there are
such generally agreed upon, often codified social rules for what counts
as respectful, considerate, and tolerant behavior can we successfully
communicate our moral attitudes toward others. Those rules create a
common language for conveying the attitudes of respect, willingness to
tolerate differences, and consideration. Similarly, incivilities draw on a
common verbal and behavioral language for displaying disrespect, intolerance, or inconsiderateness.
Because civil and uncivil acts are essentially communicative acts,
while simply treating people with respect or tolerance does not always
involve communicating our moral attitudes, civil behavior is not coextensive with respectful, tolerant, and considerate behavior. To see this,
first consider that being civil and treating people with respect, considerateness, or tolerance are sometimes two distinguishable constituents
of what we might call fully, or maximally, respectful, considerate, or
18. Responding to them as though they werent deficient in various ways (and then, perhaps, poking fun at them behind their backs) is hypocritical. But as Thomas Nagel
has observes, it is a form of hypocrisy that we make sure we teach children, that were
thankful that others engage in, and that isnt deceptive since everyone engages in socially conventional practices of polite concealment and everyone knows what might be
going on behind their backs. Thomas Nagel, Concealment and Exposure, Philosophy
& Public Affairs 27 (1998): 330.

85

MOR A L A IMS

tolerant behavior. A fully tolerant person not only permits those with
different views or life plans the same freedoms that she enjoys (for example, the freedom to pursue their interests in public spaces); she also
displays a tolerant attitude when given the opportunity to do so (by, for
example, not audibly complaining about having to share public space
with different others). She both tolerantly treats others and civilly displays her willingness to tolerate others. Because constraining ones
actions in ways that are required by a principle of toleration (or considerateness or respect) and displaying attitudes of tolerance (or considerateness or respect) are not the same thing, it is possible to be uncivil while
nevertheless treating others with some degree of tolerance, respect,
or considerateness.19 Think, for example, of the person who carefully
skirts his neighbors lawn while sarcastically declaring, Dont worry, I
wont step on your precious grass; or the employer who carefully follows affirmative-action guidelines but who tells the new employee,
You know you only got this job because youre black; or the partygoer
who rues his own self-restraint by announcing, I guess I wont tell that
(sexist) joke, since I know you gals dont have a sense of humor. These
individuals are being respectful, considerate, and tolerantalthough
not fully sobecause their actions are constrained so that they do not
damage others property, or deprive them of equal opportunity, or insult
them with demeaning jokes. Thus complaints of trespass, racist hiring,
or sexist joke-telling have no toehold in these examples. One might,
however, think that something is missing for fully respectful, considerate, and tolerant treatmentnamely the civil display of the corresponding moral attitudes.
Second, that civility does name a virtue different from simply treating people with respect or tolerance is also evident if one keeps in mind
that not all cases of treating people respectfully or considerately or tolerantly involve any sort of communicative interaction. For example,
contributing to charities is a way of treating unknown others considerately and respectfully, but that considerateness is not displayed to
the recipients of charity. Charitable donors behave well, but they are
not being civil (or uncivil). Consider also the fact that treating people
19. Buss makes a parallel point in Appearing Respectful, 797.

86

THE V IRTUE OF CI V ILIT Y

disrespectfully is often accomplished by deliberately avoiding communicative interaction with the targets of disrespect. For example, when
people engage in covert trespassing, theft, forgery, tax evasion, bribing
of public officials, and drug trafficking, particular persons, or citizens
generally, are treated with disrespect. But there is no communicative
interaction with the targets of disrespect, and thus there is no opportunity to display any attitudes at all to the targets of disrespect. Thieves
and drug traffickers behave badly, but they arent guilty of incivility. (Of
course, they arent being civil, either.)
Third, even when one is in ongoing communicative interaction with
other people, treating people disrespectfully is not always accompanied
by a display of disrespect. Instead, those who violate principles of respect, considerateness, or tolerance often try to conceal their wrongdoing. Consider conducting a discrete adulterous affair; making racist,
sexist, or other demeaning comments about ones coworker behind her
back; or engaging in discriminatory hiring practices that are carefully
hidden from job candidates. Because the targets of disrespect, inconsiderateness, and intolerance are kept ignorant of how they are being
treated, there is no uncivil display to the target. 20 Adulterers and discriminatory employers behave badly, but they are not guilty of incivility unless they flaunt their wrongdoing before the target of disrespect.
Of course, covert adulterers and discriminatory employers, even if they
arent guilty of incivility, are not to be praised for their civility either,
because presumably they conceal their misbehavior for self-interested
reasons rather than to avoid sending a disrespectful message.
Finally and most important, in morally imperfect social worlds, we
may have to choose between being civilthat is, successfully communicating our attitude of respect or toleranceand behaving in ways that
are genuinely respectful or tolerant. In such cases, it becomes quite clear
that civility cannot be equated with respect, tolerance, or considerateness. To take a familiar example, consider how opening doors for women
has been, and continues to be, a socially conventional way of displaying
20. In a broader sense of display than I am using, demeaning ones coworker behind her
back is a display of disrespectone has made ones attitude public. On my view, for
incivility, it matters to whom one makes this display.

87

MOR A L A IMS

respect for women. In many social environments, any man who plunges
ahead first through a doorway will be interpreted as rudely displaying
a disrespectful attitude. Yet such ladies-first policies, one might think
(as most feminists now do), are not really respectful. They are rooted
in demeaning assumptions about womens weakness and need for male
protection. In a morally more perfect world where womens equality is
built into our social conventions, there would be no ladies-first policies.
That is not our world, however. As a result, men often have to choose
between making a comprehensible civil display of a respectful attitude
and treating women in the way they ought always to be treated, were our
society a gender-egalitarian one.
To take a second example, consider the fact that asking people to
closet a nonheterosexual identity (at work, church, family gatherings,
the military, and the like) generally does not violate our social norms
for tolerance. Thus the military could seriously present its dont ask,
dont tell policy as a tolerant one. In short, pressuring people to stay
closeted is generally not uncivil. 21 Yet such dont-tell policies, one might
think, are not really tolerant. They are rooted in demeaning assumptions
about gay mens and lesbians moral depravity and sexual licentiousness.
In a morally more perfect world, where prejudices are not built into our
social conventions, there would not be dont-tell policies. That is not our
world, however. As a result, nonheterosexuals often have to choose between accepting without ire a civil display of tolerance and protesting
treatment that would not be acceptable were our society a sexually unprejudiced one.
Examples like this show that the decision procedure for answering
How can I treat P with genuine respect, considerateness, or tolerance?
differs from the decision procedure for answering How can I display respect, considerateness, or tolerance to P? If I am concerned with treating others with the respect owed them as moral persons, my interest is in
determining how they ought to be treated, regardless of what treatments
21. There have been considerable changes in social and legal norms concerning the visibility of same-sex sexual orientation since this chapter, begun in 1996 and published
in 2000, was written. Pressuring people to stay closeted would now be uncivil in many
areas of social life.

88

THE V IRTUE OF CI V ILIT Y

are acceptable under existing social norms. Existing social norms may
sanction as natural, normal, and legitimate treatment that is in fact demeaning, unjust, cruel, or intolerant of alternative conceptions of the
good. They may also sanction giving what are in fact unfair privileges to
some and denying to others their rightful due. To decide what genuinely
respectful or considerate or tolerant treatment would amount to, I must
set aside socially established moral understandings and adopt a socially
critical moral point of view, such as a utilitarian or Kantian framework.
From that critical point of view, treating others with moral respect may
sometimes require violating existing social norms. Similarly, if I am interested in determining what I owe others in the way of tolerating their
differences, I cannot appeal to social norms because, as a result of defects in socially shared moral understandings about the status of different groups, social norms may sanction tolerating what is in fact bigotry.
Instead, I must adopt a socially critical moral point of view in order to
sort out genuinely tolerant from intolerant beliefs and behaviors. (I will
return to this point in section IV.)
By contrast, if I am interested in displaying respect, my task is not
to figure out how people ought to be treated but, rather, how I can successfully communicate moral attitudes. Displaying respect is essentially
a communicative action. That communication requires a common
language. Social norms provide that common language because they
embody shared moral understandings. Social norms for what is due
others make possible successfully delivering an insult, a snub, a demeaning gesture. They also make it possible to offer tokens of respect or
considerateness or tolerance. Because some gap between critical moral
conceptions and social conceptions of what counts as respectful treatment is likely, a socially critical moral point of view that enables us to
determine how persons ought to be treated cannot reliably tell us how to
display the moral attitude of respect. In at least some cases, others may
perceive our effort to treat them with respect as either, depending on the
case, insufficiently respectful or exceedingly kind. For example, from a
socially critical moral point of view, one might conclude that treating a
hostess considerately requires that male (and not just female) guests help
clean up. Social norms, however, exempt men from this form of considerateness, thus hostesses are likely to view male help as exceedingly kind.
89

MOR A L A IMS

Only within a hypothetical world, where critical moral understandings

are also socially normative, are our acts of treating others with genuine respect, tolerance, or considerateness guaranteed to be correctly
interpretable by others. In morally imperfect worlds, correctly treating
others and communicating respectful moral attitudes are often two different activities. Consequently, our final judgments about what to do in
such imperfect worlds will often involve weighing two competing moral
considerations: (1) the value of successfully communicating basic moral
attitudes (civility), and (2) the importance of treating people with genuine respect, tolerance, and considerateness.
I want to underscore that these are competing considerations in morally imperfect social worlds. Any moral framework used to determine
what counts as genuine respect, tolerance, and consideratenessand
thus what our social norms ought to bewill surely attach some importance to communicating moral attitudes. Utilitarianism obviously does;
but so does any deontological theory containing a duty of beneficence
or a duty to promote others self-esteem. 22 However, if what we want
out of a moral framework are guidelines that will enable us to engage
in social criticism, that moral framework cannot afford to weight misguided feelings too heavily. Bernard Williams has made this point quite
forcefully about utilitarianism. 23 Utilitarians will not be able to criticize
existing social arrangements if they factor in too heavily the pleasures
taken in existing arrangements and the pains of disrupting them. So,
for example, utilitarianism wont yield a socially critical moral framework if it gives significant weight to offense taken at not being given
what social norms mistakenly specify as ones due. Similarly, Kantians
cannot afford to attach significant weight to communicating respect,
because the acts that successfully communicate respect may be highly
inegalitarian (think, for example, of what blacks have historically had
to do in order to communicate respect to whites in the United States).
In short, any moral framework that is designed to enable us to criticize,
revise, and sometimes reject existing social norms cannot afford to be
22. An anonymous reviewer for Ethics proposed a variant of this point.
23. Bernard Williams, A Critique of Utilitarianism, in his and J. J. C. Smarts Utilitarianism: For and Against (Cambridge: Cambridge University Press, 1973), 104106.

90

THE V IRTUE OF CI V ILIT Y

one that places a lot of value on successfully communicating attitudes

of respect, tolerance, and considerateness. Thus, in a morally imperfect
world, what a socially critical moral framework recommends may well
be at odds with what we feel called upon to do in order to communicate
our moral attitudes toward those we live with.

IV. THE BOU NDS OF CI V ILIT Y

So far, I have suggested that civility is the virtue and incivility the vice
with respect to communicating moral attitudes in contexts governed
by social norms. Civility names a distinctive feature of some actions:
displaying that one takes another to be worth respecting, tolerating,
or considering. Both civilities and incivilities rely on a common social
language. Thus civilities and incivilities are directly specified by social
norms. Codified or tacitly shared rules of fair debate, clean campaigning, neighborliness, hosting, turn taking, considerate driving, personal inquiries, proper dress, voicing criticisms, raising controversial
subjects, and so on set the terms for displaying respect, tolerance, and
considerateness.24
These displays of respect are not morally negligible. First, civility
signals others willingness to have us as co-participants in practices
ranging from political dialogues, to campus communities, to funerals, to sharing public highways. Second, for those who are not already
coerced into sharing social practices with us, civility may be a precondition of their willingness to enter and continue in cooperative ventures with us. Third, civility supports self-esteem by offering token
reminders that we are regarded as worth respecting, tolerating, and
considering. Finally, civility, particularly toward members of socially
24. One consequence of the view that there is a social language for conveying respect and
disrespect is that incivility is not a function of persons intentions. Because actions have
social meanings, what a person does may display disrespect even if he does not intend
to do so. Making baldly sexist comments to a woman displays disrespect and is uncivil
regardless of what the speaker means to be doing. The speaker may evade being held responsible for this incivility if he can come up with a passable excuse for being ignorant
of the social meaning of his speech (generally a hard thing to do).

91

MOR A L A IMS

disesteemed groups, protects individuals against the emotional exhaustion of having to cope with others displays of hatred, aversion,
and disapproval.
Granted that civil displays have moral benefits, one might still
object that the virtue of civility cannot consist solely in following established social norms for displaying respect, tolerance, and considerateness. Moral virtues should produce conduct that is correct by socially
critical moral standards. Compliance with established social norms can
hardly be said to produce such conductat least not reliably. After all,
in markedly inegalitarian societies, established social norms are likely
to require that we pay out to dominant groups larger measures of respect, tolerance and considerateness. They are also likely to require
subordinate groups to put up with more meager allotments. Thus social
norms of civility may fail to condemn the contemptuous treatment of
socially disesteemed groups, because they interpret such contempt as
civilly displaying the appropriate measure of respect. Until just recently,
for example, much of what we now call sexual harassment was socially
interpreted as innocent flirting, or as a response invited by some womens impropriety, and thus not a lapse of civility. Far from condemning
such moral misbehavior, social norms may instead condemn the disesteemeds protests as uncivil.
If civility doesnt look much like a moral virtue on my account,
wouldnt it be better to go back to the idea, mentioned at the very beginning, that genuine civility is not about displays? It is about treating
people with real respect, considerateness, and tolerance, and thus we
cant be genuinely civil without adopting a socially critical moral point
of view. (A socially critical moral point of view, recall, is just a moral
framework, like utilitarianism or Kantianism, that we can use to evaluate and revise social norms so that they reflect more nearly correct understandings of what we morally owe to others.) On this latter approach,
being genuinely civil could never mean complying with what is in fact
unjust. So, for example, genuine civility could never, under any social
circumstance, require racial deference, or closeting ones sexual orientation, or putting up with sexual harassment. Instead, a concern for genuine civility might lead us to critically reassess social norms of civility.
Although civility would then turn out not to be a virtue distinct from
92

THE V IRTUE OF CI V ILIT Y

respect, tolerance, and considerateness, we would at least have more

reason to lament its decline.
However attractive this alternative, socially critical moral analysis
might be, I think it will not ultimately yield a plausible account of civility. In this last section, I will suggest that equating genuine civility
with what we (philosophers who appeal to some socially critical moral
framework) conclude really does embody respect, considerateness, and
toleranceas opposed to what a social group, perhaps mistakenly, takes
to embody these attitudeswill result in setting the bounds of civility
in a way that undermines a principal point of the virtue. By bounds of
civility I mean the point where speech and action are sufficiently disrespectful, inconsiderate, and intolerant not to warrant a civil response.
Since civility is not the virtue of being nice no matter what, civility
norms need to tell us which intentional misbehaviors on others part we
are required to respond to civilly and which we arent. To extend Orwins analogy, civility norms need to tell us when bad neighbors are bad
enough to deserve eviction, not a bigger fence. Those who write about civility generally understand their task to include a specification of when
the bounds of civility have been reached. A sizable portion of the letters
written to Miss Manners, for example, request clarification on where the
bounds of civility are set. Those letters narrate atrocious misbehavior
and ask, in effect, Cant I be uncivil to that?
In her replies, Miss Manners takes a socially critical moral approach
to setting the bounds of civility. Gutmann and Thompson do, too, in
their discussion of mutual respect (which I take to be, in effect, a discussion of civility). Indeed, Gutmann and Thompson give us a particularly clear example of what someone who thinks that genuine civility
expresses a socially critical moral point of view would say about the
bounds of civility.
Gutmann and Thompson argue that in political dialogues mutual
respect (civility) is owed only to those expressing genuine moral positions. Whenever a view can be shown not to be a genuine moral position, a respectful (civil) acknowledgment that reasonable people may
disagree is not in order. Similarly, Miss Manners sanctions such apparent lapses of civility as subjecting to social scorn, cutting dead, exclaiming How dare you?, walking out on an offensive lecture, and giving
93

MOR A L A IMS

a withering look in response to unacceptable conduct. Included on her

list of the intolerable are expressions of bigotry, sexual harassment, child
molestation, subjecting others to cigarette and cigar smoke, and in general, patent illegality and immorality.
Both appear to set the bounds of civility by appealing to a moral
framework that could also be used to reject some commonly shared
social beliefs about what is tolerable behavior or a genuine moral position, and thus deserving of a civil response. Consider first Gutmann
and Thompsons appeal to the notion of a genuine moral position. To
determine which beliefs count as genuine moral positions and which do
not, we will have to appeal to some critical moral framework. Doing so
will in turn allow us to set the bounds of civility by specifying which
beliefs are really (or really not) owed a respectful (civil) response, regardless of what the existing social understandings may suggest is owed
a civil response. For example, Gutmann and Thompson claim that a defense of racial discrimination is not owed a civil response because it is
not a genuine moral position. Since this is a critical normative claim, it
would appear to hold good in any historical period. Thus participants
in nineteenth-century debates over slavery were presumably no more
required to respond respectfully (civilly) to defenders of slavery than
anyone today would be should she encounter an advocate of slavery.
Miss Manners appears also to appeal to an unstated moral viewpoint
in setting the bounds of civility. She claims, for example, that sexual harassment has always been intolerably uncivil and thus was never owed a
civil response. 25 Setting the bounds of civility in this way has the merit
of assuring us that being civil never requires dignifying bigotry, racism,
sexism, and homophobia, and that protesting wrongful treatment will
not be uncivil.
How could this morally attractive view undermine a principal point
of the virtue of civility? It will be helpful here to recall the political

25. Martin, Miss Manners Rescues Civilization, 164. It is, however, not always clear whether
she intends to appeal to a critical moral view or to social understandings to set the
bounds of civility. Her view that reacting against bigotry is not rude so long as what
counts as bigotry has already been announced and socially accepted (358) suggests
that the bounds of civility are set by appeal to social understandings.

94

THE V IRTUE OF CI V ILIT Y

conception of civility. On that conception, civility is what fits us for life

in a pluralistic society where nonlike-minded people will have to enter
into political dialogue in order to reach compromise agreements. Controversial issues will be the primary subject of dialogue, and the controversy itself is likely to originate in the fact that the parties to dialogue
operate from different moral frameworks (or apply the same framework
to different interpretations of the facts). A principal point of having
norms of civility is to regulate discussion of controversial subjects so
that dialogue among those who disagree will continue rather than break
down. Civility norms regulate discussion, first, by imposing speech
constraints that prohibit the parties from expressing themselves in ways
that might give a reasonable interlocutor cause to back out of the conversation. More important, civility norms regulate discussion by equally
requiring all parties to respond respectfully to the same set of positions
that are on the table for discussion, regardless of what they may privately
think about those positions. In other words, civility norms bar dialogue
participants from exercising their own, individual judgment about what
views are utterly contemptible, intolerable, and not worth a respectful
hearing. There may, of course, be positions that are off the table; but here
again they will get taken off the table not because you or I happen to
think they arent owed a civil response. If they get taken off the table, it
will be in a way that equally exempts everyone from civilly responding
to the same set of positions.
To imagine that civility requires we display tolerance only to what
we, as individual reasoners, have concluded are tolerable opinions or
genuine moral positions is to imagine a norm of civility that cannot regulate disputes. Because civility has its point and place precisely with respect to views that are under dispute in a society, civility norms must
require civil responses to some views regardless of what individual reasoners think about them. Thus individual judgment cannot determine
the views whose expression or enactment is not owed a civil response.
This is true no matter how careful, objective, and well informed individual moral reasoners attempt to be. It is simply a fact about our collective rational life together that we often differ in our judgments. Our
disagreements concern not only better and worse behavior but also
the morally intolerable that deserves no civil response. In the midst of
95

MOR A L A IMS

disagreement over what is morally intolerable (and thus does not deserve a civil response), it is a display of intolerance to insist on using ones
own judgment to decide what deserves a civil response.
For the same reason, which views are and are not owed a civil response cannot be a matter for moral philosophers to decide by appeal to
some socially critical moral framework. A socially critical moral view is,
after all, a particular normative view and thus likely to be held by some
people and not by others. It is Miss Mannerss critical moral view that
sexual harassment and subjecting others to cigarette smoke are intolerable and not owed a civil response. It is Gutmann and Thompsons critical
moral view that a defense of racial inequality does not count as a genuine moral position. The objective intolerability of sexual harassment and
racial supremacy seem obvious to us now. Neither was obvious in earlier
historical periods, and there is presently substantial disagreement about
what in fact counts as intolerably subjecting others to smoke. To suggest
that enlightened individuals in earlier historical periods who realized
the wrongness of what we now call sexual harassment and racial bigotry
were also correct to ignore the fact of deep social disagreement over
these behaviors (or the fact of widespread agreement on their moral innocuousness), and to judge for themselves what is owed a civil response,
is to leave civility norms up to individual judgment in one of two ways:
(1) either everyone is entitled to use his preferred moral framework and
decide for himself where the bounds of civility are setproducing civility anarchy, since we will likely not all agree on what is and is not
owed a civil response;26 or (2) one particular moral framework is simply
declared the correct one and is used to set the bounds of civility. Miss
Manners avoids civility anarchy by setting herself up as the supreme
legislator of civility. For example, she assumes the prerogative of deciding what the smoke rules are, and thus what kinds of smoking do not
deserve a civil response. Philosophers who want to set the bounds of

26. A good example of this civility anarchy is the variety of nonsmokers judgments about
which behaviors on the part of smokers are intolerable. Smoking near fellow passengers, in offices, in restaurants, in bars, on public streets, and in areas designated for
smokers might, depending on the person, be regarded as exceeding the bounds of the
tolerable.

96

THE V IRTUE OF CI V ILIT Y

civility from a socially critical moral point of view can similarly avoid
civility anarchy only by setting themselves up as the supreme legislators
of what counts as the true critical morality. Left to individual judgment
in either one of these ways, standards of civility cease to regulate dispute.
They are instead tied to the very moral frameworks that are under dispute. In short, if we appeal to any particular moral framework to determine the bounds of civility, we must treat as settled the very questions
that civil dialogue was supposed to resolve.
If a list of intolerable views and behavior that are not owed a civil
response cannot be derived from any critical moral view, because that
view may itself be under dispute, is there any way of specifying what is
not owed a civil response? For that matter, is there any way of specifying
what is owed a civil response? Yes. It is no accident that Miss Manners
and Gutmann and Thompson both choose sexual harassment and racial
discrimination as examples of the intolerable. These are moral matters
on which there is presently extensive social consensus (which is not to
say unanimity). Standards of civility reflect that social consensus. We
need not respond civilly to a view or behavior once there is social closure
on its intolerability. At that point, civility would not further the work of
enabling the nonlike-minded to continue political dialogue or social interaction. However, when there is social dispute over the tolerability of a
view or behavior, being civil has a point. That the dispute is occasioned
by others moral misguidedness is irrelevant to the question of whether
we owe others a civil response. All that is relevant is the fact of social
dispute.
In sum, analyses of civility that equate being civil with treating
people in genuinely respectful, considerate, and tolerant ways, and that
set the bounds of civility by appealing to some socially critical moral
framework to determine what is genuinely intolerable, misconstrue
what civility is about and why there are bounds to civility. They assume
that civility is a virtue we are required to exercise toward others only if
those others pursue genuinely morally acceptable views and behavior.
(Of course, moral acceptability will have to be judged from the point of
view of some particular socially critical moral framework, and others
may disagree that ours is the best one.) Thus civility is owed only to
people who have (in ones own best judgment) gotten it more or less
97

MOR A L A IMS

right. People one judges to have gotten hold of a morally pernicious view
are not owed a civil response. This makes civility a close kin to integrity,
since our refusing to respond civilly to a view we judge morally pernicious is one way of standing up for our moral views. Indeed, using a socially critical moral framework to set the bounds of civility assumes that
those bounds are there to safeguard our integrity by exempting us from
dignifying what we, as individual moral reasoners, take to be morally
pernicious views.
By contrast, I think civility is a virtue that we are required to exercise toward others only if they pursue socially acceptable views and
behavior. At no point do norms of civility presuppose socially critical
moral judgments about either what views are worth respecting or what
counts as respectful, considerate, and tolerant behavior. Instead, they
presuppose social understandings about what views are still debatable,
as well as social understandings about what actions are sufficiently respectful, tolerant, and considerate to be worth a civil response. Only
by appealing to social understandings can civility norms successfully
regulate disputes and interactions with others under social conditions
where our different critical moralities lead us to differ over what positions deserve respect and what actions treat others with respect, tolerance, and considerateness. In addition, as I argued in section III, only by
appealing to social understandings can civility norms provide us with
a common language for displaying respect, tolerance, and considerateness to each other under social conditions where our different critical
moralities create dispute over what genuine respect, tolerance, and considerateness amount to. Civility is thus akin not to integrity but to civil
obedience. 27 Both civility and civil obedience may require compliance
with social norms or laws that are objectionable from a socially critical
moral point of view. Neither aims to safeguard our integrity; both aim to
safeguard the possibility of a common social life together.
Because standards of civility are tied to social understandings, there
is no guarantee that those standards will exempt us from civilly responding to what we as individual moral reasoners judge to be intolerable. That
27. Rawls takes using the imperfections of the law as an excuse for civil disobedience to be
an instance of incivility; A Theory of Justice, 355.

98

THE V IRTUE OF CI V ILIT Y

we now collectively regard racism and sexual harassment to be beyond

the bounds of what is owed a civil response reflects a social achievement.
Standards of civility can also reflect social failures to acknowledge the
real moral intolerability of some views and conduct. Civility may require
respectful dialogue about morally contemptible views and tolerant responses to morally intolerable conduct. Given extensive social disagreement over the moral status of homosexuality, for example, civility may
require what, from ones own socially critical moral viewpoint, seems
excessive accommodation to prejudice. This suggests that standards of
civility may directly conflict with morally admirable refusals to dignify
what, in ones own best judgment, is morally intolerable. They may thus
require forgoing speaking and acting with moral integrity.
It is precisely this sort of result that may incline one to think that civility cannot be a moral virtue on my account. There is nothing morally
virtuous, one might naturally think, about obediently complying with
merely social norms. A trait that is not directly regulated by a socially
critical moral perspective cannot be a moral virtue. And any trait that
cannot be brought into unity with such an important moral virtue as
integrity must not itself be a moral virtue.
These reasons for not counting civility among the moral virtues
are, I think, plausible only so long as one ignores how deeply social the
enterprise of morality is. It is deeply social in two respects: morality is
fundamentally about the social relations among people, and morality is
always practiced within a social world that shares some moral understandings and disputes others. Morality calls on us not just to do right by
others, for example, to refrain from what in our best judgment amounts
to cruelty or coercion or stinginess. It also calls on us to communicate
fundamental moral attitudes of respect, toleration, and considerateness.
This, I have argued, can only be done by relying on socially shared moral
understandings of what counts as displaying these attitudes. Civility
requires obedience to social norms not for their own sake but for the
sake of one important moral aim: the communication of moral attitudes
to fellow inhabitants of our moral world. It is this that makes civility a
moral virtue.
In addition, morality calls on us not just to be critically reflective, to
search for moral justifications, and to enact what we take to be the most
99

MOR A L A IMS

defensible moral views. It also calls on us to aim for mutual agreement

on moral norms. Indeed, it is hard to imagine a commitment to morality
that isnt also a commitment to seeing that that morality gets instantiated in our social world. Reaching real mutual agreement (as opposed
to hypothetical agreement in, say, a Rawlsian original position) requires
regulating moral dialogue so that conversations do not break down.
This, I have argued, can only be done by relying on socially shared moral
understandings of what positions are reasonably disputable and thus
worth a respectful hearing. Civility norms work to regulate disputes
precisely because they do not appeal to socially critical moralities that
may themselves be under dispute. Thus not being regulated by a critical
morality is central to civilitys being a moral virtue.
And what of the conflict with integrity? Given the complexity of
moral life, it may be unwise to expect or desire a unity of the virtues.
As moral participants we must function in two very different roles. On
the one hand, we must be socially critical moral reasoners, exercising
our best judgment as individuals who aim to get it right. On the other
hand, we must engage in the communal practice of morality, relying on
a common language for displaying respect and striving for communal
progress toward better moral practices. Whether one can have integrity
and be civil will largely depend on whether shared moral understandings are reasonably decent. In inegalitarian societies, the biases embedded in standards of civility mean that the socially disesteemeds pointed
demands for more respect are sometimes construed as uncivil; those
biases also mean that expressions of contempt for the socially disesteemed are often construed either as not incivilities at all or, at worst,
as tolerable incivilities that are owed a civil response. In such morally
imperfect social worlds, the choice may have to be made between being
civil and acting with integrity. One may have to choose, for example,
between a moderate, conciliatory response to the exclusion of ones
same-sex partner from family events and a response that makes clear
how intolerably disrespectful such an exclusion is.
In morally imperfect social worlds where civility norms fail to protect the disesteemed from treatment that is genuinely disrespectful, inconsiderate, and intolerant (even though not socially understood to be
so), it is tempting to reject the value of civility altogether. This, I think,
100

THE V IRTUE OF CI V ILIT Y

is a mistake. Members of disesteemed social groups are more likely to

experience displays of contempt, intrusions on their privacy, intolerance
of their conceptions of the good, and the discounting of their feelings
and aims as less important. The last thing they need is for the privileged
to be acting out, without restraint, their personal views, whether they
be about homosexuals, or independent women, or Jews, or blacks. What
they need is precisely for the privileged to feel constrained to control
their hostile, contemptuous, disapproving, and dismissive attitudes.
Those constraints will be supplied, if they are supplied at all, by norms
of civility, because civility is the display of respect, tolerance, and consideration toward others no matter what we might privately think of
them. What the disesteemed also need is for there to be shared social
understandings about the intolerability of prejudiced and oppressive behavior. Those shared social understandings, if they exist, will define the
bounds of civility. While it is true that in morally imperfect social worlds
civility norms fail to protect the disesteemed, the problem is not that civility is overvalued, and the solution is not to care less about being civil.
The problem is in the shared understandings embedded in our norms of
civility. These need to be contestednot the value of civility in general.
When women first expressed outrage at mens sexualized behavior at
work and in the classroombehavior that at the time was socially construed as innocent flirting, not harassmentthey were not rejecting the
value of civility because existing norms required them to respond civilly
to sexual advances. They were pressing for an evolution in our shared
social understandings, and with it, a new way of being civil.28
Finally, what is there to say in response to the objection that civility is
at best a minor virtue because, when push comes to shove, treating people
with genuine respect (and demanding it for ourselves) matters more than
communicating respectful attitudes or keeping dialogue and social interaction going? Shouldnt the virtue of civility weigh only lightly on the
moral scales? My own view is no. This is in part because I do not share what
seems to be a common conviction among moral theorists that morality
is, first and foremost, about getting it right as individualsgetting the
28. Lawrence Cahoone makes a similar point in his Response to Alan Wolfe in Civility,
ed. Leroy S. Rouner (Notre Dame, IN: University of Notre Dame Press, 2000), 148.

101

MOR A L A IMS

right critical morality and acting on it. In the ways I have just suggested,
morality is also something we do together. The more seriously we take
the social practice of moralitycommunicating attitudes, collectively
revising moral norms, sustaining the activities we care morally about
(parties, funerals, friendships, being neighbors)the heavier civility
weighs on the scales. In part, too, I am inclined to weigh civility heavily
on the scales because I find something odd, and oddly troubling, about
the great confidence one must have in ones own judgment (and lack of
confidence in others) to be willing to be uncivil to others in the name of
a higher moral calling. When one is very sure that one has gotten it right,
and when avoiding a major wrong is at stake, civility does indeed seem
a minor consideration. But to adopt a principle of eschewing civility in
favor of ones own best judgment seems a kind of hubris.

102

C ha pt e r4

Common Decency
Charles Dickenss Ebenezer Scrooge is a portrait of a man without
common decency. Scrooges central failing is not his miserliness or callousness toward suffering. His sometimes spectacularly contemptible
failingsas when he suggests the poor should simply get on with dying
and reduce the surplus populationare connected to a less spectacular but more pervasive failing: Scrooge has removed himself from the
daily commerce of favors, mercies, small kindnesses, forgivings, expressions of gratitude, and social pleasantries that are the stuff of common
decency. He gruffly rebuffs his nephews invitation to Christmas dinner.
He grumbles at being expected to let his employees off Christmas day.
He threatens to take a ruler to a Christmas caroler. And he refuses even
the smallest compliance with the convention of charitable giving during
the Christmas season. Though we see Scrooges faults at the Christmas
season, his failing is not seasonal. Scrooge routinely fails to behave like
a decent human being; and for that reason no one ever stops him in
the street to say, with gladsome looks, My dear Scrooge, how are you?
When will you come to see me? No beggars implored him to bestow a
trifle, no children asked him what it was oclock, no man or woman ever
once in all his life inquired the way to such and such a place, of Scrooge.1
That Scrooge has no truck with simple favors, like telling others the
time or giving them directions, signals his lack of common decency.
Paying Bob Cratchit barely a living wage, relentlessly collecting debts
from the already impoverished, and displaying an indecently callous
attitude toward the destitute are simply more egregious examples of
1. Charles Dickens, A Christmas Carol and Other Haunting Tales (New York: Doubleday,
1998), 260261.

103

MOR A L A IMS

Scrooges general inability to live up to our moral expectations about

how minimally well-formed agents will behave.
In disappointing expectations about how a minimally well-formed
agent will behave, Scrooge does not invite others resentment or moral
indignation. Nor is guilt what he comes to feel about his past bad behavior. In A Christmas Carol, Scrooges lack of common decency is most
often met with surprise, pity, contempt, mockery, and cooled affections.
What Scrooge himself comes to feel about his lack of common decency
is not guilt but a mixture of shame and loss of human connection.
But for all that Scrooge is an odious, stingy, hard, unfeeling man, 2
there is no one whom he clearly wrongs. It may be indecent to insist that
his employees work on Christmas day, but he correctly observes that
since they also expect to be paid, he does not owe them this day off. Nor
does he owe his nephew pleasantries, or Christmas carolers something
for their cheer, or Bob Cratchit higher wages than agreed upon, or his
debtors a grace period in meeting their debts, or any particular charitable organization a donation. These are all gifts that he is within his
rights to refuse to bestow. As for what Scrooge owes others, Dickens
gives us no reason to think Scrooge fails to render what is due. On the
contrary, Scrooge is obsessed with debts. He wants nothing more from
others than exactly what they owe him. In return, he will give others exactly what he owes, and not a bit more. His business and moral ledgers
carefully track debts payable to and by him, making no allowance for
giving or receiving that exceeds the obligatory. And this is the source of
his failure of common decency. For Scrooge, others are morally entitled
to expect only what is rightfully theirs. He is unable to see the moral
legitimacy of their expectation that he will give them the grace periods, sympathetic ear, relief from work duties, living wages, the time of
day, and sociability that are just matters of common decency. Scrooge
sees nothing morally objectionable about removing himself from commerce with others needs. Its not my business, he says. Its enough for
a man to understand his own business, and not to interfere with other
peoples.3
2. Ibid., 316.
3. Ibid., 268.

104

COM MON DECENCY

The common decencies and failures of common decency at the

center of Dickenss A Christmas Carol are familiar ones. Yet, from a philosophical point of view, common decency is puzzling. Acts of common
decency seem to occupy a shadowy territory between the obligatory and
the supererogatory. On the one hand, Scrooge seems within his rights to
withhold the kindnesses and mercies that are emblematic of common
decency. He in fact doesnt owe his debtors grace periods or his nephew
pleasantries upon their meeting. Yet those around him also seem justified in responding with moral contempt and a severing of social bonds.
They rightfully find moral fault with his behavior. But how can one be
faulted for failing to give what was never owed? What sense can be made
of our treating acts of common decency as though they were not obligatory, but not purely elective either?
In what follows, I suggest that the normative expectations connected with common decency do not derive from a conception of what
we owe each other. Instead, they derive from a constructed conception
of what can be expected of a minimally well-formed moral agent.

I. T WO SPECIES OF COM MON DECENCY

Since the term decency has many uses, let me say a bit about what I
have in mind by common decency. Then we can turn to the puzzling
normative status of common decencies.
The term decent, like the terms good or mediocre, is a grading
term. Anything that can be graded could receive the grade of decent.
We speak, for example, of a decent cup of coffee, a decent selection of
items, a decent society, a decent system of law, and decent housing. The
core meaning of decent in all these cases is adequate or minimally acceptable as good. What is decent just satisfies the standard for items of a
particular kind. A decent cup of coffee is a good cup, but only just. Its
decent offers only faint praise and draws attention to what is only a cut
above the shamefully inadequate.
Sometimes what is minimally acceptable and only a step from
shamefulness is the agents moral performance. The notion of minimally
acceptable moral performanceand thus of common decencycan
105

MOR A L A IMS

be understood in two different but related ways. First, we sometimes

equate common decency with fulfilling ones minimal moral duties.
Philosophers in particular tend to construe common decency this way,
although they disagree on whether all duties or just some duties are
minimal.4
Doing what morality minimally requires is, I think, an important
form of common decency. 5 But it is not the philosophically most interesting form of common decency. Think back to Ebenezer Scrooge.
Scrooge lacks common decency not because he shirks his minimal
moral obligations but because he fails to live up to others very strong
expectations about the moral giftspleasantries, mercies, kindnesses,
and favorsthey can count on receiving from any agent who is at least
minimally well-formed. This form of common decency is, as John Kekes
puts it, especially connected with moral attitudes that call upon one
to go beyond the rules.6 It involves good will toward fellow members
of the society, a reluctance to injure others in pursuit of our own ends,
even if we have the right to pursue our ends. It is the attitude opposite to
extracting our pound of flesh. 7 Thus although it is true that we sometimes say of a person At least she had the common decency to do what
4. Susan Wolf, Above and Below the Line of Duty, Philosophical Topics 14 (1986): 131
148, observes that the goal of a theory of duty is to set minimal standards of moral
decency (135). Those standards tell people who wish to be decent that they must at
least do this much (139140). Even some utilitarians try to specify the minimal moral
obligations that are matters of common decency; see, J. O. Urmson, Saints and Heroes,
in Moral Concepts, ed. Joel Feinberg (London: Oxford University Press, 1969). This conception of decency also appears to be at the heart of Avishai Margalits The Decent Society
(Cambridge, MA: Harvard University Press, 1996). A decent society, in his view, is one
in which people are treated in nonhumiliating ways in the basic institutions of society.
5. There are two basic positions on how many of our moral duties are minimal ones. First,
one might think that all moral duties are minimal because morality is not very demanding and does not require much of us by comparison to the full range of morally good
things that we might do for others. If morality is not demanding, then one can expect
any minimally well-formed agent to be able to do her duty. This will be just common
decency. Alternatively, one might think that some moral duties are quite demanding and
that our minimal duties are those that do not tax the motivational capacities of a minimally well-formed moral agent.
6. John Kekes, The Great Guide of Human Life, Philosophy and Literature 8 (1984): 236
249, 243.
7. Ibid., 248 (emphasis mine).

106

COM MON DECENCY

she agreed to do, common decency is not simply a matter of living up

to minimal moral obligations. Common decency also names the basic
sorts of things that we expect any minimally well-formed agent will elect
to do for others absent any requirement to do so. 8
These two forms of common decencyfulfilling minimal moral
duties and giving those moral gifts that are only to be expectedshare
a common core: common decency has to do with what can be expected
from any minimally well-formed moral agent. To have common decency
is to be a good or acceptable moral agent, but just barely.
I now set aside the common decency of fulfilling minimal moral obligations and turn exclusively to the common decency of giving those
moral gifts of kindness, mercy, pleasantness, and so on that are only to
be expected of a minimally well-formed agent.

II. COM MON DECENCY, SUPER EROGATION,

A ND OBLIGATION
These are the common decencies that appear to occupy a shadowy territory between the supererogatory and the obligatory. Consider first their
relation to supererogation. Common decencies differ from typical supererogatory acts because they are expected of agents and are shameful to
omit. But common decencies share with supererogatory acts the feature of
being non-obligatory.9 As Scrooge understood so well, common decencies
are electivegifts one is morally free to give (or not). Because of this, the
8. The O.E.D. defines a decent person as someone who is kind, accommodating,
pleasantnot as someone who does her minimal duties (O.E.D Online, definition 5b).
Acts of common decency, in this sense, belong on the same scale as George Baileys uncommonly decent acts in the movie Its a Wonderful Life (1947). Both common and uncommon decency involve interfering for the better in others lives through moral gifts
of kindness, compassion, generosity, charity, mercy, forgiveness, patience, pleasantness,
thoughtfulness, and the like.
9. David Heyds observation in Supererogation: Its Status in Ethical Theory (Cambridge:
Cambridge University Press, 1982), 148, about one kind of common decencyfavors
applies generally to acts of common decency: They may be deserved or undeserved,
done spontaneously or as a response to a request. Yet they are never deserved as a matter
of right, and a refusal to do a favour cannot be criticized as morally wrong. We can ask for
a favour, but never claim it.

107

MOR A L A IMS

kindnesses, mercies, favors, and the like that constitute common decency
seem to fit quite naturally within the basic categories of supererogatory
acts.10 Those categories are: (1) favors, (2) acts of beneficence, (3) volunteering, (4) mercy and forgiveness, (5) praising, congratulating, and honoring, (6) gratitude, (7) gift giving, and (8) saintliness and heroism.11
Saintly and heroic acts, obviously, are not matters of common decency.
But each of the remaining seven categories contains some mixture of
common decencies that are expected of all minimally well-formed agents
and especially virtuous acts that could only be expected from unusually
well-formed agents and thus are left fully to the agents discretionary
judgment. How do we determine which acts are common decencies? Let
me propose for the moment that, as a general rule, any act falling into categories (1) through (7) that has been socially conventionalized, so that it
is just what is done, will be a matter of common decency. Giving ones
child a birthday present is, for example, socially conventionalized. So too,
in many organizations, is volunteering to take ones turn at some undesirable task (for example, serving as department chair). Holding a strangers
place in line, giving directions or the time to those who ask, opening the
door for those whose hands are full, and giving up ones bus seat to the
elderly are familiar conventionalized favors. Such conventionalized giftings, volunteerings, and favors are matters of common decency.
Although common decencies resemble supererogatory acts in
being morally good but nonobligatory, they also differ in one important respect: they are not fully morally elective. An act is fully morally
electivewhen
1. Omitting the act is not morally criticizable.
2. No ought stronger than an ought of moral advice-giving is
appropriately used to recommend it.
3. Choosing the act is meritorioussomething we commend or
admire the agent for doing rather than take as owed or simply
to be expected.
10. These categories are derived, with some alteration, from Heyds list in Supererogation,
chap.7.
11. Obviously, there is some overlap between these categories. Gift givings (particularly
charitable gifts), favors, and volunteerings can be forms of beneficence.

108

COM MON DECENCY

4. Gratitude untempered by any thought that one has some

moral title to the gift bestowed is the proper response to
theact.

Common decencies are not fully morally elective in any of these

senses. First, people who dont manage to do what is just a matter of
common decency are criticizable. They are not criticizable for wronging others, but their failure to give expected moral gifts does open them
to the charge of being petty, mean-spirited, contemptible, disappointing, irritating, and a poor excuse for a moral agent.12 Such criticism
underlines the subpar nature of the moral performance. Contempt,
pity, cooled affections, resentment, and (the agents) shame are all appropriate reactive attitudes to failures of common decency. By contrast,
supererogatory acts are ones whose omission does not warrant moral
criticism or negative reactive attitudes.
Failures of common decency thus have an odd status. On the one
hand, they are not wrongs. On the other hand, they are not morally acceptable omissions. Scrooges mercilessness toward his debtors, for example, was clearly taken by others to be an offense, yet one that he was
not morally obligated to avoid. Failures of common decency thus seem
to have the interesting status of being morally disvaluable acts that are
nevertheless permitted rather than forbidden. Roderick Chisholm has
called acts having these features permissive ill-doings (or offenses);13
and Julia Driver has described them as suberogatory.14

12. Common decency thus does not fit David Heyds familiar description in Supererogation
of the supererogatory: decisions that concern the truly supererogatory are free not
only from legal or physical compulsion, but also from informal pressure, the threat of
moral sanctions, or inner feelings of guilt. It is purely optional (175).
13. Roderick M. Chisholm, Supererogation and Offence: A Conceptual Scheme for
Ethics, Ratio 5 (1963): 114.
14. Julia Driver, The Suberogatory, Australasian Journal of Philosophy 70 (1992): 286
295. Drivers category of the suberogatory is possibly a bit broader than what I have in
mind by the decent. For her, the suberogatory are bad acts that involve a failure to act
on an ideal (e.g., the ideal of nonwastefulness). I want to stress that they are failures to
live up to the standard of being a minimally acceptable moral agent. The differences
here, however, are not large, and her piece is full of important insights about the nature
of these types of acts.

109

MOR A L A IMS

Second, common decencies are not fully morally elective because

even if the ought of obligation does not apply to common decencies, an
ought that is considerably stronger than mere advice giving does apply.
Joel Feinberg is perhaps best known for making this observation.15
Feinberg uses the example of one kind of common decencysimple
favorsto argue that there are some actions which it would be desirable for a person to do and which, indeed, he ought to do, even though
they are actions he is under no obligation and has no duty to do. He invites us to imagine being approached by a stranger who politely asks for
a light. Ought I to give him one? he asks, and replies, I think most
people would agree that I should, and that any reasonable man of good
will would offer the stranger a match.16 The sense of ought here falls
somewhere between a command to do ones duty and the observation
that this is one among many morally good acts that one might elect.
Third, common decencies are not fully moral elective because they
establish our minimal acceptability as a moral agent; they do not signal
our achievement of a virtuously high standard of moral agency. This is
why omitting common decencies is criticizable. It is also why choosing
to behave with common decency is not meritorioussomething that
we commend or admire the agent for doing rather than take to be owed
or simply to be expected. In the United States, for example, tipping waitpersons 15 to 20 percent is a common decency, only to be expected of
any minimally well-formed agent who is familiar with tipping conventions. It is not an indication of commendable virtue.17 By contrast, supererogation is the domain of commendable and admirable virtue.
Fourth, the proper response to a fully elective moral gift is gratitude. The proper response to being shown common decency is at most
perfunctory gratitude. Because we are normatively entitled to expect
common decency from others, gratitude in excess of simple thanks for
15. Joel Feinberg, Supererogation and Rules, Ethics 71 (1961): 276288.
16. Ibid., 276277.
17. An agent could not omit common decencies and still claim to have a virtue like generosity. But to be generous is to be disposed to treat others also in some ways that exceed
mere common decency. Purchasing the prize goose for Bob Crachits family exceeds
common decency, and it provides some evidence of Scrooges commitment to becoming a better, more generous person.

110

COM MON DECENCY

commonly decent treatment would be misplaced. Given this difference

between common decencies and supererogatory moral gifts, a good
way to discern which favors, mercies, volunteerings, and the like are
just matters of common decency is to ask oneself, What favors (mercies, volunteerings, etc.) could I ask of others without putting myself in
the position of incurring a debt of gratitude for a meritorious display of
goodwill? Some ways of filling in requests like Would you do me a favor
of... ? Could you spare... ? Would you mind letting me... ? and
Could you tell me... ? clearly impose on others goodwill and would, if
granted, incur a debt of gratitude. In other cases, we simply assume that
others should be willing to grant our request because we arent asking
for a meritorious display of goodwilljust common decency.

III. UTILITA R I A N A ND K A NTI A N OBLIGATION

TO SHOW COM MON DECENCY
In sum, common decencies appear to occupy a hybrid category, sharing some features of obligation and some of supererogation. One might,
however, balk at this idea. In particular, one might object that if common
decencies are what we ought to do, then common decencies are obligatory.18 Both utilitarians and Kantians would probably insist that the
injunction You ought to do that; its just common decency points to
an obligation. A utilitarian might take common decencies to be strictly
obligatory, as a rule, because common decencies benefit others but cost
the agent little. And utilitarians think that we are always obligated to do
whatever will maximize welfare.
Kantians would probably categorize common decencies among imperfect duties. If common decencies are imperfect duties, this would explain why the ought recommending common decency seems weaker
than the ought of obligation. No act fulfilling an imperfect obligation
18. Heyd, Supererogation, raises this worry in connection with Feinbergs description of
favors, but the point applies equally to all common decencies. As Heyd points out, if
the ought means just the best thing to doan advicethen favours are supererogatory; and if ought means a kind of requirement, how can Feinberg say that favours are
never obligatory? (150).

111

MOR A L A IMS

is strictly required. Imperfect obligations simply require that one perform some acts of a particular kindfor example, some possible favors.
The ought recommending, for example, doing someone the favor of
holding her place in line is thus not the strong ought of perfect obligation that commands what we must do now. It is the weaker ought
requiring that we do some favors, but not necessarily this one now.
Both utilitarian and Kantian approaches solve the puzzle over the
normative status of common decencies by denying there ever was a
puzzle: the ought recommending common decency just is the ought
of obligation. Neither approach, however, enables us to retain much of
the ordinary conception of common decency. Consider, first, the utilitarian view. A utilitarian would have to insist that common decencies are obligatory in just the way that keeping promises and telling
the truth are obligatory. Thus the utilitarian would have to insist that
Scrooge wasnt just criticizable for not giving his debtors a grace period,
but that he actually wronged them. A utilitarian would also have to
drop the idea that common decencies are more strongly required than
are saintly mercies and kindnesses, but are less strongly required than
minimal promise-keeping, truth-telling, and so on. All acts that maximize utilitywhether acts of promise keeping or of common decency
or of saintly beneficenceare equally obligatory. Of course, a utilitarian could try to argue that our commonsense distinctions between
the strictly obligatory, the commonly decent, and the saintly are useful
fictions to preserve.19 Perhaps we gain something when people are left
free to elect to be decent, and freer yet to be saintly, rather than feeling
obliged. But this still amounts to jettisoning, at the metalevel, our everyday distinctions between different degrees of oughtness.
The Kantian, too, must reject the idea that there are different degrees of oughtness. First, both common decencies and unusually virtuous moral gift givings are simply different ways that agents might elect
to discharge their imperfect obligations. The same ought of imperfect
obligation applies to both types of act. There is thus no obvious way of
19. Urmson, Saints and Heroes, for example, tries to give a utilitarian justification for
preserving a distinction between the obligatory and the higher flights of morality
that constitute the supererogatory. I thank Michael Smith for reminding me that these
are still fictional distinctions.

112

COM MON DECENCY

capturing the idea that common decencies are normatively expected in

a way that other moral gifts are not. Second, the Kantian cannot capture the idea that particular, individual acts of common decency are what
we ought to perform. One might ordinarily think that, absent a special
excuse, you really ought to give the match to the person who asks you for
one, and you are criticizable if you refuse. From a Kantian point of view,
however, what is criticizable is adopting a policy of refusing to render
assistance. Omitting a particular act that would discharge an imperfect
obligation cannot be criticized. Within a Kantian framework, the only
way to capture the individually criticizable nature of failures of common
decency would be to treat common decencies as matters of perfect obligation. That move, too, has a serious drawback. One of the distinguishing features of common decencies is that they involve not standing on
ones rights when one is entitled to. This is most obvious for the common
decency of not insisting on taking ones fair share; it is also obvious for
the common decency of being merciful or forgiving.
In sum, if common decencies are governed by the ought of
obligationeither strict or imperfectthen much ordinary talk about
common decency must be set aside as confused. The alternative is to see
if we can make sense of there being an ought that is weaker than the
ought of obligation but stronger than the ought of moral advice. Is
there some way of making sense of the idea that there are elective acts that
we would be criticizable for not performing?

IV. CONSTRUCTING THE CATEGORY

OF THE DECENT
What I want to propose is that the category of the decentwith its peculiarly hybrid propertiesis constructed out of an antecedently determined domain of supererogatory acts. What I have in mind is this:
We begin from some moral theory that enables us to determine what
acts are obligatory and what acts are supererogatory. The determination of the domain of the obligatory sets boundaries on what could
possibly be a matter of common decency. Something that is itself obligatory cannot be a matter of common decency, a moral gift that we are
113

MOR A L A IMS

within our moral rights not to give. Nor can violations of obligation be
common decencies, because they are morally prohibited. 20 Only supererogatory (elective and morally valuable) acts are candidates for common
decencies. The actual list of commonly decent acts is constructed from
those candidates. By constructed I mean that, unlike the obligatory
and the supererogatory, norms of common decency emerge only from
within a social practice of morality. Those norms articulate what moral
gift-giving participants in a particular social practice of morality are expected to elect.21 The expectation here is normative. It is not just that
we happen in point of fact to expect other people to be willing to do us
simple favors, or forgive us for small failings, or volunteer to take a turn.
We also take ourselves to be justified in having those expectations and
to have a legitimate basis for criticizing those, like Scrooge, who disappoint us.
This takes us to the central question. From what source does this
subset of nonobligatory, morally good acts that we call common decency get its heightened normativity?

Conventions
One account (which I will ultimately reject) of the heightened normativity attached to common decencies draws on the value of having and
sustaining social conventions of moral gift giving. The argument goes
like this:
It is often remarked within moral philosophy that securing the reliable performance of some acts has a special urgency. Our ability to
carry out any life plan at all would be seriously undermined if we could
not rely on others not to injure or kill us, to keep their agreements, to
respect our privacy and property, and to communicate with us truthfully. This form of reliabilityreliable forbearance from undermining
20. Thanks to Julia Driver for this point.
21. It is conceivable that a social practice of morality might operate without a conception of
common decency. In that case, it would not treat any supererogatory acts as ones that
agents are expected to elect; and it would not supply any basis for criticizing agents who
stand on their rights and refuse to show mercy, volunteer, forgive, do favors, and help
out when doing so is not obligatory.

114

COM MON DECENCY

others security or agencyis, indeed, of great moral importance. And

the concept of moral obligation works to secure that reliability. However, our need to rely on others extends well beyond matters of basic
security and nonmanipulated agency. Like Blanche Dubois, we find
that we unavoidably depend on the kindness of strangers. We need help
in carrying out our plans, emotional support, occasional release from
promises, forgiveness and mercy for errors, a grace period for repaying debts, and so on. That is, we depend on people electing to give us
moral gifts. Personal planning and social coordination are enhanced,
however, if some of what others might elect to do for us is routinized so
that we can have advance knowledge of the contexts in which we can or
cannot depend on others to help out. For example, when giving directions, telling the time, and lending a match are converted from fully
elective, supererogatory gifts into socially institutionalized, expected
gift givings, we can venture out in the world unburdened with maps,
watches, and lighters. Or, for example, when forgiving those who are
five or ten minutes late for appointments is conventionalized, we are
spared from always having to allow extra time to arrive. In short, optimal social functioning depends not only on individuals fulfilling their
moral duties toward others but also on the reliable exchange of moral
gifts. Converting fully elective supererogatory acts into normatively
expected ones by institutionalizing them in the shared, everyday moral
practice of a group of people produces that reliability. When socially
institutionalized, formerly fully elective acts such as picking up items
dropped by another, giving up ones seat on a bus to the elderly, and
letting those with only a few items go ahead of oneself in line become
things that a decent person ought to do, even if others cannot demand
them as a right.
Drawing on this idea that acts of common decency are part of an
institutionalize practice of moral gift giving, we can explain the heightened normativity of common decencies in one of two ways. First, it
is advantageous for there to be moral gift-giving conventions, rather
than leaving it entirely up to individual discretion which, if any, favors,
mercies, forgivings, volunteerings and the like they will do for others.
Supporting those conventions thus has moral value because those conventions are useful ones.
115

MOR A L A IMS

Alternatively, one might observe that the fact that common decencies are institutionalized practices of moral gift giving from which
everyone benefits means that those who insist on their right to refuse to
be decent are a kind of free rider. Whether they wish to or not, they in
fact benefit in myriad ways from others participation in the practice of
bestowing those moral gifts that constitute common decency. Indecent
people, like Scrooge, reap the benefits of moral gift-giving conventions
without doing their part in this system of reciprocal favors, mercies, volunteerings, and forgivings. And that is unfair.
Now, here is what I think is the problem with this way of explaining
why we ought to treat others with common decency and are criticizable
if we dont. An appeal to the social utility of moral gift-giving conventions and the unfairness of free riding on those conventions justifies too
much. Common decencies turn out to be not just obligation-like. They
are obligatory. Many have argued, for example, that the usefulness of a
conventionalized practice of promising and the unfairness of free riding
on that practice ground an obligation to keep promises. So, if we are
going to make sense of the electiveness of common decencies, we need
an account of their normativity that does not draw on the moral value of
sustaining useful conventions or of avoiding free riding on them.

Minimal Agency
A second accountwhich I think is the better accountshifts our attention from the status of norms recommending decent conduct to the
status of the identity that behaving decently sustains. Scrooge doesnt
just behave badly. He disappoints our expectations for how any minimally well-formed agent will behave. The moral importance of the
identity minimally well-formed agent generates the normativity of
common decency. That identity is morally important because any functioning practice of morality must presume that its practitioners are capable of meeting a minimal standard of moral performance. I now turn
to a more detailed explication of the central ideas in this second account
of the normativity of the ought recommending common decency.
The thought that we can expect any minimally well-formed agent
to do x, y, and z arises both for obligations and for elective moral gifts.
116

COM MON DECENCY

That is why there are, as I observed earlier, two related forms of common
decency: one pertaining to minimal moral obligations, and one pertaining to minimal moral gift giving. Consider, first, our expectations about
obligatory moral performance. Although all of us stand under the obligation to do our duty and moral failures meet with criticism, we nevertheless tolerate a good deal of moral backsliding. We tolerate it in the
sense that much wrongdoing seems unsurprising and a normal hazard
of everyday moral practice. We expect to meet with and accommodate
a good deal of moral misbehavior that results from a variety of character shortcomings. We know that variations in natural and acquired
dispositions, moral education, and strength of will result in variations
in both individuals moral performances and their overall success as
moral agents. There is, however, a baseline that we expect agents, no
matter their individual character and temptations, to be able to manage
to achieve. Even if it would be unreasonable to expect that fellow moral
agents will always do what they ought, there are at least some things
it is reasonable to expect. 22 Those who disappoint these expectations
compound the wrongfulness of what they do with the senselessness of
subjecting others to what even the most minimally well-formed agent
should have been able to manage to avoid.
In general, acts that are reasonably expected of even minimally wellformed agents are, first, acts that are not motivationally taxing. They cost
the agent very little. Doing them is, as it were, no skin off ones nose. Nor
do they presuppose any appreciable degree of virtue. As a result, excuses
appealing to temptation or understandable failures of virtue are unavailable. Second, they are acts whose moral value in the situation at hand is
obvious and unambiguous. So excuses like I didnt realize I should...
or I wasnt sure I ought... are not plausible. Third, in virtue of their
being motivationally nontaxing, obvious, and unambiguous, they are

22. Failures to meet that baseline typically meet with a different response than do other
sorts of moral failures. Our reactive attitudes of resentment, indignation, and contempt
are typically calibrated to the expectation that the moral agents with whom we share a
daily practice of morality will behave as minimally adequate moral agents. Failures of
common decency and failures to fulfill minimal moral obligations generally meet with
heightened indignation, resentment, and sometimes contempt.

117

MOR A L A IMS

the sorts of acts whose omission is not open to standard excuses, and
this is why we so strongly expect people not to omit them.
Some obligatory acts are like this. Some supererogatory acts are like
this, too. The domain of the supererogatory covers acts that vary widely
in the degree to which they tax agents motivational resources. Some supererogatory acts, particularly the saintly and the heroic, entail significant losses for the agent. Because of that, their performance requires
exceptional motivational resources. So we understand why people do
not usually elect these forms of supererogation. The domain of the supererogatory, however, also includes many unspectacular acts that are
motivationally nontaxing. Although everything in the domain of the
supererogatory is elective, the further one moves away from the saintly
and heroic, the more reasonable it becomes to wonder why one would
not elect to do this or that morally valuable act. As we imagine motivationally less and less taxing supererogatory actssuch as doing simple
favors or engaging in idle pleasantrieswe find it increasingly difficult
to make sense of a persons refusing or neglecting to elect them. This is, in
part, because the level of goodwill, concern for others welfare, and commitment to the value of rational agency that moves a person to satisfy her
minimum obligations should also move her to elect some morally good,
but nonrequired acts. Someone who only did what duty required and
elected no supererogatory acts would, thus, not be a plausible candidate
for a minimally acceptable agent.23 On the contrary, when someone like
23. An analogy may help press this point. It would be odd to equate a minimally adequate
professor with one who only does what duty requires. Suppose duty requires that a faculty member have some number of office hours, give some written evaluation of student
papers, and do some committee service but leaves as a matter of election how much.
Those who meet with students as little as possible, who return papers with hardly a
word of comment, and who decline to serve on all but the least demanding committees
shirk no professional obligations. However, they also do not live up to the expectations
for minimally adequate professorial performance. More is expected of them precisely
because the same commitment to academic ideals that provides a reason to fulfill their
professorial obligations also provides a reason for discretionary elections that advance
those ideals. In short, a minimally well-formed professor would choose more than what
her obligations require. Similarly, a minimally well-formed moral agent would choose
more than what her obligations require. A commitment to the values that provide a
reason to fulfill her moral obligations should also provide a reason for discretionary
elections that advance those values.

118

COM MON DECENCY

Scrooge doesnt elect even the least motivationally taxing supererogatory acts, we have to suppose that something has gone wrong with his
moral psychology. He suffers, perhaps, from excessive self-absorption
or deficient sympathies. In this way, reflection on what can be expected
of a minimally well-formed moral agent leads us to construct a conception of commonly decent moral gift givings from the larger domain of
the supererogatory.24 Those gift givings retain their elective character,
but their incorporation into our conception of what any minimally wellformed moral agent would elect heightens their normativity.
Clearly, however, not every supererogatory act that is motivationally nontaxing is a matter of common decency. There are endless favors,
mercies, kindnesses, forgivings, volunteerings, praisings, and present
givings that we could do for others that are relatively cost free. Most
are not expected of all minimally well-formed agents. Stooping down
to tie a strangers shoelace when his hands are full of packages, for example, is no more motivationally taxing than stepping forward to open the
door for him. Yet shoe tying is not a matter of common decency, while
door opening is. So why are some motivationally nontaxing moral gifts
matters of common decency and others not? The obvious difference between shoe tying and door opening is that opening doors for others is a
socially conventionalized moral gift giving; tying strangers shoelaces
is not. Such conventions convert supererogatory acts into common
decencies.
Social conventions can convert supererogatory acts into common
decencies in part because they make it obvious and unambiguous
what it would be good to elect. When there are no conventions, giving
people moral gifts can be problematic in all the ways that giving people
ordinary material gifts sometimes is. We may give the appearance of
bribing, currying favor, being paternalistic, taking liberties, showing
favoritism, or seducing. This was the problem with tying the strangers
24. Lawrence Blum, Community & Virtue, in How Should One Live: Essays on the Virtues,
ed. Roger Crisp (Oxford: Clarendon, 1996), also draws a somewhat similar distinction between levels of virtuenoteworthy virtue and ordinary virtue. He describes
ordinary virtue in a way that captures what I have in mind by common decency. Acts of
ordinary virtue are simply what are to be expected of a normal moral agent; they are
not regarded as meriting distinct praise or esteem (235).

119

MOR A L A IMS

shoe. What was intended as a kindness may come across as an invasion of privacy, presumptuousness, paternalism, or a bit of seduction.
So while tying the strangers shoe may be motivationally nontaxing, its
uncertain reception makes it neither obviously nor unambiguously a
good thing to do. Conventions disambiguate. They render obvious and
unambiguous the desirability of, say, opening doors for strangers with
their hands full.
Conventions also affect what agents do and do not take to be motivationally taxing. When there are moral gift-giving conventions in
place, agents expect the costs associated with those conventions. When
you board a bus, you expect to give up your seat to an elderly passenger. When you go to a dinner party, you expect to bring a token gift.
When you teach a course, you expect to give some grace periods. Such
expected costs are not burdensome because our plans and expectations
for ourselves already include their possibility. We dont feel particularly
burdened by giving up our seat because doing so is not an additional
cost of riding the busit comes with the territory of riding the bus.
So, too, bringing a token gift comes with the territory of dinner parties,
and showing occasional mercy to students comes with the territory of
teaching.
In short, gift-giving conventions determine which elective acts will
be motivationally nontaxing and obviously and unambiguously desirable. But this means that there is no one standard for being a minimally
well-formed moral agent. The moral gift-giving conventions of actual
moral practices supply the standard. Common decency is thus always
a local construction.
Decent people are, then, like decent cups of coffee or decent housing. Their decency is relative to local standards. A decent cup of coffee in
Nebraska is not a decent cup of coffee in Italy. Decent housing in rural
South Carolina is not decent housing in San Francisco. This is not to say
that there are no objective limits to what could count as decent housing
or coffee. Any decent housing must provide some protection from the elements. Any decent coffee must use noncontaminated water. But these
are very general guidelines. Local conventions supply the substantive
content. Those conventions may set the bar higher or lower for decent
coffee or decent housing.
120

COM MON DECENCY

So, local moral gift-giving conventions supply the substantive content for the concept of common decency. Here, too, there will be objective limits to what could count as common decency. Common decencies
cannot strain human nature with their motivational demands. But just
as the standard for a decent cup of coffee may vary with locale, so may
the standard for common decency.
Conceptions of common decency can vary horizontally: Among the
vast array of motivationally nontaxing supererogatory acts, different
moral practices might conventionalize different sets. So, for instance,
California Bay Area residents conventionally gift each other with enormous forbearance in wearing perfumed products; but they lack conventions for doing drivers the favor of permitting them to change lanes.
Elsewhere, one finds conventions for doing fellow drivers favors, but
none for avoiding perfume.
Conceptions of common decency might also vary vertically. Some
locales may have lower standards all around for commonly decent behavior. The villagers in Le Chambon during World War II constructed
what seems to us an extraordinarily high standard of decency. They
clandestinely assisted approximately three thousand, largely Jewish refugees at a time when doing so was severely punishable. What to us seems
like grave risk-taking to protect strangers from Nazi capture came to be
simply what was expected. As Lawrence Blum observes, knowing that
many others were involved in aiding the refugees had a double effect:
it made the worthwhileness of taking the risk to help more obvious and
unambiguous, and it reshaped the villagers sense of undue burden,
making it motivationally easier to choose to take those risks. 25

V. CONCLUSION
This, now, is what we might say to Scrooge: You take yourself to be a
minimally well-formed moral agent. Indeed, you pride yourself on
paying your debts and exacting the debts from others that they owe you.
But you have misconceived what it means to be a minimally well-formed
25. Ibid.

121

MOR A L A IMS

moral agent. If you really had the basic competences to practice morality with others, including caring about others well-being and agency,
you would at least elect those supererogatory acts that are motivationally nontaxing and obviously and unambiguously desirable. Being pleasant to your nephew, giving your employees Christmas off, and showing
some mercy to your most destitute debtors should have been obvious,
unambiguous, and easy moral gifts for you to give because they are conventional practices in your social world. In refusing to give those gifts,
you show yourself to be a shamefully inadequate moral agenta being
without common decency.

122

C ha pt e r5

Standing for Something

We admire and trust those who have integrity, take pride in our own, rue
its absence in politics, and regret our own failures to act with integrity.
Clearly, integrity is a virtue, but it is less clear what it is a virtue of or why
we might prize it.
Three pictures of integrity have gained philosophical currency, particularly through the work of Bernard Williams, Gabriel Taylor, Lynne
McFall, and Jeffrey Blustein.1 I will call these the integrated-self, identity,
and clean hands pictures of integrity. On the integrated-self view, integrity involves the integration of parts of oneselfdesires, evaluations,
commitmentsinto a whole. On the identity view, integrity means fidelity to those projects and principles that are constitutive of ones core
identity. On the clean-hands view, integrity means maintaining the
purity of ones own agency, especially in dirty-hands situations.
I am going to sketch out each of these pictures of integrity and suggest two general criticisms. First, each ultimately reduces integrity to
something else with which it is not equivalentto the conditions of
unified agency, to the conditions for continuing as the same self, and to
the conditions for having reason to refuse cooperating with some evils.
Second, all three accounts are of integrity as a personal, but not also a
1. Bernard Williams, Persons, Character, and Morality and Moral Luck, in Moral Luck:
Philosophical Papers 19731980 (Cambridge: Cambridge University Press, 1981), and
A Critique of Utilitarianism, in J. J. C. Smart and Bernard Williams, Utilitarianism:
For and Against (Cambridge: Cambridge University Press, 1973); Gabriel Taylor, Integrity, in Pride, Shame, and Guilt: Emotions of Self-Assessment (Oxford: Clarendon,
1985); Lynne McFall, Integrity, Ethics 98 (1987): 520; Jeffrey Blustein, Care and
Commitment: Taking the Personal Point of View (New York: Oxford University Press,
1991). Although what I refer to as three pictures of integrity are analytically distinct,
these authors work with them as components of a complex account of integrity.

123

MOR A L A IMS

social virtue. This limits the analysis both of what integrity is and of why
it is a virtue. In the last section, I will suggest a way of understanding
integrity as a social virtue.

I. THE INTEGR ATED-SELF PICTUR E

OFINTEGR IT Y
Etymologically, integrity is related to integer, a whole number, and
to integration, the unification of parts into a whole. The integratedself picture of integrity begins from this etymological observation,
and the resulting description of the person of integrity as a whole integrated self owes a good deal to Harry Frankfurts work on freedom and
responsibility. 2
On this view, the integration of the self, and hence integrity, requires first that one not be a wanton. Frankfurt imagines wantons
to be individuals who either lack the capacity or simply fail to deliberate and make up their minds about which of their desires they want to
be volitionally effective. As a result, wantons act on whichever desire
happens to be psychologically strongest at the moment. Because the
wanton is passive in relation to what moves him, Frankfurt concludes
that the wantons desires are, in an important sense, not his and, as a
result, neither are his actions. Such a being lacks integrity altogether.
He does not, in Frankfurts view, have a self, because it is only by endorsing a particular desire that an agent claims it as his own and thereby
constitutes his self. 3
Integrity, however, requires a good deal more than simple nonwantonness with respect to ones first-order desires. First, both weakness of
will and self-deception undermine the individuals ability to act on her
actual or professed endorsements. The weak-willed person ends up not
having the will he wants, but one that is imposed upon him by a force
2. Harry Frankfurt, Freedom of the Will and the Concept of a Person, Journal of Philosophy 68 (1971): 520; and Identification and Wholeheartedness, in Responsibility,
Character, and the Emotions: New Essays in Moral Psychology, ed. Ferdinand Schoeman
(Cambridge: Cambridge University Press, 1987).
3. Frankfurt, Identification and Wholeheartedness, 38.

124

STA N DI NG FOR SOM ETHI NG

with which he does not identify and which is in that sense external to
him.4 The self-deceived person is unable to see what actually motivates
her. She thinks it is one thing (for instance, cautiousness) when in fact it
is something else (cowardice). As a result, the will she has is not the one
she claims to want. In both cases, what the agent does is not integrated
with what she endorses or claims to endorse.
Second, in a variety of ways, wantonness can infect ones
endorsementsthat is, ones second-order desires. Thus, even individuals who reflect on the sort of person they want to be may fail to do so in
an adequately self-constituting way. As Gabriel Taylor argues, how one
comes to endorse a first-order desire matters. If a person adopts values
only because her group does, without having any reasons of her own for
thinking that these are the right values, then her second-order volitions
will not really be her own. [S]he has to find out from others which desires to identify with, or indeed what sorts of desires she should have.5
In addition, as Taylor also observes, unless the individual regards her
endorsements as prima facie committing her to making the same endorsements on future occasions, she will be no more than shallowly
sincere, wholeheartedly identifying with one set of desires today and a
different set tomorrow.6 Both the crowd follower and the shallowly sincere exhibit second-order wantonness and a lack of integrity.7 Such wantonness appears avoidable, and integrity achievable, only if a persons
endorsements are determined by her own practical reasoning.
Frankfurt raises a further possibility that reflective individuals may
fail to identify wholeheartedly with their volitions. They may have inconsistent second-order desires. Or alternatively, they may be ambivalent about whether they want to identify with a particular desire. Both
inconsistency and ambivalence result in there being no unequivocal
answer to the question of what the person really wants.8 The individual
4. Ibid., 33.
5. Taylor, Integrity, 116.
6. Ibid.,113.
7. Gary Watson, Free Agency, Journal of Philosophy 72 (1975): 205220, raised the problem of wantonness in higher-order volitions. Frankfurt addresses himself specifically to
Watsons critique.
8. Ibid.

125

MOR A L A IMS

cannot wholeheartedly say I will because there is no unified self to

back the willing. She lacks integrity. Wholeheartedness, and with it
integrity, would require integrating competing desires into a single ordering, as well as separating some desires from the self and relegating
them to outlaw status. It is these acts of ordering and of rejection
integration and separationthat create a self out of the raw materials of
inner life.9
This picture of integrity has intuitive appeal. It captures our sense
that people with integrity decide what they stand for and have their own
settled reasons for taking the stands they do. They are not wantons or
crowd followers or shallowly sincere. Nor are they so weak willed or selfdeceived that they cannot act on what they stand for. The actions of persons of integrity express a clearly defined identity as an evaluating agent.
One might, however, wonder whether integrity is nothing but a
matter of self-integration. On the integrated-self picture, any person
whose actions are fully determined by her own endorsements has integrity. But consider Thomas E. Hill Jr.s example of an artist who
lacks self-respect and, it seems, lacks integrity as well: Suppose an
artist of genius and originality paints a masterwork unappreciated by
his contemporaries. Cynically, for money and social status, he alters
the painting to please the tasteless public and then turns out copies in
machine-like fashion. He does it deliberately, with full awareness of his
reasons.10 His pandering to public opinion, silencing his own aesthetic
judgments, and selling out his standards for material gain reveal a lack
of integrity. Yet there seems no reason to think that he does not fully determine his actions. He does, but without integrity. Integrity, one might
intuitively think, involves not subordinating ones own judgment about
what makes art worthy of being produced and appreciated to considerations of personal comfort, gain, status, and expediency. (In the final
section, I will suggest why this is so.)
One might also wonder if what Frankfurt calls wholeheartednessthe consistency of and nonambivalence about ones various
9. Ibid., 39.
10. Thomas E. Hill Jr., Self-Respect Reconsidered, in Autonomy and Self-Respect (Cambridge: Cambridge University Press, 1991), 19.

126

STA N DI NG FOR SOM ETHI NG

endorsementsis really a necessary condition for having integrity.

Wholeheartedness might instead be an ideal of unified agency. That is,
as agents, we might wish we could be wholehearted about what we do.
But being of two minds might not make what we do any less ours and
thus might not pose any special threat to integrity. Because the notion
of wholeheartedness regularly occupies a central place in philosophical accounts of integrity, it is worth probing whether it should.11 Taking
inconsistency and ambivalence in turn, I will sketch out two examples
that suggest that integrity may sometimes in fact require resisting the
impulse to resolve inconsistencies and ambivalence.

Inconsistency
Maria Lugones has repeatedly argued for the value of conceptualizing
oneself as a duplicitous or multiplicitous being whose identity is differently constituted in different cultural worlds or meaning systems.12
The identity Latina, for example, is differently constituted in Hispanic
and in racist Anglo cultures. Racist oppression consists, in part, in the
suppression of the Hispanic cultural understanding of what it means to
be Latina. And thus for Lugones, struggling against racist oppression
partly consists in endorsing and affirming her identity as a Latina as it is
constituted within Hispanic culture. Many people, however, confront
multiple oppressions. Lugones, for instance, is both Latina and lesbian.
In struggling against multiple oppressions, she is faced with the task of
affirming not only her Latina identity as it is constituted within Hispanic
culture but also her lesbian identity as it is constituted within nonheterosexist lesbian communities. But the meaning and value systems (for
example, concerning gender, sexuality, and family) that make those two
identities possible are in conflict. Within Hispanic culture, lesbianism
is an abomination. Within the lesbian community, Hispanic values and
11. Taylor, McFall, and Blustein all take wholeheartedness to be central to integrity.
12. Maria Lugones, Playfulness, World-Traveling, and Loving Perception, Hypatia 2
(1987): 319. See also Maria Lugones, On the Logic of Pluralist Feminism, in Feminist Ethics, ed. Claudia Card (Lawrence: University Press of Kansas, 1991), 3544, and
Hispaneando y Lesbiando: On Sarah Hoaglands Lesbian Ethics, Hypatia 5 (1990):
138146.

127

MOR A L A IMS

ways of living do not have central value. As a result, Latina lesbian is

not a coherent identity nor is there a single, unified conceptual and normative perspective that could count as the Latina lesbian perspective,
and thus no single perspective from which to take issue with both racist
and heterosexist oppression.13 I do not know, she writes, whether the
two possibilities can ever be integrated so that I can become, at least in
these respects, a unitary being. I dont even know whether that would be
desirable. But it seems clear to me that each possibility need not exclude
the other so long as I am not a unitary but a multiplicitous being.14
What Lugoness case illustrates is that lack of wholeheartedness
does not necessarily signal some personal failure on the part of the agent
to make up her mind what she really wants. Agents can have reasons
to resist resolving value conflicts. In Lugoness case, taking a stand
against oppressionssomething a person with integrity might well
doinvolves endorsing and struggling to preserve meaning and value
systems that conflict with each other. To insist that, even in these cases,
integrity requires wholeheartedness would be to make practical deliberation over whether a value conflict ought to be resolved oddly irrelevant
to integrity.

Ambivalence
A similar point may be made about ambivalence. In his autobiography Cures: A Gay Mans Odyssey, Martin Duberman describes his

13. The point here is not that one could not construct a unified identity and conceptualnormative perspective. The point is that such a unified identity would be neither Latina
nor lesbian, and endorsing it would be inconsistent with giving priority to combating
racist and heterosexist oppression that consists, in part, precisely in the suppression of
Hispanic and lesbian identities as they are constructed in their home cultures. The
point is also not that one cannot be critical of the identity and culture one endorses
for example, that one cannot be critical of heterosexism in ones Hispanic community.
The point is that the criticism must be internal; it must take place on the background
assumption that certain conceptions and evaluations of gender, sexuality, and family
that are constitutive of Hispanic culture have weight. To engage in external criticism
of Hispanic culture (say, from the point of view of the lesbian community) would be to
dismiss the significance of that culture from the outset.
14. Lugones, Hispaneando y Lesbiando, 138139.

128

STA N DI NG FOR SOM ETHI NG

ambivalence about his therapists suggestion that they team-teach a

seminar at Princeton where Duberman was a history professor. The
therapist, Karl, claims that team teaching will help cure Dubermans
homosexuality by allowing him to work closely with a caring male. Because your father was so distant, the therapist tells him, you cannot
believe to this day that an adult male could care about youand indeed
thats the main reason you pursue males sexually, and especially unavailable males like hustlers: its a way of belatedly trying to get your
fathers love while simultaneously confirming that you cant.15 Duberman, however, suspects that the team-teaching idea has more to do with
his therapists ego than with therapy. In response to Karls suggestion,
he says, I could feel myself stiffen with distrust. And then, two seconds
later, with self-distrust, as I instantly questioned whether my suspicion
about Karls motives wasnt precisely the reflexive skepticism about an
older mans kindly interest in me that we had just finished analyzing.16
Caught between his own suspicions and his therapists authoritative
judgment, Duberman is faced with the choices of dismissing his therapists judgment in favor of his own, or of acceding to his therapists
judgment and silencing his suspicions, or of remaining in a state of
ambivalence.
One might think that, as a person with integrity, Duberman
should have stood up for his own suspicions. Indeed, one might generally think that whenever ones own and others interpretations of
ones motives conflict, one ought to resolve that conflict in favor of
ones own judgment. The integrated-self picture of integrity suggests
just this conclusion. Feminists have also tended toward this view.17
Recognizing that ambivalence is generally endemic among members
of oppressed groups who suspect that dominant interpretations of
their motives and actions are mistaken, but for whom there are as
yet no clearly articulated arguments discrediting dominant views,
15. Martin Duberman, Cures: A Gay Mans Odyssey (New York: Plume, 1992), 139.
16. Ibid., 140.
17. See, for example, Sarah Lucia Hoagland, Lesbian Ethics: Toward New Value (Palo Alto,
CA: Institute of Lesbian Studies, 1988); and Kathryn Morgan, Women and Moral
Madness, in Science, Morality, and Feminist Theory, ed. Marsha Hanen and Kai Nielsen
(Calgary, Alberta: University of Calgary Press, 1987).

129

MOR A L A IMS

feminists have regarded such socially produced ambivalence as destructive of integrity. For reasons that will become clearer in the last
section, I am unpersuaded that this is so. Anyone who regards herself
as an equal in autonomous judgment to others cannot be indifferent
to what others think. When ones own and others judgments come
into serious conflict, ambivalence may be a way of acknowledging
that equality. Ambivalence does not necessarily signal a failure on
the agents part to make up his mind about what he really believes
and wants. Agents can have reasons to resist resolving ambivalence.
In particular, they may think it important to acknowledge a basic assumption underlying practical deliberationnamely the equality of
deliberators.
In sum, the integrated-self picture of integrity, though outlining
some important, necessary conditions of integrity (for example, not
being a mere crowd follower), reduces integrity to volitional unity.
As a result, it obscures the fact that persons can have reason to resist
resolving conf licting commitments and ambivalence about their
own desires, and thus that resisting wholeheartedness may sustain
integrity rather than be symptomatic of its absence.18 In addition,
the integrated-self picture of integrity places no restrictions on the
kinds of reasons that can motivate persons with integrity. But simply
acting on ones own reasons seems insufficient for integrity. Some
sorts of reasons seem incompatible with integrityfor instance, a
primary concern with ones own comfort, material gain, pleasure,
and the like at the expense of ones own judgments about what is
worth doing.
18. I think it is important to be skeptical about any account of integrity whose implication
is that members of oppressed groups are particularly likely not to have integrity or that,
for them, acting with integrity requires acting in morally unsavory ways (for example,
ignoring all but one oppressive system or adopting a dismissive stance toward social
judgments). Lugoness and Dubermans cases suggest that achieving the ideal of an integrated self does not depend solely on an agents internal capacities. It also depends
on social conditions. The illusion that integration is entirely up to the individual may
reflect a particularly privileged social positionfor example, one from which the question of where one stands with respect to multiple and conflicting oppressions does not
regularly come up and within which ones own self-interpretation receives substantial
social confirmation.

130

STA N DI NG FOR SOM ETHI NG

II. THE IDENTIT Y PICTUR E OF INTEGR IT Y

A second picture of integrity owes a good deal to Bernard Williamss
work.19 On this view, integrity is a matter of having a character and being
true to it. To have a character, as Williams sees it, is to have some ground
projects with which one is so strongly identified that in their absence one
would not be able to find meaning in ones life or have a reason for going
on. Because both Kantianism and utilitarianism require that agents be
prepared to give up their ground projects in the name of impartial good
ordering or the maximization of good states of affairs, both moral systems are, in his view, hostile to agents integrity.
Picturing integrity as fidelity to projects that the individual
deeply identifies with has intuitive appeal. It captures in a way that the
integrated-self picture did not the idea that persons with integrity stand
for something. On the integrated-self picture, a person stands for all of
the desires that she does not regard as alien or outlaw forces, no matter
how trivial those desires might be. Thus, ones integrity is implicated in
everything one does. The identity picture, by contrast, discriminates between desires that are basic to ones sense of self and those that are not.
A person with integrity stands for those desires that are constitutive of
her core self. This explains why such persons might prefer death to the
betrayal of what they stand for.
Although Williams was explicitly concerned with integrity, his
discussions of integrity all occur within the context of formulating objections to Kantian impartiality and the utilitarian conception of negative responsibility. He was, in particular, concerned with securing a
space for individuals partiality to their personal, identity-constituting
projects against the seemingly relentless demands of morality. A central part of his argument was that individuals will not have a reason to
care about their own future, including their future in a morality system,
unless they have some ground projects whose pursuit propels them
into the future.20 Even if Williams was right to insist that Kantian and
utilitarian morality demand too much of agents, one can still question
19. Williams, Persons, Character, and Morality, Moral Luck, and Integrity.
20. Williams, Persons, Character, and Morality, esp.14.

131

MOR A L A IMS

whether integrity really is, and is nothing but, being true to what one
deeply identifies with.

Identity Without Integrity

Those who endorse the identity picture of integrity admit that, on this
view, one might have integrity even though ones identity- conferring
projects are nonmoral, or even morally despicable. This is because
deeply identifying with what one does puts ones integrity beyond
question. The Gauguin portrayed by Williams, for example, stakes his
deepest sense of self on his desire to realize his painterly gifts. Gauguin is not pictured as thinking that he will have earned his place in
the world, if his project is affirmed: that a distinctive contribution to
the world will have been made, if his distinctive project is carried forward. The point is that he wants these things, finds his life bound up
with them, and that they propel him forward, and thus they give him a
reason for living his life. 21 Although taking his moral obligations seriously, this Gauguin does not regard them as identity-conferring in the
deepest sense. Morality, for him, is not a ground project. Thus, when
moral obligation conflicts with his deep identity as a painter, preserving his integrity requires that he betray his moral commitments.
Agreeing that integrity can take nonmoral forms, must we also
agree that Gauguin acts with integrity just because he so deeply identifies with painting? This, I think, depends on what we mean by identity
and identification. It is possible, first, to understand identity as a psychological phenomenon. From a psychological point of view we might
understand who we are in terms of our deepest impulses and what feels
natural or unforced. Identifying with a desire would not, in this case,
entail that the agent also endorses the desire she identifies with. If we
have any reason to doubt Gauguins integrity, it is because we suspect
that identifying with a project may differ from endorsing it and that
Gauguins reason for pursuing his painterly project is his identification with it, not his endorsement. To clarify this distinction between
21. Ibid., 1415. Williamss specific discussion of Gauguin is in Williams, Moral Luck.

132

STA N DI NG FOR SOM ETHI NG

psychological identification and endorsement, I draw once again on an

example from Dubermans Cures.
During his two-decades-long pursuit of a psychotherapeutic cure for
his homosexuality, Duberman accepted the then dominant view of homosexuality as a neurotic and pathological barrier to a loving, committed relationship. Making what was called a heterosexual adjustment
was, he thought, his only hope for a healthy, happy life. Repeatedly entering therapy for a cure, he just as repeatedly quit, being both unwilling
to follow his therapists injunction to stop acting out his homosexuality and convinced that he could not change. He vacillated between
terminating relationships for enforced celibacy and arranging his life to
accommodate frequent trips to New York gay bars.
His refusal to endorse his desire for men seems clear from the narrative. He says, Accepting it [namely the decision to quit therapy] means
accepting my life, being satisfied with it. And I cant.... 22 But it seemed
equally clear both to himself and to his aggravated therapists that he did
not identify with the therapeutic goals he endorsed. He was in his words
an onlooker, an auditor, rather than a participant in the therapeutic
process.23
Cases like Dubermans, where identification and endorsement part
company, force us to get clearer about what we mean by a ground project or identity-conferring commitment. If such desires and commitments are simply ones that are connected to the individuals deepest
psychological impulses, then they would not necessarily be endorsed.
One simply does, as a matter of psychological fact, care deeply about a
particular project. Williams sometimes speaks this way. In his words,
a person who has a ground project simply finds his life bound up with
it. Understood this way, there is no reason to suppose that what one
psychologically identifies with is necessarily also what one endorses
and what makes ones life meaningful and worth living. Thus there is
no reason to suppose that losing such identity-conferring projects necessarily poses any special threat to integrity. In trying to cure himself
of what he found his life bound up with, Duberman assumed that he
22. Duberman, Cures, 36.
23. Ibid.

133

MOR A L A IMS

was acting with integrity, not undermining his integrity. And insofar as
we imagine that Gauguin, in pursuing what he found his life bound up
with, acted merely on a psychologically deep impulse without critically
reflecting on the value of doing so, we may suspect him of not acting
with integrity.
In short, integrity involves fidelity to ones endorsements, not
merely to psychologically deep identifications. Although it may happily
be true of many of us that we want to be who we arethat endorsement
and psychological identification coincidethis is not inevitable. One
may deeply identify oneself with some nonendorsed desires, and living
up to ones endorsements can exact a terrible toll on psychological identity. When endorsement and identification conflict, the price of trying
to become a self we take to be better is not our integrity.

Integrity Beyond Identity-Conferring Commitments

One might try to preserve the basic idea that integrity is connected to
identity and fidelity to self by shifting to a deliberative notion of identity. From a deliberative point of view, we might understand who we
are in terms of our considered judgments about what is of value, what
principles ought to be endorsed, and how they should be hierarchically
ranked.24 Thinking of identity this latter way, Gabriel Taylor observes
that some of a persons evaluations concern trivial matters and do not
contribute to her identity. 25 Those that do contribute to identity are
more properly described as identity-conferring commitments. Such
24. One might think that Williams meant to connect ground projects to this deliberative
notion of identity and to an agents deepest endorsements. But understood this way,
there is no reason to suppose that ground projects could conflict with the agents own
view of what morality demands. If Gauguin endorses painting as his ground project
that is, as what has evaluative prioritythen he has already answered for himself the
question of what morality demands. It does not, in his view, demand eliminating a
space for partiality to ones own projects. Gauguins endorsement of his painterly project, reflecting as it does an antecedent rejection of utilitarian value maximization and
Kantian impartiality, cannot then be offered, without begging the question, as a reason
for thinking that utilitarianism and Kantianism are mistaken. Nor can it be offered as a
reason for thinking that either morality system poses a threat to the agents integrity.
25. Taylor, Integrity, 131.

134

STA N DI NG FOR SOM ETHI NG

commitments, in Lynn McFalls words, reflect what we take to be most

important and so determine, to a large extent, our (moral) identities, as
well as what we can do and survive as the persons we are. 26
This idea that integrity requires fidelity to our core values sounds
right. But one might question whether integrity is just a matter of being
true to (and unself-deceived about) identity-conferring commitments.
If integrity is just a matter of standing on principles or values that are
central to ones identity, it would follow that betraying or being selfdeceived about principles or values that are more peripheral to ones
sense of self would not cost a person her integrity. This is precisely the
conclusion Jeffrey Blustein draws. He says, Not every instance of weakness of will, of acting contrary to ones better judgment, and not even
repeated akratic failure, necessarily indicates a lack of integrity. There
must be a deficiency in self-control with respect to commitments or
principles that have some bearing on the agents broad conception of his
or her lifes direction or sense of self-identity. 27 He draws a parallel conclusion about self-deception. 28 It would seem, then, that on matters that
are not strongly connected to ones sense of self-identity, one cannot act
without integrity. But this does not seem right. We recognize persons
with integrity not only by their willingness to incur great losses for the
sake of what they hold most dear but also by their conscientiousness
in smaller matters having no strong bearing on the agents broad conception of his or her lifes direction. We expect persons of integrity not
only to stand up for their most deeply held and highly endorsed commitments but also to treat all their endorsements as ones worthy of being
held by a reflective agent.29

26. McFall, Integrity, 13.

27. Blustein, Care and Commitment, 100.
28. Ibid., 106.
29. What does seem right about Blusteins position is that some self-deception and weakness of will is compatible with an all things considered assessment of a persons character. In answering the question, Is this the kind of person that, all things considered,
we would describe as having integrity? it is most relevant to look at how a person
stands with respect to her core commitments. A person might be pervasively weak
willed with respect to very low-order principles or to the application of core principles
in fairly trivial cases, but she might exhibit great strength of will and courage in sticking

135

MOR A L A IMS

In sum, the identity picture of integrity equates the conditions

under which we can go on as the same self with the conditions for integrity. But acting on the deep impulses that define our psychological
sense of self seems to have little to do with integrity, given that agents
may repudiate their deepest impulses. Acting on those deeply held and
highly endorsed commitments that define our sense of self, though constituting part of what it means to act with integrity, does not appear to
constitute the whole of it.

III. THE CLEA N-H A NDS PICTUR E

OFINTEGR IT Y
Running throughout both pictures of integrity presented so far is the
thought that integrity is importantly connected to an agents endorsements. The clean-hands picture offers a different take on this same
theme. On this picture, integrity is a matter of endorsing and, should
the occasion arise, standing on some bottom-line principles that define
what the agent is willing to have done through her agency and thus the
limits beyond which she will not cooperate with evil. A person has integrity when there are some things she will not do regardless of the
consequences of this refusal. In bottom-line situations, she places the
importance of principle and the purity of her own agency above consequentialist concerns.
Williams has also been a key advocate of this conception of integrity, although philosophical discussions of dirty hands and choosing the
lesser of two evils generally square off standing on principle and integrity against compromising with evil to secure a better outcome. Like the
to her convictions when core principles or more serious cases are at stake. If so, we
might well be prepared to say that, all things considered, she has integrity. It does not
follow, however, from the fact that a person lacks integrity, all things considered, only
if she is weak willed or self-deceptive about basic goals and concerns that a person
acts without integrity only if she acts contrary to her basic goals and concerns. People
can act without integrityor courage, strength of will, temperance, kindness, and so
onon particular occasions while still being, all things considered, persons who have
integrity, or who are courageous, strong willed, temperate, kind, and so on.

136

STA N DI NG FOR SOM ETHI NG

other pictures of integrity, this one, too, has intuitive appeal. It captures,
in a way the identity picture does not, the kind of thinking we expect
behind principled refusals: not I couldnt go on as the same person if
I did this but I would be doing a wrong. It also captures better than
the identity picture what it means to stand for something. Standing for
something is not just a matter of personal identification with certain
values; it is also a matter of insisting on the endorsability of those values.
Like the other two pictures of integrity, this one, too, emerges within
a larger philosophical context. Williams was interested in challenging
what he took to be two tenets of utilitarianism: (1) we are just as responsible for preventing others from doing evil as we are for refraining from
evil ourselves, thus agents must be prepared to dirty their hands and
perform morally repugnant deeds if doing so will prevent others from
committing even worse deeds; and (2) so long as we maximize beneficial outcomes we have no reason to feel regret, guilt, shame, and the like
no matter what we have had to do to maximize outcomes. 30 Both tenets,
in Williamss view, are incompatible with agents maintaining a sense of
their own moral integrity. To have integrity is to view some actions as
morally disagreeable apart from their consequences and to reflect that
view in ones actions and sentiments. Thus persons with integrity will
sometimes refuse to maximize good consequences when this means
doing something morally disagreeable. They will also regret doing morally disagreeable acts on those occasions when circumstances require
doing a lesser evil in order to prevent a greater one.
I want to come at the criticism of the clean-hands picture of integrity via a more indirect route than I took with the preceding pictures.
Specifically, I want to begin by examining this thought that some moral
theories are more hospitable to acting with integrity than others. At
one end of the spectrum of moral theories is strict consequentialism.
Here, standing on principle when one could instead make the best of a
bad situation would never be justified; and so, the reasoning goes, consequentialism accords integrity little or no moral value. On the other
end of the spectrum is utopian deontology, where being morally justified hinges on acting on those principles that would be acceptable in an
30. Williams, Integrity.

137

MOR A L A IMS

ideal moral world. Here, standing on principle is de rigueur no matter

how dreadful the consequences of doing so. (Recall, for example, Kants
insistence on dealing truthfully with the murderer at ones door.) Thus
one would always be justified in refusing, on principle, to cooperate
with evil; and so, the reasoning goes, utopian deontology makes integrity a supreme value. In the middle are various moderate positions that
accord both principles and consequences justifying weight. A moderate
position might sometimes require cooperating with evil and sometimes
require standing on principle, depending on what the lesser evil is. 31 Alternatively, or in addition, a moderate view might regard standing on
principle as a permissible but not a required option. 32 Moderate moral
theories, it might be thought, place some, but not supreme, value on
acting with integrity. Thus they sometimes recommend acting with integrity and sometimes recommend compromising ones integrity.
Because some of the more striking examples of acting with integrity involve refusing to compromise ones principles, it is indeed
tempting to think that advocating a particular theory of moral justification entails placing a higher or lower value on acting with integrity,
depending on how much justificatory weight is put on deontological
principles versus consequences. But that temptation should be resisted.
It does not follow from the fact that persons of integrity act on principle, and the fact that deontological theories recommend acting on
principle, that deontological theories are integrity-friendly theories.
Consideration of the preceding two pictures of integrity has suggested
that acting with integrity involves acting on ones own principles. However, when a theory of justification recommends acting on principle,
it is not recommending that people act on their own principles. It is recommending that people act on the right principles. Thus a deontological theory may sanction acting on principle without sanctioning the
agents acting on her own principlesthat is, without sanctioning her
acting with integrity.
31. Terrance C. McConnell, Moral Blackmail, Ethics 91 (1981): 544567, works out such
a moderate position.
32. Thomas E. Hill Jr. argues for this on Kantian grounds in Moral Purity and the Lesser
Evil, in Autonomy and Self-Respect (Cambridge: Cambridge University Press, 1991).

138

STA N DI NG FOR SOM ETHI NG

In short, a theory of moral justification places value on having good

reasons for action. If it also places value on acting on principle, it does so
only insofar as the principle supplies a good reason. By contrast, to value
integrity is to place value on an agents acting from her reasons, whether
they are good ones or not. 33 This means that no theory of moral justification is inherently hospitable to integrity. Both deontology and utilitarianism may recommend courses of action that conflict with the agents
own principles. Both deontology and utilitarianism only contingently
sanction acting with integrity. That sanction depends on the agents first
endorsing the moral theory, thereby making theoretically good reasons
also the agents own reasons.
Now, even if one gives up the idea that utilitarianism is uniquely
unfriendly to integrity, one might still think there is something to the
clean-hands picture of integrity. Integrity, one might think, requires
having at least some nonconsequentialist principles to stand on, even
if they are the wrong ones, and thus reason sometimes to regret cooperating with evil. The consequentialist has no such principles. There
is nothing she would not do to optimize consequences. Thus even if
she is justified in repeatedly dirtying her hands to fix a bad world, she
cannot claim to have integrity. But this seems wrong. Although there
is nothing she would not do to optimize consequences, there are things
the utilitarian would not donamely nonoptimific acts. On the old
Star Trek series, for instance, Mr. Spock was portrayed as a diehard
consequentialist on life-and-death issues, always ready to sacrifice the
few for the many; and he was also portrayed as a person of impeccable
integrity, willing to be one of the sacrificed few and unwilling to compromise his utilitarian principle in the face of his crewmates insistence
on the wrongness both of letting the numbers count and of cooperating
with evil.
33. A person can be absolutely mistaken about which principles one may justifiably stand
on yet act with integrity in taking the stand. When Dan Quayle stood on his pro-life
principle, refusing to sanction an abortion for his own young daughter and even for a
twelve-year-old raped by her father, one might have thought him hopelessly misguided;
but that thought alone would not have been reason to think he lacked integrity. (If one
suspected him of lacking integrity, it was because it is hard to imagine he really believed
what he said.)

139

MOR A L A IMS

In sum, given that a person believes an act is wrong apart from its
consequences, having integrity may indeed require that she not do it, or
at least regret doing it. But integrity does not require believing that there
are such consequence-independent wrongs. 34 The only necessary condition of moral integrity is that one do what one takes oneself to have most
moral reason to do. For consequentialists, that will mean cooperating
with evil. For nonconsequentialists, it will mean not cooperating or regretfully cooperating with evil.

Selling out
Underlying the clean-hands picture of integrity, I suspect, are often
the ideas that (1) there is a right course to take when presented with
the choice between two evils, or the option of compromising with opponents, or the choice between protesting and remaining silent about
injustice; (2) rightness is not fully determined by consequences; and (3)
having integrity just is a matter of taking the right course. 35 Thus the
person without integrity is the one who cooperates with evil or compromises with opponents when she ought not, or who fails to protest
when she should. There is something to this last statement, though not
what the equation between getting it right and having integrity suggests. I have argued that integrity hinges on acting on ones own views,
not the right views (as those might be determined independently of the

34. This is not to say that the thought that a person is morally mistaken has no bearing on
the question of his integrity. Sometimes it is hard to imagine how someone could care
about what principles they act on, be unself-deceived, sincere, critically reflective, nonhypocritical, concerned with more than their own comfort, and get things morally so
wrong.
35. This would have to be Williamss view if utilitarianism is going to be singled out as the
enemy of integrity. From a deontological point of view, utilitarianism requires agents
to do the wrong thing. Thus if having integrity is equivalent to doing the right thing,
utilitarianism will (again, from a deontological point of view) require that agents act
without integrity. What Williams could have been pointing to was the fact that utilitarianism makes external demands on agents. If integrity is a matter not of doing the
right thing but of acting on ones own (internal) views, then utilitarianism would again
be an enemy of integrity. But in this case, deontology would have to be depicted as
equally inimical to integrity, because it too makes external demands on the agent.

140

STA N DI NG FOR SOM ETHI NG

agents own opinion). If people without integrity do indeed cooperate,

compromise, and remain silent when they ought not, the force of ought
not cannot be the wrong thing as determined by some (deontological)
moral theory. Rather, people without integrity violate their own views.
They cooperate with evil, compromise with opponents, and remain
silent when their own principles and values tell them they ought not.
If this is so, how does lacking integrity differ from weakness of will?
Surely not all weak-willed failures to act on ones own best judgment
signal lack of integrity. Breaking a diet privately embarked on because
one is lazy, or craving sugar, or just plain hungry is weak willed, but not
necessarily a cost to integrity, especially if the person reproaches himself for his weakness. Self-reproach is exactly what one expects of the
person of integrity who lets himself down.
To lack integrity, I suggest, is to underrate both formulating and exemplifying ones own views. People without integrity trade action on
their own views too cheaply for gain, status, reward, and approval, or for
escape from penalties, loss of status, and disapproval (as did the artist
who cynically altered his work for gain). Or they trade their own views
too readily for the views of others who are more authoritative, more in
step with public opinion, less demanding of themselves, and so on. 36 The
person who allows himself to be cajoled, bullied, bribed, or embarrassed
into breaking a diet he endorses, or who rationalizes his failure with the
thought that most people have lower standards of fitness that would not
have required dieting in the first place, is a prime example of a person
without integrity. Integrity becomes an issuesomething that one
risks losing and must act to preserveparticularly in contexts where
there is some incentive to act on someone elses best judgment. Williamss
well-known example of George illustrates the point. 37
36. In thinking about integrity in terms of the value a person places on her own views,
I have been influenced by Thomas E. Hill Jr.s analysis of self-respect (Servility and
Self-Respect, in Autonomy and Self-Respect [Cambridge: Cambridge University Press,
1991], 418) as, in part, a matter of being unwilling to trade ones own rights cheaply.
I read his Self-Respect Reconsidered as in fact a discussion of integrity, because the
issue there is not the value a person attaches to her own rights but the value she attaches
to having and acting on views of her own.
37. Williams, Integrity.

141

MOR A L A IMS

George, an opponent of chemical and biological warfare, is offered

a chemical-biological warfare research job by a utilitarian who urges
George to take it, thereby preventing a more zealous researcher from
doing so. George thinks that he should refuse on principle to participate
in this research, regardless of the consequences. As Williams constructs
the case, utilitarianism makes an external demand on agents to abandon their convictions that some acts are wrong apart from their consequences. However, any morality system, utilitarian or not, if personified
and figured as a kind of stern, moralistic father who demands ones compliance with a view not ones own, will pose a threat to integrity. Agents
may give in to the demand, abandoning their own judgment and acting
without integrity. As Blustein correctly points out, this has nothing
particularly to do with the content of the demand that the utilitarian is
making of this person.38 It has everything to do with abandoning ones
own judgment for anothers. The more authoritative or more coercive the
external demand that one do x rather than the y one thinks one ought to
do, the more intense the integrity question becomesnamely the question of whether one will act on ones own or on an external judgment.
Also central to the case is the fact that others will have strong and
reasonable grounds for reproaching George if he refuses the job. Both
the utilitarian employer and the pragmatically minded opponents of
chemical and biological warfare will think he has done the wrong thing.
His wife, too, may reproach him for taking a principled stance that does
not give concern for her welfare high priority. To all of these reproaches
he will have little to offer but the thought I did what I thought right.
The greater the risk of being held to accountreproached, condemned,
penalizedby others for acting on ones own judgment, the more central becomes the question of whose judgment to make ones guide. That
is, ones integrity becomes the issue.
Finally, central to the case is the tension between what the world
as it is presently structured may require and what an ideal world would
require. In an ideal world, some things ought never to happen and some
acts no one should ever be called upon to do. In Georges view, chemical and biological weapons have no place in an ideal world, and no one
38. Blustein, Care and Commitment, 70.

142

STA N DI NG FOR SOM ETHI NG

should ever be called upon, as he is now, to advance their development

in order to prevent a greater evil. One does not have to be a deontologist to appreciate that fact. The more deeply entrenched the views, and
the more pervasive the actions which produce a nonideal world, the
more intense the integrity question becomesnamely the question of
whether to accede to others construction of the world by acting as best
one can in present circumstances, or to act on ones own judgment that
the world is a bad one and calls upon people to do what no one should be
called upon to do. 39
In sum, in contrasting acting on principle to maximizing outcomes,
the clean-hands picture of integrity mislocates the heart of the integrity question. It is not consequentialism that threatens integrity, but our
own vulnerability to other peopletheir bribes and threats, authoritative demands, reproaches and accusations of unreasonableness, their
lower standards that make it easy to get away with violating our own,
and their collective construction of a world that calls upon us to act
against our ideals. We find ourselves tempted to give in, accede, pander,
bow, and stoop to views we do not endorse, and to sell out, abandon,
recant, conceal, and compromise too readily those we do.

IV. PER SONA L A ND SOCI A L V IRTUES

I have argued that each of the three pictures of integrity reduces integrity to something else: to the conditions for unified agency, to the conditions for continuing as the same self, and to the conditions for having a
reason to refuse to cooperate with some evils. Although persons with integrity will sometimes stand up for what they wholeheartedly endorse,
or for what is central to their identity, or for deontological principles,
integrity is not equivalent to doing these things. Continuing to be of two
minds, conscientiousness about small matters, and dirtying ones hands
can also be matters of integrity.
39. That we have two integrity concernsacting on our best judgment given the world
as it is and acting on our best judgment of what we ought never to be called upon to
dosuggests that worlds can be sufficiently bad that acting with complete integrity is
hopeless. In making her choice, Sophie confronted just such a world.

143

MOR A L A IMS

I said at the beginning of this chapter that I thought there was a

second problem with the three pictures of integritynamely that they
proceed on the assumption that integrity is a personal virtue, and that
this assumption wrongly limits what can be said about both the nature
and value of integrity. It is to that second critique that I now turn.
Some virtues are personal, others are social, yet others are both. A
personal virtue, like temperance, consists in having the proper relation
to oneselfin this case, to ones desires. Social virtues consist in having
the proper relation to others. Civility, for instance, is a social virtue, a
desirable mode of conducting oneself among others. Some virtues are
both personal and social. Self-respect, for instance, might be thought to
involve having both a proper regard for ones own moral status (and thus
the right relation to oneself) and a proper regard for ones place among
other moral beings (and thus the right relation to others); it is a virtue
exercised both by holding oneself to standards and by demanding rightful treatment from others.40
On the integrated-self, identity, and clean-hands pictures, integrity characterizes an agents relation to herselfto her desires (they
are wholeheartedly endorsed or else outlawed), to her character (she
cultivates and protects its depth), and to her agency (she takes special
responsibility for what gets done through it and governs herself by at
least some deontological principles). Given this understanding of integrity as a personal virtue, guarding ones integrity must be largely selfprotective. It is for the sake of my autonomy, my character, my agency
that I stand by my best judgment. Or alternatively put, it is for the sake
of some specially valued feature of selves, of which I am one, that I stand
by my best judgment.
Characterizing integrity as a purely personal virtue does not imply
that there is anything self-indulgent about striving to have integrity.
But it does imply that integrity is not essentially connected to how we
conduct ourselves among others and that its fitting us for proper social
relations is not what makes it a virtue. Is there any reason to think that
integrity is less like temperance, a purely personal virtue, and more like
40. Thomas E. Hill, Jr., Servility and Self-Respect and Self-Respect Reconsidered, explores both the personal and the social dimensions of self-respect.

144

STA N DI NG FOR SOM ETHI NG

self-respect, a personal and social virtue? Taking the notion of standing

for something and the self-indulgence criticism of integrity in turn, I
want to suggest two reasons for not confining the analysis of integrity to
understanding its nature as a personal virtue. First, doing so fails to provide us with an adequate explication of what it means to stand for something. Second, although such analyses can counter the self-indulgence
charge, they cannot make the person of integritys relation to other persons central to that defense.

Standing for Something

I take it that the notion of standing for something is central to the
meaning of integrity. Indeed, the intuitive appeal of the integrated-self,
identity, and clean-hands pictures lay in their articulating part of what
is meant by standing for something. When, however, the analysis of
integrity is confined to understanding it as a personal virtue, standing
for something ultimately reduces to standing by the line that demarcates
self from not-self. On the integrated-self, identity, and clean-hands pictures, the adoption of principles and values as ones own establishes the
line between self and not-self. Acting with integritythat is, on ones
own judgmentis thus intimately tied to protecting the boundaries
of the self, to protecting it against dis-integration, against loss of selfidentity, and against pollution by evil. Acting without integrity undermines the boundaries of the self, whether that be accomplished through
the abandonment of ones autonomy, the betrayal of ones deepest commitments, or the contamination of ones agency through association
with evil. On all three views, loss of integrity signals loss of some important dimension of selfhood.
To the extent that integrity is, indeed, a personal virtue, this account
of the significance of standing by ones principles and values rings true.
What drops out of these accounts, however, is the centrality of standing
for principles and values that, in ones own best judgment, are worthy
of defense because they concern how we, as beings interested in living
justly and well, can do so. When President Clinton capitulated to the
Joint Chiefs of Staff and members of Congress, such as Sam Nunn, about
the military ban on gays and lesbians, he was criticized, particularly by
145

MOR A L A IMS

the gay and lesbian community, for lacking integrity. The force of that
charge was not that he had failed to sustain (or had misrepresented) the
boundaries of his self. The force of the charge was that he had treated as a
matter of little significance the representation and defense of views that
in ones own best judgment are the better ones. He did so either by misrepresenting his own view of the ban in the first place or by too readily
conceding to a view he considered wrong. This, in the eyes of his critics,
constituted less a self-betrayal than a betrayal of those counting on him
to stand up for what they took to be the better view. Moreover, not standing up for ones best judgment about what would be just or what lives
are acceptable forms of the good suggests that it does not really matter
what we as a community of reasoners endorse. The person of integrity,
one might plausibly think, is precisely the person who thinks this does
matter. Integrity here seems tightly connected to viewing oneself as
a member of an evaluating community and to caring about what that
community endorses. That is, it seems to be a social virtue.

Self-Indulgence
The depiction of integrity as a personal virtue aimed at securing the
boundaries of the self tends to provoke charges of self-indulgence. This
self-indulgence charge can, I think, be countered. Even so, a further
question remains as to whether accounts of integrity as a personal virtue
enable us to say all the things we want to say about what makes integrity
a virtue.
The self-indulgence critique goes something like this: advocates of
integrity seem to place evaluative weight on the fact that a view is ones
own. This looks self-indulgent; the identity-picture of integrity is especially prone to this criticism. On one version of the identity picture, the
core principles of ones deliberative viewpoint are core principles not because one thinks them worthy of endorsement but simply because one
so thoroughly identifies with them. But all three pictures, because they
value standing on ones own views, are vulnerable to charges of egoism
and self-indulgence.
The proper line of defense to this charge is to point out that value
is being attached not to the ownness of a view but to something else of
146

STA N DI NG FOR SOM ETHI NG

which formulating and acting on ones own views is an integral part.

Briefly reconstructing how such arguments would go, one might say the
following.
The integrated-self picture of integrity attaches value to autonomy.
The project of becoming a person with integrity just is the project of
becoming a fully autonomous person whose actions are determined
by herself rather than by desires and values that are not truly her own.
Having and acting on views of ones own is thus valuable not because of
the sheer fact that they are ones own but because having and acting on
views of ones own is integral to being an autonomous, free, and responsible being, which itself is valuable.
What the identity picture of integrity attaches value to is somewhat
harder to specify. The thought might be that the depth of character that
comes with deep commitments is an admirable characteristic of persons. Or the thought might be that deep attachments are part of any life
that could count for us as a good, full, and flourishing human life. Or the
thought might be that only a life containing deep attachments will be rich
enough to compel our continuing interest in staying around and participating in morality. Having and acting on identity-conferring commitments is thus valuable, not because of the sheer fact that they are ones own
but because having and acting on deep commitments is part of any admirable, flourishing life worth living, and that kind of life is what has value.
What the clean-hands picture of integrity attaches value to is again
not easy to specify. One thought might be that special value attaches
to taking responsibility for ones own conduct. In a quite different vein,
one might claim that value attaches to adopting a deontological rather
than consequentialist perspective, and thus to acting on principle itself.
In either case, that the principles happen to be ones own principles is
incidental and inevitable given that deliberation about which principles
are endorsable will have to be conducted from within ones own deliberative viewpoint.
Although I will not attempt to do so here, I think all three views of
what makes integrity a virtue might be articulated in either Kantian or
utilitarian terms.41
41. I owe this thought to Geoffrey Sayre-McCord.

147

MOR A L A IMS

However, even if the integrated-self, identity, and clean-hands views

succeed in accounting for the virtue of integrity, one might still criticize
them for excluding some important considerations from their account.
Some Kantian and utilitarian arguments for the value of integrity will
be ruled out as arguments also for the virtue of integrity.42 In On Liberty,
for instance, Mill argues that the unrestricted representation and exchange of ideas are critical to the discovery of truth. But the discovery of
truth would seem to depend not just on the freedom to speak but also on
the integrity of the speakersthat is, on their commitment to publicly
standing for their own best judgment of what the truth is. Kantians, too,
might see some value in standing before others on ones own best judgment. From a Kantian point of view, persons are not just autonomous
agents with special responsibility for their own conduct. They are also
members of a community of co-legislators. The embodiment of this colegislative aspect of persons would seem to require agent integritythat
is, a commitment to standing before others on ones best judgment, submitting it to others critiques, and defending its fitness for co-legislation.
From the standpoint of the integrated-self, identity, and clean-hands
pictures of integrity, however, these considerations only provide additional reasons for valuing integrity, not for thinking it a virtue. For the
latter to be true, we would have had to start from an account of integrity
as a social virtue. That is, we would have had to start from the thought
that acting on ones own best judgment is integral to some common
project (such as the search for truth or co-legislatable principles) or to
a way of comporting ourselves among others. Only if we assume that
integrity is not, or not just, a matter of the individuals proper relation
to herself, but is a matter of her proper relation to common projects and
to the fellows with whom she engages in those common projects, would
42. To show that a trait is a virtue is to show that something about the trait itself is intrinsically valuable. For instance, on the integrated-self picture, autonomy is both central
to the trait we call integrity and has intrinsic value. Thus integrity is a virtue. Traits
can also have extrinsic value. For instance, one might think that even though integrity
is not to be defined as a trait that fits us for membership in a truth-seeking or moral
community, this is one welcome effect of integrity. If so, its fitting us for community
membership provides us with an additional reason for valuing integrity, though not an
additional reason for thinking it a virtue.

148

STA N DI NG FOR SOM ETHI NG

the utilitarian and Kantian considerations just mentioned count as articulating what makes integrity a virtue.
Contrary to the integrated-self, identity, and clean-hands pictures of
integrity, I am strongly inclined to think that integrity is a social trait and
that its fitting us for community membership is precisely what makes it
a social virtue. Looking at integrity as a social virtue enables us to see
persons of integrity as insisting that it is, in some important sense, for
us, for the sake of what ought to be our project or character as a people,
to preserve what ought to be the purity of our agency that they stick by
their best judgment. It is to a picture of integrity as a social virtue that
I now turn.

V. THE SOCI A L V IRTUE OF INTEGR IT Y

What, then, is the social virtue of integrity? I begin with this picture: I
am one person among many persons, and we are all in the same boat.
None of us can answer the question, What is worth doing? except
from within our own deliberative points of view. This What is worth
doing? question can take many specific forms. What evils, if any, ought
one morally to refuse to do no matter the consequences? What, for philosophers, is worth writing about? What is worth keeping, what worth
reforming in the social identity black or woman or gay? What principles take precedence over what others? What is one, if not the only,
worthwhile way of conducting a good life? That they are answerable
only from within each persons deliberative viewpoint means that all of
our answers will have a peculiar character. As one among many deliberators, each can offer only her own judgment. Although each aims to do
more than thisto render a judgment endorsable by allnothing guarantees success. The thought It is just my judgment and it may be wrong
cannot be banished no matter how carefully deliberation proceeds. But
given that the only way of answering the What is worth doing? question is to plunge ahead using ones own deliberate viewpoint, ones best
judgment becomes important. As one among many deliberators who
may themselves go astray, the individuals judgment acquires gravity. It
is, after all, not just her judgment about what it would be wrong or not
149

MOR A L A IMS

worthwhile to do. It is also her best judgment. Something now hangs for
all of us, as co-deliberators trying to answer correctly the What is worth
doing? question, on her sticking by her best judgment. Her standing for
something is not just something she does for herself. She takes a stand
for, and before, all deliberators who share the goal of determining what
is worth doing.
To have integrity is to understand that ones own judgment matters
because it is only within individual persons deliberative viewpoints, including ones own, that what is worth our doing can be decided. Thus
ones own judgment serves a common interest of co-deliberators. Persons of integrity treat their own endorsements as ones that matter, or
ought to matter, to fellow deliberators. Absent a special sort of story,
lying about ones views, concealing them, recanting them under pressure, selling them out for rewards or to avoid penalties, and pandering
to what one regards as the bad views of othersall these indicate a failure to regard ones own judgment as one that should matter to others.
The artist who alters his work of genius, making it saleable to a tasteless public, lacks integrity because he does not regard his best aesthetic
judgment as important to anyone but himself. He abandons the codeliberative perspective. And those who act for the sake of preserving
their identity, but without asking whether it is worth preserving, lack integrity; this is because they do not even raise the What is worth doing
question. Whatever sells and whatever is me cannot ground action
with integrity because these reasons do not address the co-deliberative
question of what is worth doing.
That hypocrites lack integrity is a common observation. Analyses
of integrity as a personal virtue, however, do not plausibly explain why.
On the integrated-self and identity pictures of integrity, one would have
to say that hypocrites lack integrity because their actions are not integrated with their endorsements; or because in the course of pretending
commitment, they are untrue to their real, identity-conferring commitments; or because sustained pretense undermines the agents ability
to be clear and not self-deceived about what she really does endorse.43
43. Taylor and Blustein both stress the way sustained hypocrisy may result in selfdeception and unclarity about what one really endorses.

150

STA N DI NG FOR SOM ETHI NG

Although hypocrisy may be bad in these ways for the hypocrite, this is
not typically why we charge hypocrites with lacking integrity. Hypocrites mislead. And it is because they deliberately mislead people about
what is worth doing that they lack integrity. Jim Bakker, for instance,
persuaded a lot of people to invest money in his doing Gods work. His
embezzling revealed that he had misled them either about the value of
doing Gods work or the value of his doing it. Neither the integrated-self
nor the identity picture of integrity can explain why misleading others,
by itself and not because of its deleterious effects on the hypocrite, has
anything to do with lacking integrity. If, however, integrity is not a
merely personal virtue, but is the social virtue of acting on ones own
judgment, because doing so matters to deliberators common interest in
determining what is worth doing, then hypocritical misrepresentation
of ones own best judgment clearly conflicts with integrity.
This view of integrity also helps to explain the shame at failure to
abide by ones own judgment as something more than mere shame at the
unsturdiness of ones will or the guilty awareness of violating a standard.
If an agent passes herself off as someone who insists on the importance
of private spaces, and then secretly indulges in reading anothers private
letters, the thoughts I have no self-control and This is wrong are different from the thought I have no integrity. Neither the weakness nor
the wrongness of the act immediately reveals lack of integrity. Rather,
the thought I have no integrity accompanies the revelation of ones
inability to stand for something before others.
Finally, looking at integrity not as the personal virtue of keeping
oneself intact but as the social virtue of standing for something before
fellow deliberators helps explain why we care that persons have the courage of their convictions. The courageous provide spectacular displays of
integrity by withstanding social incredulity, ostracism, contempt, and
physical assault when most of us would be inclined to give in, compromise, or retreat into silence. Social circumstances that erect powerful deterrents to speaking and acting on ones own best judgment undermine
the possibilities for deliberating about what is worth doing. We thus have
reason to be thankful when persons of integrity refuse to be cowed.
Understanding integrity as a social virtue also shifts our sense of
what the obstacles to integrity might be. On the integrated-self picture,
151

MOR A L A IMS

the primary obstacles to integrity are internal: self-deception, weakness

of will, shoddy practical reasoning, inconsistency among and ambivalence about ones endorsements. These are no doubt obstacles. But what
of contempt, ostracism, loss of a job, penal sanctions, the breakdown of
friendships and familial relations, or being labeled confrontational,
difficult, overly sensitive, or militant, not to mention the inexhaustible confidence of others that one is wrong? These are public obstacles to
acting with integrity. Even the thickest skinned and toughest willed may
find them hard to stand up against, especially on a continuing basis.
If integrity is the virtue of having a proper regard for ones own judgment as a deliberator among deliberators, it would seem that integrity
is not just a matter of sticking to ones guns. Arrogance, pomposity,
bullying, haranguing, defensiveness, incivility, close-mindedness, and
deafness to criticism (traits particularly connected with fanaticism)
all seem incompatible with integrity. All reflect a basic unwillingness
or inability to acknowledge the singularity of ones own best judgment and to accept the burden of standing for it in the face of conflict.
Moreover, acknowledging others as deliberators who must themselves
abide by their best judgment seems part of, not exterior to, acting with
integrity. Untempered by the thought This is just my own best judgment, standing for something puts ones own and others integrity at
riskones own because of the temptation to supplement standing for
with coercive pressure, and others because coercion may work. This is
to say that when what is worth doing is under dispute, concern to act
with integrity must pull us both ways. Integrity calls us simultaneously
to stand behind our convictions and to take seriously others doubts
about them. Thus neither ambivalence nor compromise seems inevitably to betoken lack of integrity. If we are not pulled as far as uncertainty
or compromise, integrity would at least demand exercising due care in
how we go about dissenting. Because we so often seek exemplars of integrity retrospectively, identifying those who championed causes that
to us now are clearly worthy, it is easy to overlook what, from their earlier vantage point, acting with integrity must have looked like. Socrates,
Galileo, Luther, and King acted against the best judgment of their peers,
including some whom they admired. To think that caving in to their
peers posed the only threat to their integrity oversimplifies the nature of
152

STA N DI NG FOR SOM ETHI NG

integrity. Hubristic denial that others best judgment matters posed an

equal threat. However admirable those with the confrontational courage of their convictions may be, even protesters risk losing their integrity to arrogance.

Concluding Remark
What I have had to say about integrity suggests that integrity may be a
master virtuethat is, less a virtue in its own right than a pressing into
service of a host of other virtues: self-knowledge, strength of will, courage, honesty, loyalty, humility, civility, respect, and self-respect.44 My
aim was to understand that service. What is a person who tries to have
integrity trying to do? I have not rejected (though I have revised) the
ideas that she is trying to be autonomous, or loyal to deep commitments,
or uncontaminated by evils. But I have tried to argue that this is not the
whole story. She is also trying to stand for what, in her best judgment, is
worth persons doing.

44. I owe this observation to Owen Flanagan.

153

PA R T I I I

CON V E N T ION A LI Z E D
W RONG DOI NG
The preceding chapters focused on the importance of social consensus on
moral norms and the process by which this takes place. Social consensus establishes the bounds of civility, specifying which views and behaviors are not
owed a civil response and thus when agents are morally permitted not to display respect, tolerance, and considerateness toward others. Social consensus
also gives content to the idea that minimally well-formed moral agents can
be normatively expected to engage in some acts of what I called moral gift
giving. Social norms both specify which elective acts are sufficiently nonmotivationally taxing and obviously and unambiguously good to be expectable
from any minimally well-formed agent, and render those elective acts motivationally nontaxing and obviously and unambiguously good in the first
place. Central to the process of establishing social consensus on moral norms
are both civility and integrity. Norms of civil speech and behavior prevent
negotiations over which moral views and ways of living are acceptable from
breaking down and thus facilitate the process of reaching social consensus.
Integrity, understood as standing before others for ones own best judgment,
ensures that candidate views for social consensus are put forward both under
conditions where there is social dispute about what the right views are and
under conditions where there is social consensus.
This next pair of chapters takes up a problem only briefly raised in
the discussion of civility: the possibility that social consensus will settle on
the permissibilityor worse, the obligatory natureof what is in fact

MOR A L A IMS

thewrongful treatment of others, thus conventionalizing injustice. This is a

particularly serious issue for moral theorizing, given the actual pervasiveness
of conventionalized injustice within real-world social practices of morality.
It is also especially serious because cultural ideologies, factual misinformation, and socialization to accept existent social norms make conventionalized injustice exceptionally difficult to spot. Chapter6 considers the question
of what changes we would need to make in the tools we use for critical moral
reflection in order to enhance our ability to detect conventionalized injustice
and to explain what makes some cooperative schemes wrong despite the fact
that those who are wronged appear to willingly comply with them. This is
a large topic and I look at only one tiny bit of it: how could we reconstruct
Kants universalizability test (the CI procedure) so that it is more useful in
detecting and explaining the wrongness of conventionalized injustice?
The tools we need in order to be adequately critical of conventionalized
wrongdoing often emerge from social movements and from areas of academia
other than moral philosophy. Feminist consciousness-raising groups, womens rights organizations, and womens studies programs, for example, generated many of the conceptual tools for detecting and explaining the wrongness
of conventionalized mistreatment of women. These individuals and groups
operated at the frontiers of new moral knowledge in what I call abnormal
moral contexts, where their insights into wrongdoing were not broadly
known or, if known about, not accepted as genuine moral insights. Chapter7
looks at conventionalized wrongdoing, not from the perspective of the moral
theorizer interested in sharpening her tools but from the perspective of the
participant in social practices of morality. How will participants who are just
doing what everyone knows is permissible tend to regard being told that
they are morally ignorant, being excused for that ignorance, and being given
moral education? How will enlightened participants who refrain from conventionalized wrongdoing tend to view themselves and be viewed by others?
And should those who are on the receiving end of conventionalized wrongdoing refrain from reproaching those whose moral ignorance excuses them
from responsibility? I argue that in abnormal moral contexts, reproaching
only those who are morally responsible undermines the social functions of
reproach, namely to encourage people to see themselves as self-legislators and
to refrain from wrongdoing.

156

C ha pt e r6

Kant and Compliance

withConventionalized Injustice
Kants first formulation of the Categorical Imperative (CI) procedure for
testing a maxims moral acceptability requires that the agent ask herself
whether she could will her maxim to become universal law. Kant apparently takes the CI procedure to be straightforward and workable; Kant
interpreters typically do not. The CI procedure is notoriously problematic; some would say, unsalvageable. It is unclear at what level of generality maxims are to be framed. Framed too generally, they fail to describe
the agents intention in action and, worse, result in the CI procedures
generating overly rigorous, exceptionless duties. Framed to narrowly,
intuitively impermissible maxims pass the CI procedure. It is also unclear what kind of contradiction in conception we are to look for. If we
are to look for the logical impossibility of our action in a world where
our maxim has been universalized, then very few of our maxims will
fail the CI procedure. Moreover, natural acts like killing and torture
appear logically possible in a world of universal killing and torture.1 If
we are to look instead for the practical impossibility of our action in such
a world, then all maxims that involve coordination problems will fail no
matter how morally innocent our plans intuitively seem. 2 Finally, it is
unclear whether the purpose of the CI procedure is to generate duties
1. See, for example, Barbara Herman, Murder and Mayhem, in The Practice of Moral Judgment (Cambridge, MA: Harvard University Press, 1993); and Christine M. Korsgaard,
Kants Formula of Universal Law, Pacific Philosophical Quarterly 66 (1985): 2447.
2. Barbara Herman, Moral Deliberation and the Derivation of Duties, in The Practice of
Moral Judgment (Cambridge, MA: Harvard University Press, 1993), 132158.

157

MOR A L A IMS

or, as Barbara Herman suggests, to aid us in moral deliberation. 3 If the

CI procedure is used to generate duties, we seem saddled with the problem of rigorism. If it is used simply to aid deliberation, it would seem we
must already have established moral presumptions against which to test
maxims that propose exceptions to these presumptions.
In this chapter, I want to set aside this now familiar list of the CI procedures troubles and turn to a different sort of worry: Can the CI procedure reveal injustice when injustice has been conventionalizedthat
is, when there is general social compliance with an unjust maxim? Although all four of Kants Groundwork examples of the CI procedure are
problematic, it is in the case of unjust maxims that the procedure seems
to work most straightforwardly. In his false promising example, an agent
proposes making an exception of herself to a conventional social practice of promise keeping. Widespread compliance with the convention
of promise keeping is what enables the agent to use a false promise to
her advantage. That is, promising (whether falsely or not) works only
when the maxim of keeping promises is already widely accepted. It is
this dependence of all promising, including false promising, on general
compliance with the maxim of promise keeping that makes universalizing the agents maxim of false promising incoherent. Now, there are two
ways of understanding the relation between successful promising and
general compliance with a promise-keeping maxim. One might think
the relation is logical. By definition there could be no promises in the
absence of a convention of promising. Or one might think the relation
is practical. In point of fact, we as rational beings would do things differently (for example, we would not accept promises) if the patterns of
general compliance were different.4 I find the latter interpretation most
plausible (and I will have more to say about this later). But it raises the
following problem: Suppose that what the agent proposes is not making
an exception of herself, but quite the opposite. She proposes complying
with a conventional practice. Now, suppose that practice is unjust. What
3. Ibid.
4. See Onora ONeill, Acting on Principle: An Essay on Kantian Ethics (New York: Columbia
University Press, 1975), especially the example of bank robbery, 7072. Were attempted
bank robbery sufficiently widespread, we would devise different, more effective security
systems than we do now.

158

K A NT A ND COMPLI A NCE W ITH INJUSTICE

then? In this case, the thought We as rational beings would do things

differently, and differently enough to thwart the agents plans can gain
no foothold. When the agents maxim is already actually universalized
(in the sense that there is general compliance), it would seem that the
CI procedures hypothesis of universal and natural lawlike compliance
cannot reveal anything new about how we as rational beings would do
things if the agents maxim were universalized.

I. SOM E INITI A L WOR R IES

One might suspect that there is something fishy about this alleged problem. Exactly what kind of universalization is being imagined in talk
about general compliance with unjust maxims? Conventionalized injustice in our society typically takes the form of treating different social
groups differently. Racism, sexism, heterosexism, anti-Semitism, and
so on all have this form. Thus, it seems, conventional unjust practices
exhibit a failure to universalize maxims, not the universalization of an
unjust maxim. I will argue that this response is question begging. If it is
ever legitimate to treat different people differently (as surely it is), then
the claim that racism, for example, involves the failure to universalize
ones maxims must be the conclusion of a moral argument demonstrating why, in this case, differential treatment is unwarranted. To make it a
premise begs the moral question.
In a different vein, one might be dubious that Kants application of
the CI procedure to the false-promising case relies on any appeal whatsoever to existing social conventions. Although Kant states that the CI
procedure reveals both the fact that we wish only to make an exception for ourselves and the irrationality of this noncompliance, and that
in any transgression of duty [w]e only take the liberty of making an
exception to the law for ourselves (or just for this one time) to the advantage of our own inclination,5 it is a mistake to think that what he
has in mind is noncompliance with or making ourselves an exception to
5. Immanuel Kant, Grounding for the Metaphysics of Morals, trans. James W. Ellington (Indianapolis, IN: Hackett, 1981), 32 [424].

159

MOR A L A IMS

social conventions. The relevant exception is not to social norms but to

the moral law. If so, the CI procedure should work regardless of whether
or not the moral laws we propose transgressing are entrenched in social
norms. In either case, it should show us what is wrong with excepting
ourselves from moral lawsnot what is wrong with excepting ourselves
from conventional social norms.
But consider how Kants argument against lying promises goes.
An agent proposes noncompliance with a purely conventional social
practicepromise making, accepting, and keeping. She proposes
making a promise but not keeping it. The CI procedure asks her to
universalize her proposal. In willing that she will make false promises
when convenient, she has willed the very conditions under which she
will be unable to realize her original plan. No one will take her promises
seriously.
The argument works independently of any moral assessment of the
practice of promising itself. At no point is the question raised whether
rational beings must, or could, will the practice of promising itselfthat
is, whether the practice is either morally required or morally permissible. One could perhaps use the contradiction-in-willing test to show,
for example, that rational but humanly finite beings could not rationally will a world devoid of promise making, accepting, and keeping. But
Kant did not see the necessity of doing so. In his view, the CI procedure
has done all its work once it shows how the agent is depending on the
very practice whose rules she proposes transgressing. Her noncompliance presupposes general compliance. And that is what is wrong with
her proposal.
Suppose, by contrast, she proposes keeping her promise. This maxim
will sail right through the CI procedure. It will do so not because of any
moral feature of the practice of promising. Rather, it passes simply because her proposal does not involve making an exception for herself to a
practice that she assumes others will continue upholding.
This is worrisome. The CI procedure appears tailor-made to catch a
particular form of moral failing in its net, namely the temptation to make
an exception of ourselves to generally followed social practices that
are assumed to be morally acceptable ones. Because it is not designed
to assess the practice itself, all maxims of compliance with generally
160

K A NT A ND COMPLI A NCE W ITH INJUSTICE

followed practices pass it. 6 Such maxims lack the internal incoherence
of maxims of noncompliance.
The suspicion that the CI procedure is tailored to a particular moral
failing is buttressed when one considers Kants account of temptation.
In his view, the moral law takes the form of an imperative for finite rational beings because such beings suffer from inclinations that tempt
them to act on maxims that could not be universalized, to use others
as means to ends that those others do not share, and to govern their
actions from a personal rather than an impersonal legislative point of
view.7 In short, rather than thinking from a fully social point of view,
we are tempted to take private advantage and personal happiness as
supplying sufficient reasons. As a result, we endorse policies for ourselves without first asking whether we could will that they be policies
for all.
Kant generally appears blind to a moral failing that, from a twentyfirst century point of view, might seem more salient and that was salient to many of his Enlightenment peers: the temptation to use the very
conventionality of social practices as sufficient reason for action. In his
analysis of temptation, Kant appears more the Protestant than the Enlightenment figure, more concerned to check the power of desire than
to question the legitimacy of some social conventions. In the Metaphysics of Morals and Lectures on Ethics, Kant emerges as the defender of
conventional attitudestoward suicide, masturbation, homosexuality,
6. This is a variant of the Hegelian objection that the wrongness of not returning deposits
depends on an antecedent and unstated assumption about the desirability of a system of
private property.
7. This is not to say either that inclinations are in themselves bad or that they cannot provide additional support for the motive of duty (something Kant clearly thought they
could do). It is to say that Kant presents the conditions necessary for the possibility of
moral failure primarily in terms of the internal psychology of the agent, rather than in
terms of the influence of external social conventions and beliefs. What corrective policy
regarding inclinations Kant thought we should adopt is unclear. Although, as one reviewer points out, Kant claims in Religion Within the Limits of Reason Alone that it is
not only futile to want to extirpate them but to do so would also be harmful and blameworthy; yet Kant also remarks in the Grounding for the Metaphysics of Morals that the
universal wish of every rational being must be... to be wholly free from them (35 [428])
and that reason in the consciousness of its dignity despises such incentives and is able
gradually to become their master (22 [411]).

161

MOR A L A IMS

marriage, the death penalty, civil revolution, and womens civic status.
Even when, in the case of honor, conventional practices of dueling and
killing illegitimate infants appear to him misguided, he is willing to
give conformity to convention excusing weight rather than condemn
such conformity. 8
Kants stance against any right to revolution also suggests that the
dangers of departing from established social rules worry him more than
the possible injustice of those rules. Subjects of states, in his view, are
obligated to put up with even what is held to be an unbearable abuse
of supreme authority9 by governments that proceed in a thoroughly
brutal (tyrannical) fashion.10 That is, subjects have no right to sedition
or rebellion.11 On one interpretation, Kants thought is that it is better
to support a system of public law that declares for all what is just, however corrupt that laws conception of justice may be, than to return to a
state of nature in which each does what seems just to himself and where
there is no mechanism for adjudicating between conflicting perceptions of justice.12 Although inferences from Kants political views to his
moral views are hazardous, one might imagine that Kant would have
regarded rebellion against well-established social conventions to be a
worse option than conformity with those conventions, however corrupt
they might be. Socially agreed upon norms at least approximate, even if
imperfectly, the legislation of a kingdom of ends.
On this background of doubt that Kant ever intended the CI procedure to condemn maxims of compliance with unjust social conventions,
I now turn to examining Kantian arguments against compliance with
one specific unjust practiceslavery. I will be arguing that standard
Kantian arguments against slavery do not work. They typically depend

8. Immanuel Kant, The Metaphysics of Morals, trans. Mary Gregor (Cambridge: Cambridge
University Press, 1991), 144145 [336337].
9. Ibid., 131 [320].
10. Immanuel Kant, On the Proverb: That May be True in Theory, But is of No Practical
Use, in Perpetual Peace and Other Essays on Politics, History, and Morals, trans. Ted
Humphrey (Indianapolis, IN: Hackett, 1983), 79 [299].
11. Ibid.
12. Jeffrie G. Murphy, Kant: The Philosophy of Right (New York: St. Martins, 1970); David
Cummiskey, Kantian Revolutions, unpublished manuscript.

162

K A NT A ND COMPLI A NCE W ITH INJUSTICE

on an overly narrow definition of slavery. Worse, because they assume

that we must always universalize slave-holding maxims across all rational beings, they beg the central question in all disputes about the justice of inegalitarian practices: Are there differences that would warrant
treating two social groups differently? In the second half of this chapter,
I will be arguing for a different understanding of the CI procedure, one
that does not assume all inegalitarian practices are wrong but, instead,
enables agents to assess which inegalitarian practices could be made
universal law.

II. PROBLEMS IN A PPLY ING THE CI PROCEDUR E

TO SL AV ERY
One might suppose that the CI procedure can be applied straightforwardly to unjust systems such as slavery. One might, for example, suppose that would-be slaveholders adopt the maxim I will that I own
humans as property. The CI procedure asks the agent if he can conceive
of the universalized maximAll will own humans as property
obtaining as a universal law of nature. But, obviously, not all can own
humans if there are to be any humans available to be owned. Bluntly,
we cant all be masters.13 Insofar as this running of the CI procedure
brings out the fact that slaveholders do not want their slaving policies
applied to themselves, it has intuitive appeal and is in the spirit of Kantian thinking. However, this particular argument relies on what Kant
would regard as an ill-formed maxim. There are, for Kant, no maxims
of the flat form I will p. Maxims specify the reasons for endorsing a
course of action. American colonial slaveholders, for instance, did not
just will slaveholding. They willed using slaves to cultivate indigo, rice,
sugarcane, and tobacco for the purpose of advancing their agricultural economic interests. The Catholic Church endorsed slaveholding
in order to instruct heathen Africans in Catholicism, presumably for
13. I take this to be Onora ONeills argument in Consistency in Action, in Constructions
of Reason: Explorations of Kants Practical Philosophy (Cambridge: Cambridge University Press, 1989), 96.

163

MOR A L A IMS

the purpose of the greater glory of God.14 So at the very least, a slaveholding maxim must specify the reason-supplying purpose of owning
humans. Once we specify the purpose, universalization will not require
imagining that everyone is a master. It will require imagining that everyone who shares a particular purpose uses enslavement as the method
of choice for achieving it.
Once properly formed, will slaveholding maxims produce a contradiction in conception when universalized? Leslie Mulholland has
argued quite persuasively that they will. On Mulhollands view,
if we take the concept of a slave to be the concept of a human being
who is the rightful property of another, we find that it is analytically impossible for a slave to have rights to property. Hence, it is
analytically impossible for a slave to have slaves. Consequently,
it is analytic that if one person enslaves another the other cannot
enslave anyone when he needs their service.15

Here, we no longer suppose that all have the status of masters. Rather,
we suppose that slavery is the universal method of choice for promoting ones economic gain, the glory of God, and the like. But slavery, by
definition, rules out the possibility that slaves may use this method in
advancing their own economic or religious interests. Thus, we cannot,
without contradiction, conceive of a world in which everyone is using
slavery as the method of choice for advancing their own interests.

Definitional Narrowness
One major difficulty with this persuasive application of the CI procedure is that it depends on a narrow definition of slavery as a system
in which the enslaved are completely and permanently stripped of all
14. Claudine Hunting, The Philosophes and the Question of Black Slavery, Journal of
the History of Ideas 39 (1978): 405418. Louis XIVs 1685 Code Noir ordered masters
to have their slaves instructed in Catholic religion, etc., with heavy penalties should
they not comply (409).
15. Leslie Mulholland, Kant: On Willing Maxims to Become Laws of Nature, Dialogue
18 (1978): 92105, 97 (emphasis mine).

164

K A NT A ND COMPLI A NCE W ITH INJUSTICE

property rights. The enslaveds irrevocable state of propertylessness is

why they cannot adopt the slaveholding maxim to advance their own
interests. American slavery, however, was not entirely like this. Some
slaves managed to buy their freedom; others were given it.
Suppose, then, we consider a different slave system from Mulhollands, one where enslaved people retain entitlement to purchase their
own freedom and, once free, to own slaves themselves. No one is logically barred from adopting the slaveholding maxim. To be sure, purchasing ones freedom, and subsequently acting on the slaveholding
maxim, would be difficult given the economic conditions attending enslavement. But Kant does not require that others acting on our maxims
be easy, only that it be conceivable. One might recall, here, that Kant
believed heads of households had property rights over their servants
akin to property rights to things. Servants, he says, are included in what
belongs to the head of a household and, as far as the form (the way of
his being in possession) is concerned, they are his by a right that is like a
right to a thing; for if they run away from him, he can bring them back in
his control by his unilateral choice.16 What made this system of servitude acceptable to Kant was that servants contractually agreed to enter
servitude and could, in principle if not in point of economic fact, exit
servitude. They were thus, in principle, capable of possessing servants
themselves.
My point here is that only one specific form of slavery generates a
logical contradiction when universalized, namely a slave system that
stipulatively defines slaves as individuals who are permanently barred
from owning property, including other persons as property. By contrast,
any system that relies on economic impoverishment, rather than on
the stipulative denial of property rights, to keep individuals in a state
of servitude or enslavement will pass the CI procedures contradictionin-conception test. In such systems, enslaving relies on economic facts
about persons, much the way that killing relies on natural facts about
persons. As a result, we will have the same difficulty in generating a logical contradiction with respect to enslavement that we do in the case of
any natural act.
16. Kant, Metaphysics of Morals, 101 [283].

165

MOR A L A IMS

Barbara Hermans discussion of convenience killing usefully illustrates the problem. Applying the CI procedure to maxims of convenience killing requires us to imagine a world where killing is the universal
method of choice for advancing ones interests. But what of this imagined world? she asks.
If everyone killed as they judged it useful, we would have an unpleasant state of affairs. Population numbers would be small
and shrinking; everyone would live in fear. These are bad consequences all right. Still a world that looks like this is conceivable:
Hobbes described it in some detail. And if there is nothing inconceivable or contradictory in thinking of a world that contains a
Hobbesian law of killing, it looks as though we must conclude
that the CC [contradiction-in-conception] test does not reject the
maxim of killing.17

Similarly, it seems, we must conclude that the contradiction-inconception test does not reject the maxim of slaveholding when slaves
inability to hold slaves themselves is due entirely to economic impoverishment. It would indeed be an unpleasant world as more and more
people are physically or economically coerced into positions of servitude that would be terribly difficult to leave. But it is conceivable.18
Hermans response to the contradiction-in-conception tests failure to demonstrate the wrongness of convenience killing is to apply
the contradiction-in-willing test instead. One might do the same for
slave systems. Even if universal adoption of some slave maxims is conceivable, it might be impossible to will their universal adoption. This
strategy has an obvious drawback. Only those maxims that fail the
contradiction-in-conception test are ones that we have a legally enforceable
17. Herman, Murder and Mayhem, 118.
18. Christine Korsgaard, Kants Formula of Universal Law, 31, makes the same point: no
amount or kind of use of the action of killing is going to make it impossible. And this is
because the existence of this kind of action and its efficacy depend only on the laws of
nature, not on any conventional practice. She takes this fact about natural acts to be a
reason for adopting the Practical Contradiction rather than the Logical Contradiction
interpretation of the contradiction-in-conception test.

166

K A NT A ND COMPLI A NCE W ITH INJUSTICE

duty of justice not to act on. Thus, even if we can use the CI procedure to
show that some slave systems are not willable without contradiction, the
CI procedure will still not have been vindicated. What is at issue is not
just the CI procedures capacity to tell us that slavery is wrong, but also its
capacity to tell us that slavery is a particular kind of wrong, namely a violation of a legally enforceable duty of justice.19

Universalization
There is a second, quite different sort of objection one might raise to the
thought that what the CI procedure asks is What if everyone were masters? or Could everyone adopt a slaveholding maxim for advancing
their interests? One might deny that all maxims must be universalized
across the entire population of rational beings. What if I intend simply
to play out my role in a cooperative scheme? Must people who occupy
a different role really be capable of acting on my role-specific maxim?
The thought here is that some differences are morally important ones,
and blindly universalizing across all rational beings is bound to obscure
them. White colonial Americans thought there were morally important
racial differences. Automatically universalizing slaveholding maxims
across all rational beings, rather than the restricted category of whites,
would have seemed question begging to them.
Does the CI procedure always require asking, Could my maxim
become a universal law for all rational beings? Kants promise-keeping
example suggests not. The practice of promising requires complementary roles. There are promise makers (promisors) and promise acceptors
(promisees). Within the practice, the two parties are bound by different
rules. Those who make promises are to keep them. Those who accept
promises are not to ask for further assurances or demand fulfillment
before the agreed-upon time. False promisors violate the rules of the
promise makers role. The CI procedure does not require universalizing
a false-promising maxim across all rational beings. It requires universalizing across the category of promisor by asking, What if all promisors
19. For a further discussion of this problem in relation to maxims of murder, see Korsgaard,
Kants Formula of Universal Law.

167

MOR A L A IMS

obeyed your maxim? In that event, the practice would break down.
The illusion that a wider universalization takes place in Kants example
comes simply from the fact that, in our society, the promisor role is open
to everyone. It would seem, then, that under colonial Americas white
supremacist system, which institutionalized different racial roles, whites
who asked Could my maxim of owning Africans become a universal
law for whites? were not obviously violating the CI procedure. 20
One might, of course, think that the openness of the promisor role,
or of any other role in a cooperative scheme, is important. The thought
might be that it is legitimate to universalize across the restricted promisor category only because any rational being can become a promisor,
and thus the rules applying to promisors also apply to all rational beings.
What is wrong with universalizing across the restricted category of
whites is that it is a closed role, not open to all rational beings, and thus
the resulting rules do not apply to all rational beings.
But is it really a priori necessary that all roles be open? Kant himself
regarded parent-child and husband-wife as closed roles. Their occupants
are determined by nature. In other cases, even though the occupants
are not determined by nature, the roles appear closed by comparison
to the openness of the promisor role. Consider teacher-student, doctorpatient, employer-employee, jailor-incarcerated, landlord-tenant, and
similar complementary sets of roles. In each case, there are qualifying
conditions for occupying the role. Not just anyone can claim to be a
teacher, and we do not usually think it wrong of teachers to endorse a
policy of reserving the job of lecturing for teachers.
Where our intuitions are that the role is legitimately closed, our intuitions will also be that universalizing must be restricted to role occupants, and not extended to all rational beings. If not restricted, the CI
procedure has quite implausible results. Consider teachers: A teacher
proposes lecturing for the purpose of educating others. If she universalizes her maxim across all rational beings, she may well conclude that
she is not permitted to lecture. If all, regardless of their qualification
20. This means that even enslavement maxims that analytically bar the enslaved from
adopting the enslavement maxim may fail to generate a logical contradiction once
universalized.

168

K A NT A ND COMPLI A NCE W ITH INJUSTICE

to teach, use lecturing as the educative method of choice, then no one

will assume that lecturing has educative value. They will regard lectures
much as they regard promises in a world of false promisingwith disbelief. In order to pass the CI procedure as we think it should, the teachers
maxim of lecturing must be universalized only across teachers. Thus,
the sheer fact that a role is closed, by natural or by acquired qualifications, does not entail that we may not restrict universalization to role
occupants.
In sum, if cooperative schemes involving complementary rulegoverned rolesincluding ones that assign roles according to natural
or acquired qualificationsare going to pass the universalization test,
then we must be careful not to overread what universalization requires.
It does not require that all persons can adopt a particular maxim as a
guide to their own behavior. For example, it does not require that students be capable of adopting and acting on maxims that apply to teachers. As I shall argue later, it does require that all involved in a cooperative
scheme find it reasonable to continue cooperating once a maxim for
action within the complementary role has been universalized within
that role (So, for instance, promisees find it unreasonable to continue
cooperating once a maxim of false promising has been universalized
within the promisor role.)21
Attention to cooperative schemes brings out a simple point: it is not
in itself wrong to treat people differently. To assume at the outset that
slaveholding maxims must be universalized across all rational beings is
to beg the question. The wrongness of a particular inegalitarian system
is something that must be proved, not assumed.
This point is obscured when slaveowning maxims are discussed as
though the agent proposes simply collaring people off the streets and
pressing them into servitude, where it doesnt matter whom he collars
and thus anyone would be fair game. If that were the proposal, then the
slaveholding maxim must be universalized across all rational beings.
But slavery, as part of an inegalitarian racial system, does not work like
this. The would-be slaveholder does not propose enslaving just anyone.
21. Thus, even in the first formula of the Categorical Imperative, which stresses universalization, it is necessary to anticipate the third formulas emphasis on co-legislation.

169

MOR A L A IMS

He proposes enslaving people of a different race. Pre- and postcolonial

Americans justified racially based slavery on the grounds that the Africans they traded in were, they thought, childlike, barbaric, heathen,
incapable of caring adequately for their own welfare, incapable of reasoning, including moral reasoning, and destined by God in the story of
Ham to be the servants of whites. These were their reasons for thinking
a social system of slavery made sense. They presumed that servitude to
anglo-Americans was both a natural and a beneficial state for Africans.
Slavery was, in their view, functional for both parties. Any formulation
of the slaveholding maxim connecting slavery solely to self-interest misrepresents the reasoning at work within the institution of slavery. The
slave system, like all inegalitarian systems, was publicly underwritten
with claims to significant differences between anglo-Americans and Africans that warranted differential treatment of the two groups.
At this point, it may well seem that the CI procedure will not be
useful in assessing unjust inegalitarian systems that are predicated on
an ideology of difference. That is, it will not reveal the impermissibility of compliance with socially underwritten unjust practices. This is
because the problem appears ultimately to be one of mistaken factual
beliefs. Had the proponents of slavery gotten their facts straight, many
might have concluded that slavery was morally impermissible. The CI
procedure cannot do this work. Or can it? I will return to this question
in the next section. 22

22. One last reason for not universalizing the slaveholding maxim across all rational beings
deserves mention. The CI procedure may have more persuasive force when we imagine
a world where all whites, rather than all persons, adopt a slaveholding maxim. Recall
that the CI procedure asks the agent to think from the standpoint of a hypothetical
world. But why would considerations about what happens in a purely hypothetical
world be persuasive? Why, for example, would an agent be moved by the thought that
her lying promise would be laughed at in a hypothetical world of universal false promising? The hypothetical worlds of the promise-keeping and beneficence examples, I suggest, move us because they do not seem that distant. We understand all too clearly the
motives prompting us to make false promises or deny aid. And we understand these as
deeply human motives. The CI procedure implicitly reminds us that others share our
same motives and thus that the practices of promising and mutual aid are fragile ones,
dependent for their existence on others exercising restraint. The CI procedure makes
salient the fact that what we really want, in making false promises or denying aid, is that

170

K A NT A ND COMPLI A NCE W ITH INJUSTICE

III. R ESISTA NCE

I turn now to the task of reconstructing the CI procedure. I have argued
that when there are socially established rolesfor example, promisorpromisee or racially differentiated rolesproposals for action within
one of those roles should be universalized across the restricted category
of fellow role occupants. This interpretation of the CI procedure looks
as though it will have disastrous consequences for arguments against
slavery. It means that, in applying the CI procedure, a white agent asks,
Could I will that my maxim of enslaving Africans holds as a universal
law of nature for all whites? The answer, it would seem, is yes. There is
no contradiction in conceiving a world where whites universally employ
racially based slavery as the method of choice for advancing their interests. This dismaying conclusion is what motivates the insistence on
universalizing across all rational beings and stipulating that slaves are
permanently propertyless persons. Such insistence seems our only hope

others continue exercising restraint, but that we be permitted to act self-indulgently. In

short, the persuasive power of the CI procedure rests, at least partially, on the psychological plausibility of imagining that others behave as we do.
Is a hypothetical world where all adopt the policy of enslaving any others within
reach psychologically plausible? What human psychology would make such a world
possible? In order to imagine all enslaving any others within reach, as though by a law of
nature, we must impute a general psychological predisposition to regard other persons
as mere things, potential items of property, nonpersons. Perhaps Hobbes could imagine this. Kant, I think, could not. Imagining humans like this means imagining humans
who are completely resistant to all moral claims made on them by others. That we are
constitutionally inclined to deceive or to refuse aid to others whom we recognize as rational beings like ourselves is imaginable. In neither case do we regard others as devoid
of humanity. That we are constitutionally inclined to treat them as mere propertythings, fully devoid of humanity, is not. In other words, it is one thing to imagine a psychology vulnerable to temptation (as in the promise-making and beneficence cases). It
is another to imagine a psychology on which the humanity of others can make no mark.
The case is quite different if we imagine one racial, cultural, or gender group
adopting the policy of enslaving a different racial, cultural, or gender group. Universalization across whites, for example, would not require that we impute a general incapacity to acknowledge the humanity of persons. We would need only to impute a
human predisposition to mistake accidental differences in appearances, culture, etc.,
for morally relevant differences. That psychological assumption is, regrettably, all too
plausible.

171

MOR A L A IMS

for defeating slavery maxims. But perhaps the problem lies neither in
restricting universalization nor in loosening the definition of slavery.
Perhaps the problem lies in our misunderstanding of what the CI procedure asks us to imagine. Perhaps there is a contradiction in conceiving
universal white action on slave maxims that we are just not seeing because we havent gotten the CI procedures hypothetical world correctly
pictured.

Promises, Promises
It will be useful to look more carefully at Kants promising example.
What is it that makes universal promise breaking inconceivable? One
might think something like this: the practice of promising is, by definition, a practice of following certain rules, including the rule Keep
your promises. It is thus like games. A particular game, such as chess,
is a specific game in virtue of the very rules that constitute it. To alter
the rules is to destroy that game; a game in which the castles could be
moved diagonally would not be chess. Analogously, altering the rules of
the practice of promising would destroy the practice.23 Thus, a world of
universal promise breaking is inconceivable. There would, by definition,
be no practice of promising in such a world.
I think this is the wrong account. First, it does not look like Kants
account. Consider what Kant says:
the universality of a law which says that anyone believing himself
to be in difficulty could promise whatever he pleases with the intention of not keeping it would make promising itself and the end
to be attained thereby quite impossible, inasmuch as no one would
believe what was promised him but would merely laugh at all such
utterances as being vain pretenses.24
23. In describing the Logical Contradiction interpretation of the contradiction-inconception test, Korsgaard (Kants Formula of Universal Law, 30) suggests a view like
this of why universal false promising would destroy the practice of promising: One
may generate the contradiction by saying that when this [universal violation of the rules
of promising] happens the practice has new rules and becomes a different practice.
24. Kant, Grounding, 31 [422] (emphasis mine).

172

K A NT A ND COMPLI A NCE W ITH INJUSTICE

And:
For by such a law there would really be no promises at all, since in
vain would my willing future action be professed to other people
who would not believe what I professed, or if they over-hastily did
believe, then they would pay me back in like coin. 25

Kants stress is not on the practice-constituting nature of a rule to keep

promises. Rather, he focuses on the cooperative nature of promise
making and accepting. Promisees cooperate with promisors. They rely
on the promisors word and make present plans on the basis of faith in
the promisors future performance. When the word of promise makers
cannot be counted on, promisees have no reason to cooperate. Instead,
they have reason to resist participating in the practice of promising. As
potential promisees, they laugh at promises rather than accept them.
And as promisors, they pay promise breakers back in like coin.
If Kants argument does not rest on an analogy between promising and games, but instead on the predicted resistance of potential
promisees to cooperating with a practice of promising, then there will
be nothing inevitable, or analytic, about the breakdown of the promising practice in a world of universal promise breaking unless we build
in some additional assumptions.26 Onora ONeill has suggested that
we must at least assume that humans learn from experience, and thus
that a normal and predictable consequence of persons systematically
acting on a policy of false promising will be the collapse of trust in promises.27 This assumption by itself, however, is insufficient. Humans undergo lengthy socialization processes in childhood that are reinforced
throughout their adult lives; and they live in social worlds that secure
conformity to social conventions and norms through sanctions against
noncompliance. Socialization and sanctions can be structured so as to
prevent unjust policies causing a breakdown of trust and provoking
25. Ibid., 15 [403] (emphasis mine).
26. An observation that Leslie Mulholland also makes, though for different reasons, in
Kant: On Willing Maxims, 98.
27. ONeill, Acting on Principle, esp.78 and 81.

173

MOR A L A IMS

resistance. That is, they can be used to ensure that the normal and predictable consequence of systematic action on an unjust maxim is trust
in and cooperation with those who act unjustly. 28
To see this, consider a possible world in which promises might be
falsely made. People in this imaginary world belong either to a low caste
or a high caste. Low-caste people believe they are naturally inferior to
high-caste people. Because of this, they deny their entitlement to have
their interests considered by high-caste persons. Also, for them, the
noblest virtue lies in their complete faith in the veracity of high-caste
persons regardless of how high-caste persons in fact behave. Through
unwavering faith in high-caste promises, they rise above their ignoble
natureor so they think. But, more realistically, they know that should
they refuse to accept a promise, they will be punished in various ways.
They will be socially shunned, made the butt of jokes, denied the economic largesse of high-caste persons, and the like. Now let us suppose
that high-caste persons routinely make lying promises to low-caste
persons. In this world, the practice of promising does not break down
because low-caste persons have reason to continue cooperating. Continued participation in the practice enables them to prove their virtue
and to avoid penalizing sanctions.29
What shall we conclude? Kants CI procedure evidently requires
that we not imagine potential promisees like these low-caste promisees
when we universalize a maxim of false promising. If universal promise
breaking is to be inconceivable, then we must make further assumptions
about potential promisees. First, we must imagine that they take themselves to be our equals. They do not believe that they are inferior in ways
28. For a chilling but compelling analysis of the factors that produce womens voluntary
compliance with their sexual inequality, see Marilyn Frye, In and Out of Harms Way:
Arrogance and Love, in The Politics of Reality: Essays in Feminist Theory (Berkeley, CA:
Crossing Press, 1983).
29. Lest one imagine that low-caste persons are not really accepting promises, consider an
example closer to home. Virtuous women were and are expected to have cooperative
faith in male goodwill, despite the alarming amount of battery, rape, murder, harassment, and discrimination that goes on. Is their faith unreal? I see no reason to suppose
so. People often hold views of other people that are contradicted by other beliefs that
they hold, particularly when they have an incentive to do so. This is what I am imagining the low-caste persons to be like.

174

K A NT A ND COMPLI A NCE W ITH INJUSTICE

that would warrant diminished consideration. Second, we must imagine them freed of coercive incentives to cooperate with false promisors.
They risk no social, economic, legal, or physical penalties for refusing to
cooperate. Third, we must imagine them to be rational in the sense that
they hold no false factual or moral beliefs about themselves or others.
They do not, for example, suffer from misguided notions about virtue.
Finally, we are to imagine that the only reasons they have for participating in cooperative practices are the reasons that fully free, equal, and
rational beings would have.
Looked at this way, what the CI procedure invites the would-be
promise breaker to do is to contemplate how fully free, equal, and rational beings would respond to conditions of universal promise breaking. Unlike the imagined low-caste persons, free, equal, and rational
beings have both reason and freedom to laugh in the face of false promisors. They have both reason and freedom to repay false promisors in like
coin. In short, they see the irrationality of complying with a practice
that can only undermine their own interests; and they have the power
to resist compliance. Among such beings, the practice of promising will
inevitably break down under conditions of universal promise breaking.

The Resistance of the Enslaved

We are now in a position to see why, when we asked Can I will that
my African slaveholding maxim hold as universal law for all whites?
this maxim appeared easily to pass the CI procedure. We were reasoning as though the enslaveds characteristicsas fully free, rational, and
equal beingswere irrelevant to the CI procedure. Indeed, we didnt
think about the enslaved at all. We simply imagined whites as universally adopting a slaveholding maxim. And so we were led to focus on
the question of whether all whites behaving as the agent proposes would
make the agents own proposed action impossible. (Clearly not.) Or, to
the extent we thought about the enslaved at all, we imagine their being
exactly the sorts of persons that life under the institution of slavery
would have made them: powerless, penniless, fearful, and so on. And
so we were led to focus on the question of whether all slaveowners behaving as the agent proposes would, via the enslaved persons response,
175

MOR A L A IMS

make the agents own proposed action impossible (Again, clearly not.)
In short, we assumed that universalization itself was all that mattered.
But, I have suggested, the CI procedure asks us to do more than universalize the maxim. It asks us to consider how fully free, equal, and rational
beings would respond to the conditions created by universalizing the
maxim. 30 Would universal (white) action on slaveholding maxims meet
with such resistance from fully free, rational, and equal (black) persons
that the practice of slavery would break down? Would black persons
laugh in the would-be slaveholders face or pay him back in like coin?31
We are also now in a position to see why a maxim of convenience killing appeared easily to pass the CI procedure. We, following
Herman, imagine that the CI procedure only asks us to universalize
the convenience-killing maxim. Because we did not consider that there
might be constraints on how we are permitted to envision the persons
among whom this maxim is universalized, we proceeded to imagine
humans as we now know them: as beings who often lack the physical
and technological resources to avoid being killed. They are beings who
cannot help but be vulnerable to the convenience killers plans, even if as
rational beings they reject being so used for anothers purposes. In short,
they are not empowered to engage in rational resistance to the conditions created by a universal maxim of convenience killing. They lack the
power to refuse to cooperate in the procurement of their own convenient
30. Ronald M. Green, The First Formula of the Categorical Imperative as Literally a Legislative Metaphor, History of Philosophy Quarterly 8 (1991): 163179, 165, gives a similar reading of the Formula of Universal Law as stating, Act only on that maxim which
you are rationally able to will as being approved, accepted, and acknowledged by all
other rational agents as a law governing everyones conduct. Ultimately he collapses
the first and third formulations. I have tried not to.
31. Kant, interestingly, attributes progress toward a fully legal condition, as well as human
self-development generally, to the resistance and war with which the selfish indulgence
of inclinations is met: [man] finds in himself the unsociable characteristic of wanting
everything to go according to his own desires, and he therefore anticipates resistance
everywhere, just as he knows about himself that for his part he tends to resist others.
Now this resistance awakens all of mans powers, brings him to overcome his tendency
towards laziness, and, driven by his desire for honor, power, or property, to secure
status among his fellows; Idea for a Universal History with a Cosmopolitan Intent,
in Perpetual Peace and Other Essays, 32 [2021]. See also his To Perpetual Peace: A
Philosophical Sketch, in Perpetual Peace and Other Essays, 120 [360] ff.

176

K A NT A ND COMPLI A NCE W ITH INJUSTICE

(to others) deaths. Run on this kind of being, the CI procedures result
reflects, not the rational choice of responsive agents, but their helplessness. If instead we build into the CI procedure the requirement that all
rational beings be empowered to resist, if they so choose, the conditions
created by universalizing a maxim, the convenience-killing maxim will
not pass the CI procedure. Convenience killing will be impossible in a
world where everyone effectively resists making him or herself vulnerable to murder, in just the way that false promises will be impossible in a
world where everyone resists making him or herself vulnerable through
accepting promises. Interpreting the CI procedure this way, there will
not be a significant difference between conventional and natural acts. 32
Analogously, in the case of slaveholding maxims, what the CI procedure asks the agent to focus on is not so much universal (white) compliance with slaveholding maxims as the empowered status of would-be
slaves. The agent is barred from factoring in his own groups social, legal,
economic, and physical power to force African compliance. Indeed, he
must instead imagine anglo-Americans confronting Africans who have
full access to the social, legal, economic, and physical resources that
would enable them to resist if they so choose. The agent is also barred
from factoring in those beliefs (for example, in black inferiority) and
those personality traits (for example, fear and deference) that are produced in Africans through socialization into a slave society. Instead, he
must imagine anglo-Americans confronting Africans who are psychologically free to make up their own minds about what their status is vis-vis anglo-Americans. (In the last section, I will take up the question
of how the formula of universal law now differs from the formula of the
kingdom of ends.)
32. In Kants Formula of Universal Law, Korsgaard also argues that natural acts can be
dealt with by the CI procedure. She rests her argument, however, on the supposition
that the agent will not be able to accomplish his own purposes when he makes himself
the target of the same sort of treatment. I am suggesting a reading where universalization can be understood as generating both a practical and a logical contradiction by
stressing that the agents action is not possible in a world where others refuse to cooperate. Korsgaards reading commits us to finding only a practical contradiction. My interpretation is closer to Onora ONeills, in Acting on Principle, chap.5. ONeill stresses
the way that rational agents would respond to the conditions created by universalizing
ones maxim so as to make the proposed act impossible.

177

MOR A L A IMS

A number of comments about this construction of the CI procedure are in order. First, universalization appears to play a much less
important role than either Kant or Kant interpreters typically give it.
Is that a fault? I think not. I have been arguing that a contradiction in
conception does not result simply from universalizing an impermissible
maxim. It results from free, equal, and rational beings response to the
conditions created by universalizing an impermissible maxim. The CI
procedure does its work by getting an agent to think about free, equal,
and rational beings response. Now, when an agent excepts herself from
a conventional, just practicefor example, she makes false promises
universalizing her maxim enables her to get a clearer view of exactly
how free, equal, and rational beings would respond to her maxims becoming universal law. Universalization works because it compensates
for the fact that real-world persons response to promisors, including
false promisors, is shaped by existing, general compliance with promising conventions. Universalizing is, in this case, a useful technique for
testing the fitness of ones maxim for universal law. But it is only a technique, and is not essential to the Formula of Universal Law itself. (Consider, publicizing a maxim of noncompliance would do the same work as
universalizing it. 33) In the case of compliance with conventional, unjust
practices, we need a different technique for making salient the response
of free, equal, and rational beings to ones maxim becoming universal
law. The real-world Africans responses to slavery maxims were shaped
by socialization and coercion into compliance. Mere universalization
will not compensate for this. A requirement that the agent discount the
effects of socialization and coercion by imagining fully free, equal, rational, and empowered persons will.
Second, the CI procedure, constructed this way, will work even
given fairly high levels of factual misinformation about the nature of
persons in the social groups between which the cooperative scheme is
to operate. This is because the response that matters in the CI procedure
is the response based on the other groups own conception of who they
33. For an interesting discussion of Kants use of a publicity technique, see Kevin R.
Davis, Kants Different Publics and the Justice of Publicity, Kant-Studien 83 (1992):
170184.

178

K A NT A ND COMPLI A NCE W ITH INJUSTICE

are and what their ends are. Thus, much of the worrisome information
that proponents of slavery invoked to justify the institution would be
ruled off-limits. The agent may not, for example, make judgments about
how members of the other group ought to respond given her own beliefs
about their nature. (She may not, for example, judge that potential slaves
ought to want paternalistic care, given her own belief in their incapacity to take care of themselves.) Nor may she use knowledge about the
other group generated solely by members of her own group as a reliable
basis for predicting their response. (She may not, for example, claim
to know that Africans are less intelligent if the only studies of racial
intelligence have been conducted by anglo-Americans.) Thus, the CI
procedure places a severe constraint on any justifying appeal to the consideration this is in accordance with natural (or acquired) differences
between.... One may not make such an appeal when the differences
alleged to justify the practice in question have been forwarded, verified, and endorsed by only one party to the practice, or would not have
been endorsed by both parties absent intensive socialization to hold just
those practice-justifying beliefs. Reasoning on the basis of appeals to
difference is allowable only when both parties freely endorse the differences asserted.
One might object that, deprived of so much information, no agent
could successfully run the CI procedure. How is she supposed to predict
the response of free, rational, and equal Africans if virtually everything
she knows about them is ruled out? Worse, because the Africans in
her slave society are themselves the product of a slave society, she apparently cannot use their real-world responses as a guide. Two replies
are possible here. On the one hand, this problem plagues any style of
moral reasoning conducted within unjustly inegalitarian societies. Both
official bodies of knowledge and the self-narratives of the oppressed
are likely to be distorted. On the other hand, there are perhaps more
informational resources available than first seem. Agents can ask the
kinds of questions that might make them deeply suspicious that cooperation, if it occurs, results from free, rational choice. For example, who
benefits the most from this arrangement? How intense and continuous
are the socialization processes into this arrangement? How serious are
the sanctions for failure to cooperate? Who controls production of the
179

MOR A L A IMS

knowledge that is used to justify the arrangement? Agents can also pay
attention to actual resistance: Is there wholesale revolt (e.g., running
away, murdering overseers)? Is there subversive resistance (e.g., breaking tools, feigning illness, playing stupid)? How deep is the cultural
anxiety over the possibility of resistance, and how much cultural energy
goes into explaining away resistance (e.g., attributing slaves running
away to the disease of drapetomania)? Have separatist cultures developed from whose epistemic and normative standpoint conventional
social practices would clearly not be endorsed?34
In sum, applying the universal law formula requires concrete and
socially critical knowledge of how members of other groups conceive of
their ends, capacities, and status vis--vis the agents group; of the socialization techniques and coercive factors (both institutional and individual) that may distort their conception of their own ends, capacities, and
status; and of oppositional cultures35 and resistance strategies created
by that group. In the absence of such knowledge, it is not possible to determine whether a maxim of compliance with a conventional practice
that affects other social groups is conceivable as universal law, and thus
rationally justified.

IV. R EPLIES TO OBJECTIONS

Although the reconstruction I have offered enables us to frame nonquestion-begging arguments against inegalitarian practices, and provides the deliberating agent with guidelines for engaging in more
socially critical reflection, this reconstruction is not trouble free. I

34. Consider how answering these same questions also enables one to show why some inegalitarian relationshipse.g., between minister and parishioner or between teacher
and studentare morally permissible. Standard rules for ministerial interaction with
parishioners presumably would not meet resistance from free, rational, and equal parishioners. (By contrast, cult rules are taken to be ones that would meet resistance had
cooperation not been secured via manipulative and coercive techniques.)
35. The term oppositional culture is taken from Ann Ferguson, Is There a Lesbian
Culture? in Lesbian Philosophies and Cultures, ed. Jeffner Allen (Albany: SUNY
Press,1990).

180

K A NT A ND COMPLI A NCE W ITH INJUSTICE

conclude by briefly taking up some of the more serious objections that

might be raised.

Universal Law versus the Kingdom of Ends

Given Kants claim that the three formulas articulate the same Categorical Imperative, any reconstruction of a Kantian formula should enable
us to see how that formula is connected to and anticipates the other formulas. But there is an attendant risk of collapsing formulas so that they
lose their integrity as distinct formulas. In reconstructing the formula of
universal law, I have stressed the idea that maxims must be acceptable to
other free, equal, and rational beings, thus bringing this formula closer to
the formula of the kingdom of ends. Is it now too close?
If the demand that ones maxim be acceptable to free, equal, and
rational beings distinguishes the formula of the kingdom of ends from
the formula of universal law, then I have indeed collapsed the first into
the third formula. But one need not agree to this way of distinguishing
thetwo formulas. What distinguishes them may not be that the third
formula stresses acceptability to free, equal, and rational beings, but how
it does so.
In the formula of the kingdom of ends, acceptable to free, equal,
and rational beings must be understood within the context of this formulas specific procedure for generating acceptable principles; the positive assumption that free, equal, and rational beings are also ends in
themselves; and the equation of normative validity with mutual acceptability. From the co-legislative standpoint of the kingdom of ends, each
member reasons from the explicit assumption that every member is an
end in himself. This means among other things that they have a prima
facie concern to see each persons ends realized or at least to ensure each
person freedom to pursue his ends.36 It also means that from the colegislative standpoint, mutual acceptability is taken to be what validates
principles. According to Rawlsian interpretations of the third formula,
mutual acceptability is secured by a procedure of abstracting from
36. Thomas E. Hill Jr., The Kingdom of Ends, in Dignity and Practical Reason in Kants
Moral Theory (Ithaca, NY: Cornell University Press, 1992), 61.

181

MOR A L A IMS

individuating personal differences. Co-legislators do not reason from

the perspective of their own individuating characteristics (they stand
behind a veil of ignorance), and they do not legislate special rights and
duties attached to individuating characteristics (such as gender). 37
The formula of universal law, as I have reconstructed it, neither employs a decision procedure that requires abstracting from differences
nor presupposes that free, equal, and rational beings are also ends in
themselves, nor equates normative validity with mutual acceptability.
The agent does not reason as a co-legislator committed to endorsing
only mutually acceptable principles. Mutual acceptability is the outcome, not the aim, of applying the formula of universal law. Not aiming
for mutual acceptability, the agent does not adopt a Rawlsian original
position, deprived of knowledge about her actual ends, individuating
characteristics, evaluative and factual beliefs, and the like. Instead,
she is committed to endorsing only those maxims she could conceive
or will that she act on, given their universalization among free, equal,
and rational beings who may have reason not to cooperate. Thus, she
reasons as someone committed to pursuing her actual ends, but amid
others who have the capacity to find cooperating with her irrational and
thus to resist her plans. Because the first formula asks the agent whether
she could will her maxim as universal law once artificial, contingent
barriers to resistance have been removed, she need not ignore her particular ends. Nor need she ignore her own, possibly inegalitarian and
prejudiced conceptions of individuating differences. This is to say that
she is not required to engage in the kind of abstraction required of colegislators. What she must do is (1) pay attention to what enables her
to pursue those ends and act on her own conception of individuating
differences without meeting the kind of resistance from rational beings
that would make action on her maxims impossible; and (2) abstract out
all nonrational factors that might influence others cooperation. In the
promising example, she must abstract out the state of general compliance with the promising practice that creates trust in promisors. In the
slavery example, she must abstract out coercive, resistance-discouraging
sanctions against noncooperation, as well as socialization processes that
37. Ibid., 60.

182

K A NT A ND COMPLI A NCE W ITH INJUSTICE

artificially produce others agreement to her ends and her conception of

racial differences. Thus, far from abstracting from differences, the first
formula reveals the potential for differences and contestatory conflict
among free, equal, and rational beings.

Illegitimate Resistance
A resistance-based moral procedure appears to suffer from the same
defect as a consent-based moral procedure. Consent by itself does not
guarantee the validity of the norms consented to. We can consent to
things we ought not. Similarly, resistance by itself does not guarantee
the invalidity of the norms resisted. We can resist what we ought not
for example, punishment for breaking what are in fact legitimate laws.
If consent is to have validating force or resistance to have invalidating
force, it would seem that we would need to bring the third formula and
something like the original position back into the picture, asking what
fully impartial agents would consent to or resist.
To see why this is not necessary, consider first the case of punishment. The concern here is that fully empowered lawbreakers would be
able to resist punishment and thus, it would seem, a maxim of punishing lawbreakers could not pass the contradiction-in-conception test as
I have formulated it. But the question to be asked is not whether individuals would resist their being punished on the particular occasion of
their breaking a law; rather, it is whether individuals would resist a policy
of punishment that could apply to them. Individuals who become lawbreakers may well have reason to accept a system of punishment because
that system also works to their advantage (they may themselves have
reason to want the protection of a penal system). They do not, however,
have reason to accept just any system. Individuals who are poor, black,
and male constitute a disproportionate number of prison inmates. They
may have good reason to resist a policy whose only response to crime
is via penal sanctions. Similarly, those who are incarcerated may have
reason to resist specific penal policies that permit brutality, overcrowding, denial of visiting privileges, and so on. As I imagine the CI procedure, what it precludes is my endorsing a policy of punishment when
cooperation with that policy has been secured by nonrational means
183

MOR A L A IMS

and would not occur apart from those nonrational influences. What it
requires is consideration of the possibility that those who are most vulnerable to penal sanctions may be of a different mind concerning the
existence and details of a penal system than those who pride themselves
on being law-abiding citizens. Riots, rebellions, hunger strikes, civil
disobedience, and separatist movements are all resistance strategies to
policies. And it is thought about the possibility of such resistance (not
about the individual inmate making his prison break) that the CI procedure is designed to evoke.
Of more concern, I think, is the potential for mutual resistance. For
instance, while racially based enslavement maxims would, if universalized among free, equal, rational, and empowered beings, meet resistance
that would make action on them impossible, obviously so would egalitarian maxims in a racist society. Both our own and South African history bears this out. One would like to be able to say that resistance based
on egalitarian considerations has a greater claim to legitimacy.
But perhaps it is a virtue of the CI procedure as I conceive it that
it does not permit judgments about the legitimacy or illegitimacy of
particular cases of resistance. In social contexts that generate mutual
resistancefor example, where both inegalitarian and egalitarian conceptions of racial categories hold swayboth maxims predicated on
racial inequality and maxims predicated on racial equality would be
nonwillable as universal laws. This is, I think, as it should be. In societies where there is very deep conflict, any decision procedure that is as
formal as Kants CI procedure will not be able to resolve conflict over
substantive claims directly by validating, say, one conception of racial
categories rather than another. What it can do is force agents to move
to consideration of principles that do stand a change of being willable as
universal law under conditions of social conflict. Those are likely to be
principles of fair negotiation and conflict resolution.

Begging the Question

I have argued against an interpretation of universalization that would
require agents proposing a slaveholding maxim to ask whether the enslaved could also adopt their maxim. That interpretation assumes the
184

K A NT A ND COMPLI A NCE W ITH INJUSTICE

very thing that must be proved: that a particular inegalitarian system

of different rules for different parties is illegitimate. But havent I done
something similar by requiring agents to envision the Other as a free,
rational, and equal being? Arent advocates of slavery being required to
abandon their belief that Africans are childlike or animal-like and to
adopt a more egalitarian perspective at the outset?
No. What they are asked to do is to consider whether the success of
the cooperative schemes they propose rests on the contingent success of
socialization techniques and social sanction in producing in others (in
this case, enslaved Africans) a belief in their own inequality and a belief
that cooperation rather than resistance best serves their interests. The
assumptions of equality and freedom are negative requirements. Agents
may not assume that the Other would agree with their belief that the
Other is unequal absent the strategies currently employed to produce
agreement. Nor may they assume that the Other would cooperate once
contingent sanctions against noncompliance have been removed. The
assumption of rationality does not extend beyond that already being
made in using socialization techniques and coercive threats to secure
cooperationnamely that the Others are capable of holding beliefs
about themselves and their interests, and of acting on the basis of those
beliefs. Unlike the second and third formulas, the formula of universal law does not bring in freedom, rationality, and equality as ways of
characterizing beings with a special moral status (namely ends in themselves). They are brought in only insofar as they are necessary for the
agent to determine whether she can guarantee the success of her cooperative scheme and thus its fitness for universal law.

185

C ha pt e r7

Responsibility and Reproach

Feminist consciousness is often afflicted with category confusion

an inability to know how to classify things.
Sandra Lee Bartky

Feminist thinking about moral responsibility for oppressive and sexist

behavior illustrates just this kind of confusion. When wrongdoing
takes the form of social oppression, the relationship between individuals and their actions shifts in ways that render uncertain our judgments
about moral responsibility and, with those, our judgments about the
blameworthiness of individuals and our entitlement to reproach them.
Part of the uncertainty about how to assign moral responsibility derives
from the atypical character of the wrongdoing that feminists critique.
Unlike ordinary cases of individual wrongdoing, oppressive wrongdoing often occurs at the level of social practice, where social acceptance
of a practice impedes the individuals awareness of wrongdoing. Thus
determining moral responsibility raises very difficult questions about
how to weigh the social determinants producing moral ignorance
against the individuals competence to engage in moral reasoning. The
social scale at which oppression occurs complicates thinking about
moral responsibility in other ways as well. If we assume, as we often
do, that only morally flawed individuals could act oppressively, then we
will have to conclude that the number of morally flawed individuals is

186

R ESPONSI BI LIT Y A N D R EPROACH

vastly larger than we had dreamed and includes individuals whom we

would otherwise rank high on scales of moral virtue and goodwill. The
oddity of this conclusion forces serious questions about the possibility of morally unflawed individuals committing serious wrongdoing.
Finally, when wrongdoing occurs at a social levelthat is, at a level
that places whole social groups at risk, and where wrongdoing stands
a much greater chance of being perpetuated by its very normalcyour
moral and personal stake in intervening in the pattern is much higher.
The question of blame becomes not just one of blameworthiness but,
more importantly, one of our entitlement to use moral reproach as a
tool for effecting social change. Does the justified use of moral reproach
require, as it does in ordinary cases, being justified in assigning moral
responsibility and blameworthiness? In what follows, I want to explore
the differences between ordinary cases of wrongdoing and oppression
as a way of clarifying why questions about moral responsibility for oppressive behavior are so difficult to resolve. In the course of that exploration, I will argue, first, that the central difference lies in the normality
versus the abnormality of the context in which wrongdoing occurs;
second, that oppressive behavior need not proceed from some morally
culpable flaw; and third, that in what I will be calling abnormal moral
contexts, our entitlement to use moral reproach is independent of the
blameworthiness of individuals.

I. LOCATING CONFUSION A BOUT MOR A L

R ESPONSIBILIT Y
For women who have had even the slightest exposure to feminist
thinking, confusion over moral responsibility arises at an immediate,
personal level in the form of contradictory reactive attitudes. Condemnatory attitudes (anger, indignation, contempt) war with exculpating attitudes (forgiveness, tolerance, pity). Confused emotions and confusion
over what one ought to feel stem from a kind of double vision. On the
one hand, feminist consciousness is a consciousness of victimization.
To apprehend oneself as victim is to be aware of an alien and hostile

187

MOR A L A IMS

force which is responsible for the blatantly unjust treatment of women

and for a stifling and oppressive system of sex roles; it is to be aware, too,
that this victimization, in no way earned or deserved, is an offense.1
The level and extent of offense seem to call for reactive attitudes
predicated on moral responsibility. Yet at the same time, feminist consciousness is a consciousness of inhabiting a new reality of seeing what
one did not, could not, see before and what others still do not, cannot,
see. The social invisibility of offense seems to call, by contrast, for more
tolerant reactive attitudesones predicated on excuse.
In feminist literature, the language of oppression iterates the same
tension between responsibility and excuse. The sheer volume of oppressive, exploitive, sexist, and power-abusing practices documented by
feminists seems to call for a suitably large number of culpable agents.
Indeed, talk about oppression, exploitation, sexism, and power relations implicitly points to oppressors, exploiters, sexists, and abusers of
powerthat is, to the responsible villains. Yet the villains are mostly
ordinary men, with ordinary characters, living out ordinary lives as husbands, scientists, ad men, construction workers, and so forth. Pimps,
porn magnates, rapists, and self-proclaimed chauvinists are exceptional
figures in womens oppression. Without the ordinary mans participation in routine social practicesin marriage, in the workplace, in daily
conversationoppression would not take the universal form it does.
But the disanalogy between pimps and husbands may make us hesitate
to name both oppressors. The ordinary man may act oppressively, but
is he an oppressor? How do we locate individual responsibility when oppression occurs at the level of social practice?
There are no quick and easy answers. On critical reflection, neither
holding individuals responsible for their participation in oppressive
social practices nor excusing them seems an appropriate response. Why
this is so will be developed in sections II and III, but it will help to begin
by turning a critical eye on some of the arguments for responsibility and
for excuse.
1. Sandra Lee Bartky, Toward a Phenomenology of Feminist Consciousness, in Feminism
and Philosophy, ed. Mary Vetterling-Braggin, Frederick A. Elliston, and Jane English
(Totowa, NJ: Littlefield, Adams, 1981), 2627.

188

R ESPONSI BI LIT Y A N D R EPROACH

Ruth Bleier blames scientists for some actions done in the course of ordinary science:
However unreflective the process may be, scientists, such as those
[who document womens deficiency in visuo-spatial skills]
are able to stop just short of making the kinds of assertions that
their own and others data cannot defensibly support yet they can
remain secure in the knowledge that their readers will supply the
relevant cultural meaning to their text; for example, that women
are innately inferior in the visuo-spatial and (therefore) the mathematical skills, and that no amount of education or social change
can abolish this biological gap. It is disingenuous for scientists to
pretend ignorance of their readers beliefs and expectations and
unethical to disclaim responsibility for the effect of their work
and for presumed misinterpretations of their pure texts. Scientists are responsible, since they themselves build ambiguities and
misinterpretations into the writing itself. 2

In blaming scientists, Bleier uses the language of agency

disingenuous, pretending ignorance, building ambiguities and misinterpretations in, but the disclaimer however unreflective the process
may be undermines her charge, especially since Bleier notes that the
practice of science requires not reflecting on normative implications.
If doing normal science means not engaging in moral reflection, why
castigate individual scientists for pretending ignorance or for building
misinterpretations into their literature? Only if they were doing science
abnormallythat is, in a morally reflective waywould they be in a position to feign ignorance or to deliberately cultivate misinterpretation.
In a somewhat different vein, Mary Daly argues vehemently in Gyn/
Ecology for universal male responsibility, pointing out that a refusal to
name the enemy and speak in the active voice covers up the fact that it
is men, rather than abstract forces, who oppress women and benefit from
2. Ruth Bleier, Lab Coat: Robe of Innocence or Klansmans Sheet? in Feminist Studies/
Critical Studies, ed. Teresa de Lauretis (Bloomington: Indiana University Press,
1986),62.

189

MOR A L A IMS

it. She says, Despite all the evidence that women are attacked as projections of the Enemy, the accusers ask sardonically:
Do you really think that men are the enemy? This deception/
reversal is so deep that womeneven feministsare intimidated into Self-deception, becoming the only Self-described oppressed who are unable to name their oppressor, referring instead
to vague forces, roles, stereotypes, constraints, attitudes,
influences. This list could go on. The point is that no agent is
namedonly abstractions. 3

Naming only abstract forces as agents veils the fact that individuals play
roles, impose constraints, hold attitudes, and thus cause harm.
Causing harm, however, is not the same as being responsible for
harm.4 Ordinarily, when there is evidence of strong conditioning, the
individual is excused from responsibility, and Daly acknowledges just
that conditioning. Oppressive social practices, she remarks, are acted
out over and over again in performances that draw the participants into
emotional complicity. Such re-enactment trains both victims and victimizers to perform uncritically their preordained roles.5 If this is so,
why does she refuse to excuse men? Dalys resistance to accepting excuses may derive from a belief that the logic of the language of moral
responsibility changes when the moral picture includes participation
in immoral social practices and the power to legitimize those practices.
This is a possibility worth pursuing because, however wrong blaming
individuals for publicly unacknowledged faults in a social practice may
seem, not blaming them seems equally wrong.
In The Man of Professional Wisdom, Kathryn Pyne Addelson
argues that prestige structures in science enable some researchers to exercise cognitive authority over the direction that scientific theorizing
3. Mary Daly, Gyn/Ecology: The Metaethics of Radical Feminism (Boston: Beacon, 1978),29.
4. Here, I am not speaking for Daly but simply voicing a common distinction in moral
theory. I suspect that Daly reconstructs the notion of responsibility. There is some suggestion that she uses causing harm and benefiting from it as a sufficient condition for
responsibility.
5. Daly, Gyn/Ecology, 109.

190

R ESPONSI BI LIT Y A N D R EPROACH

takes.6 That authority can be misused to deprive competitive work of

serious attention, to cut off resources for competitive work, and ultimately to irrationally bias the development of science.7 Of specifically
feminist concern is the use of cognitive authority to stunt female scientists careers or to perpetuate male bias in scientific theories. In illustrating prestige structures, Addelson describes how Robert Yerkes used his
cognitive authority to promote the career and theories of Clarence Ray
Carpenter. But she says,
it would be a mistake to describe Yerkes as showing favoritism
and bias.... Researchers are also the judges of which competing
theories it makes sense to pursue or to encourage others in pursuing. If this seems to result in bias, the way to correct it is not by
blaming individual researchers for showing favoritism because they
depart from some mythical set of abstract canons [the canon that
rational criticism requires competing theories]. 8

Yerkess and similar cognitive authorities ignorance of prestige

structures and of the importance of promoting competition excuses
their abuses of cognitive authority. Thus she suggests that the way to
avoid harmful uses of cognitive authority is to broaden rational criticism in science by requiring that both philosophers of science and scientists understand how prestige and power are factors in the way cognitive
authority is exercised.9 Moreover, we should institutionalize this sort
of criticism [feminist criticism] and make it an explicit part of scientific method. We should also try using the notion of cognitive authority
and expanding the range of the criteria of scientific rationality and criticism so that it includes social arrangements within the scientific professions.10 Addelsons shift into the passive voice (should be required)
and an anonymous we help her avoid confronting the questions:
6. Kathryn Pyne Addelson, The Man of Professional Wisdom, in Discovering Reality, ed.
Sandra Harding and Merrill B. Hintikka (Boston: D. Reidel, 1983).
7. Ibid., 178.
8. Ibid, 178179 (emphasis mine).
9. Ibid., 179 (emphasis mine).
10. Ibid., 182.

191

MOR A L A IMS

Required by whom? and Which we bears responsibility for making

what amount to substantive changes in the practices of science and scientific criticism? If we take seriously Addelsons claim that only those
with cognitive authority can effectively require new directions, it would
seem that they are responsible for critiquing and changing scientific
practice. Outsiders (like Addelson) can only hope that their critiques
are heard, taken seriously, and promoted. (Feminists in general bear the
status of outsidersoutsiders to the scientific, academic, and everyday
social communities, where their critiques are not simply nonlegitimated
but also actively delegitimated. They, too, can only hope that their critiques are heard, taken seriously, and promoted by insiders.)
The problem, then, with excusing is this: If the excuse for wrongdoing is the normalcy and social legitimacy of ones actions, this same
excuse can be repeated for ongoing resistance to seeing that there is anything wrong with what one is doing. If, for instance, Bleiers scientist
ought to be excused for his lack of moral reflection because he is just
doing science normally, he ought similarly to be excused for not accepting gender bias as a good reason for rejecting his theory, because here
again he would just be doing science normally. From his point of view,
rationality requires giving greater weight to beliefs held by a consensus
of rational knowersin this case, scientists; and it is not his fault if the
community of rational knowers happens to have reached consensus on
the wrong beliefs (e.g., the belief that a theorys normative implications
are irrelevant to its scientific merits). The problem with this perspective
is that unless individuals accept moral responsibility for the practices
in which they participate, and for the social justifications of those practices, individuals will lack a motive to listen to outsiders moral critiques.

II. NOR M A L V ER SUS A BNOR M A L MOR A L

CONTEXTS
In the above examples, the ordinary man acting oppressively, the scientist employing a morally unreflective writing style, and the prestigious scientist misusing his cognitive authority all suffer from a moral
ignorance shared by their compatriots and sustained by the normalcy
192

R ESPONSI BI LIT Y A N D R EPROACH

of their wrongdoing. The culpability of their wrongdoing thus turns

on the culpability of their moral ignorance. Ought they to have known
better?
Susan Wolf suggests that the freedom necessary for responsibility
is a freedom to be determined by the True and the Good. This requires
that the world cooperate in such a way that our most fundamental
selves have the opportunity to develop into the selves they ought to
be.11 In particular, it requires that the world enable us to become knowledgeable about the Good. But the world fails to cooperate in the above
cases. Unlike the pimp who violates public moral standards, these individuals are misguided by public standards of morally permissible action.
Moreover, the esoteric, often publicly inaccessible, and socially delegitimated nature of feminist moral criticism works against their acquiring,
or taking seriously, moral doubts about those public standards. Thus it
would seem that, because the world does not cooperate, their ignorance
is not culpable.
But this is too simple. These individuals possess (we may assume)
the full array of moral competencies that can be expected in any normal
adult agent. As average moral citizens, it would seem that they are responsible for applying accepted moral canons (e.g., against exploiting
others) to cases not covered, or incorrectly covered, in the social stock
of moral knowledge. Moreover, their ignorance is not simply due to an
uncooperative world. Their participation in oppressive social practices
helps sustain the social acceptance of those practices.
In order to move forward, we need to examine more closely how
the moral ignorance in these cases and the abnormal moral context in
which it occurs differ from more familiar forms of moral ignorance in
normal moral contexts. The moral ignorance at issue here clearly differs
from that typically appealed to in excusing wrongdoing. First, it occurs
at a social rather than an individual level; and second, it occurs in individuals with generally good moral reasoning skills. Exploring these
differences matters because our intuitions about moral responsibility
and warranted reproach typically presuppose a normal moral context.
11. Susan Wolf, Asymmetrical Freedom, Journal of Philosophy 77, no. 3 (1980):
151166,160.

193

MOR A L A IMS

When this presupposition breaks down, we may find ourselves forced to

give different answers to questions about responsibility and reproach.

Moral Ignorance in Normal Moral Contexts

In normal moral contexts, the rightness or wrongness of different
courses of action is transparent to individuals, where transparent does
not mean self-evident, but simply that participants in normal moral contexts share a common moral language, agree for the most part on moral
rules, and use similar methods of moral reasoning. We expect people
to be familiar with moral languageto know, for example, that promise, respect, polluting, and drunkenness are typically used in and
signal moral contexts and that mechanism, operating, and sleepiness do not. We also expect people to know what sorts of things count
as promises, signs of respect, and so forth, as well as the moral rules
and principles that go along with moral language. The sharing of moral
knowledge allows us to assume that most rational, reflective people
could come to correct judgments about which courses of action would
be right, wrong, or controversial; and this is what I meant by the transparency of normal moral context.
In such contexts, the idea that moral agents are self-legislators makes
sense. Put differently, moral individualismin the sense of a reliance
on the individuals ability to judge correctlyis a luxury affordable in
a morally homogeneous society where individual choices are likely to
concur, thus contributing to an institution of morality.12
The shared nature of moral knowledge and the self-legislating ability of moral agents in normal moral contexts determine the form moral
ignorance can take. First, it will be an ignorance of what the moral community in general knows. Thus in normal contexts, moral ignorance is
necessarily exceptional. This is why claims of moral ignorance make
such poor excuses. It would, for example, take a very special story for I
12. Self-legislation has sometimes been presented (for example, by Kant) as purely a function of rationality. One of my assumptions throughout this chapter is that philosophical theories need to take into account the sociological aspects of moral knowledge and
action. Thus I understand self-legislation to be a function of a particular kind of moral
society.

194

R ESPONSI BI LIT Y A N D R EPROACH

didnt know that polluting public waterways is wrong or I didnt know

that dumping thousands of gallons of toxic waste into public waterways
is polluting to be an acceptable excuse. To count as an excuse, one
would have to show that ignorance originated in some atypical defect in
moral education or moral development beyond the individuals control.
That moral ignorance is an ignorance of what others know means
that it is difficult to become ignorant in the first place, and even more
difficult to sustain that ignorance. Susan Wolf gives this example of ignorances etiology:
We imagined a case... of a man who embezzled some money, fully
aware of what he was doing. He was neither coerced nor overcome
by an irresistible impulse, and he was in complete possession of
normal adult faculties of reason and observation. Yet it seems he
ought not to be blamed for committing this crime, for, from his
point of view, one cannot reasonably expect him to see anything
wrong with his action. We may suppose that in his childhood he
was given no lovehe was beaten by his father, neglected by his
mother. And that the people to whom he was exposed when he
was growing up gave him examples only of evil and selfishness.
From his point of view, it is natural to conclude that respecting
other peoples property would be foolish. For presumably no one
had ever respected his. And it is natural for him to feel that he
should treat other people as adversaries.13

The embezzlers exceptional ignorance is obvious: Whereas our childhoods fell within a range of normal decency, his was severely deprived.14
Because moral ignorance will be hard to come by in societies where
moral knowledge is shared and constantly confirmed in daily interactions without some gross defect in the individuals moral education and
development, morally ignorant people will also likely suffer from a general inability to take the moral point of view and will be poor candidates
for moral agents.
13. Wolf, Asymmetrical Freedom, 159160 .
14. Ibid., 160.

195

MOR A L A IMS

A final point about moral ignorance. Not excusing moral ignorance

matters to morality in the same way that not excusing legal ignorance
matters to the legal system. Moral and legal rules place people under an
obligation to conform to those rules. But there is little point in having
either system of obligations if people can easily avoid the sanctions
against nonconformity by pleading ignorance. In normal contexts,
moral rules have a point because moral ignorance is so rarely an excuse.

Moral Ignorance in Abnormal Moral Contexts

Abnormal moral contexts arise at the frontiers of moral knowledge when
a subgroup of society (for instance, bioethicists or business ethicists)
make advances in moral knowledge faster than they can be disseminated
to and assimilated by the general public and subgroups at special moral
risk (e.g., physicians and corporate executives). As a result, the rightness
or wrongness of some courses of action (for instance, routine involuntary sterilization of the mentally retarded) is, for a time, transparent
only to the knowledge-acquiring subgroup and is opaque to outsiders.
Because moral knowledge is not shared, the presumption that all agents
are equally capable of self-legislation breaks down. In order to normalize the moral context, channels for moral communication may be institutionalized. Following the rise of medical ethics, for instance, hospitals
instituted ethics committees, healthcare schools added ethics coursework, and there were changes in laws governing healthcare practice.
Moral ignorance in abnormal moral contexts obviously differs from
that in normal contexts. It is, first, an ignorance of what only a limited
group of others know. Moral ignorance is the norm. Second, because it
is ignorance of advances in an existing base of moral knowledge, being
morally ignorant in abnormal contexts is perfectly compatible with
taking the moral point of view and being self-legislating in other spheres
of ones moral life. One need be neither morally defective (like Wolf s
embezzler) nor morally corrupt to be at risk of wrongdoing in abnormal
contexts.
Most feminist moral critique occurs in an abnormal moral context,
because those critiques have made only limited inroads into popular
consciousness. Public consensus on the wrongness of discriminatory
196

R ESPONSI BI LIT Y A N D R EPROACH

hiring, sexual harassment, and marital rape makes the moral context
in which these oppressive acts occur a normal one. But feminists also
critique a wide range of actions and practices that would not, in popular consciousness, be considered wrong (male bias in psychological and
other theories, the design of female fashions, the use of he neutrally,
heterosexual marriage, and so on). Here the context of these actions
shifts to an abnormal one.
Because of idiosyncrasies not shared by other new areas of applied
ethics, feminist criticism creates an abnormal moral context that is particularly resistant to normalization. Most obviously, feminists lack the
sort of institutionalized channels of communication between insiders
and outsiders that bioethicists and business ethicists have. Womens
studies programs and feminist publications and professional organizations institutionalize communication within the feminist community.
More significantly, feminist moral criticism diverges from both traditional moral language and styles of moral reasoning. Other areas of
applied ethics reshape moral language by straightforwardly playing off
existing moral languagecorporate responsibility, patient rights,
involuntary sterilization, and so forth. Feminists, of course, do this,
too. Such neologisms as sexual harassment and date rape facilitate
both our seeing moral issues where we had not previously and our drawing connections between these and already acknowledged moral issues
(e.g., between rape by strangers and date rape).15 But feminists also reshape moral language in less readily accessible ways: marginalize, the
Other, silencing, rapist society, marriage as prostitution. Understanding the meaning, extension, and legitimacy of this kind of moral
language requires a much deeper familiarity with feminist criticism
than the first sort of neologisms do. Thus the language of feminist moral
criticism may obstruct its dissemination and assimilation.
Feminists reshape moral reasoning in similarly radical ways. Other
areas of applied ethics extend existing consequentialist and rights-based
theories to new areas of moral interest. Although feminists do this
too, they also reconstruct moral reasoning by, for instance, stressing
15. Jean Grimshaw, Philosophy and Feminist Thinking (Minneapolis: University of Minnesota Press, 1986), 8789, makes and develops these points.

197

MOR A L A IMS

systematic harmthat is, by assessing harm contextualized in an interlocking system of harmful practices. Consequentialists may sometimes
assess how actions contribute to other desirable or undesirable practices, but they certainly have not started by assuming that individual actions cannot be morally assessed without first understanding the system
of practices of which individual actions are a part.16 In addition, through
the work of Carol Gilligan and others, some feminists have worked on
developing an alternative moral theory, the ethics of care, which is neither clearly consequentialist nor rights based, and which emphasizes the
moral importance of personal, noncontractual relationships, compassion and sympathy, sustaining connection, and highly contextualized
moral reasoning.17 When moral knowledge advances by overhauling,
not just extension, becoming morally knowledgeable requires moral reeducation and not just supplementary coursework. Thus there are especially strong reasons in this abnormal context for having diminished
expectations for the level of moral knowledge about oppression attainable by individuals outside the feminist community.

III. R ESPONSIBILIT Y
Let us return now to the question of moral responsibility, addressing
it with sensitivity to both the analogies and the disanalogies between
Wolf s embezzler, who suffers from moral ignorance in a normal moral
16. Much feminist moral criticism concerns systematic disrespect or oppression of women.
While individual action types may in themselves be harmful to women (for example,
discriminatory hiring and promotion policies, rape, sexual harassment), they are additionally harmful to the extent that they fit into a system of harmful actions. Indeed, the
actual harmfulness of some forms of behavior becomes visible only when contextualized in a system of offenses. Failure to imagine offenses in a systemic context results
in the perception of women as overreacting. It results in a failure to understand, for
instance, that the object of womens anger is not being called honey, but being called
honey as part of a system of disrespectful linguistic practices. Some neologisms, e.g.,
rapist society, refer to such systems of harms.
17. Carol Gilligan, In a Different Voice (Cambridge, MA: Harvard University Press, 1982);
Annette C. Baier, Hume, the Womens Moral Theorist and Virginia Held, Feminism
in Moral Theory, both in Women and Moral Theory, ed. Eva Feder Kittay and Diana T.
Meyers (Totowa, NJ: Rowman & Littlefield, 1987).

198

R ESPONSI BI LIT Y A N D R EPROACH

context, and the individuals in our earlier examples, who suffer moral
ignorance in an abnormal moral context.
We might tell an excusing story for the ordinary man that is analogous to the embezzlers excuse. Imagine, for example, a man who always
refers to women as girls or ladies. He, too, is uncoerced into doing
so and is in complete possession of normal adult reasoning faculties.
Yet it seems he ought not to be blamed for linguistically infantilizing or
patronizing women, for, from his point of view, one cannot reasonably
expect him to see anything wrong with his actions. We may suppose
that in his childhood, his father and mother referred to women as girls
or ladies. He may also have come to understand that the former is flattering because it suggests youth and the latter is simply polite. We may
suppose that the people to whom he was exposed when he was growing
up gave him examples only of this linguistic use and this understanding
of its significance. From his point of view, then, it is natural to conclude
that girl is flattering rather than infantilizing and that lady is polite
rather than patronizing.
What makes this excuse more problematic than the embezzlers?
First, the embezzler is not just ignorant of property rights. More fundamentally, he lacks the capacity to take the moral point of view, a capacity
that might enable him to rise above childhood conditioning and to reassess his fathers, mothers, and others actions. By contrast, the ordinary
man, Bleiers unreflective scientist, and Addelsons prestigious scientist
are capable of taking the moral point of view and reassessing the morality of what everyone else does. They thus lack the embezzlers strong
excuse for moral ignorance, because he ought to have known better
applies, however weakly, to them as it does not to the embezzler.
This disanalogy will not bear much weight. While the ordinary
man, the unreflective scientist, and the prestigious scientist are capable
of taking the moral point of view and are equipped with many of the
tools necessary for moral reasoning, there are limits to the powers of
moral self-critique. The husband who refuses to pay child support could
know his error even before this became a media issue. He knows about
promises, parental obligations, and fair play; and it takes no heroic effort
to apply these to the child support case. But feminist moral criticism
also constructs new moral categories, new modes of moral reasoning,
199

MOR A L A IMS

and new priorities among old principles. Whenever self-criticism depends on having acquired new tools for moral reasoning, it hardly
seems reasonable to blame those who have not acquired these things
for failing to be sufficiently reflective. (This was Addelsons point.) Here
one must resist the temptation to suppose that anyone who has not been
severely deprived, in the way Wolf s embezzler was, but who nevertheless acts wrongly, must suffer some culpable vice of intellect or character. This reassuring supposition (reassuring because it places moral
rectitude fully within the power of individual will) ignores the social dimension of moral knowledgespecifically the possibility in abnormal
moral contexts of sharp disparities in the social distribution of moral
knowledge.
One might still object that there may be less exonerating reasons
for these individuals failure to reassess the practices in which they participate. Self-interest can motivate the suppression of moral reflection.
Business executives, for example, may suppress moral reflection about
their own business practices because they tacitly recognize that ethics
and profit maximization rarely coincide. And people can certainly take
advantage of abnormal contexts, pretending or cultivating ignorance
when prudent to do so. One of the points stressed by feminists is that
mens benefiting from oppressive social practices provides them with a
motive for resisting critical reflection and for exercising self-deception
about their own motives and about the consequences for women of their
actions. This possibility of motivated ignorance makes excuses suspect.
How strong is this objection? Self-deception is a matter of being
motivated not to examine ones actions or reasoning too carefully lest
something unpleasant turn up. Suppose a man who uses he neutrally
says, when challenged, Whats wrong with that? After all, its proper
English. Is he deceiving himself that there is no moral issue here or that
its being proper English is a good reason? If he has had no exposure to
feminist moral criticism and hears almost everyone around him using
proper English, he will have no motive to examine his grammar or his
reasons for using he neutrally. But this is very different from being
motivated not to examine his grammar. Lacking a motive to be morally
reflective is not self-deception. And a motive to be morally reflective is
exactly what people will lack when moral ignorance is the norm.
200

R ESPONSI BI LIT Y A N D R EPROACH

While the embezzler and the ordinary man acquire their ignorance
in very different contexts, in neither case does the charge that he ought
to have known better stick. At most we have learned to be more cautious about excusing in abnormal contexts, asking such questions as,
Could he, with a reasonable amount of extra effort, have come to the
conclusion that his actions were wrong? Was his moral ignorance motivated and self-deceiving?

IV. THE SA NCTIONING FORCE OF EXCUSING

I have argued that the individuals in section Is examples are not morally
responsible for participating in accepted but unacceptable social practices, because their wrongdoing occurs in an abnormal moral context
where items of moral knowledge that are crucial for assessing those practices are not socially available to them, and hence their moral ignorance
is not culpable. Yet our interest in questions of moral responsibility is
more than an intellectual one, satisfiable by achieving judgments about
responsibility. We also take a practical interest in determining how we
ought to respond to wrongdoers. I said at the beginning of this chapter
that in abnormal moral contexts our entitlement to respond with moral
reproach is independent of the blameworthiness of individuals. I turn
now to that argument.
Imagine an analogue to Wolf s embezzler who embezzles in an abnormal moral context. He lives in a society where employees routinely
embezzle money whenever they think they can get away with it (an extreme version of employees habit, in our own society, of appropriating office supplies). His father, mother, and the people to whom he was
exposed had embezzled varying amounts, and employers had come to
expect this. Indeed, because embezzling is so commonplace, people find
it hard to view embezzlement as theft. From his point of view, it is natural for him to feel that embezzling is not wrong, or at worst, only a minor
infraction.
While we may believe that his ignorance is excusable, the consequences of our acting on that belief differ substantially from the
consequences of excusing Wolf s original embezzler. Because Wolf s
201

MOR A L A IMS

embezzler appeals to extremely unusual causes for his ignorance, we can

afford to take pity on him. In the second case, the embezzler appeals to
causal factors that also cause widespread ignorance of the wrongness of
embezzling. Here, acting on our belief in his excusability has the effect
of sanctioning embezzling and of committing ourselves to putting up
with routinely having our money embezzled. To see why this is so, we
need to look at what we do when we excuse individuals both in face-toface encounters with wrongdoers and in moral theorizing about excusable wrongdoing.
Suppose a student calls his female professor Mrs. _____, although he knows of her doctorate and uses Dr. to address his male
professors. If she excuses him for doing so because he could not have
known how insulting this is, showing that she excuses him means letting
the insult pass. Pointing out that using Mrs. rather than Dr. is insulting, even if done with a smile, would show that she blames rather than
excuses him. This is because, in normal contexts, which he will assume
this is, people usually are responsible for their deeds. Thus Using Mrs.,
not Dr., is insulting says something not only about the action but also
about the doers responsibility for it. It is partly because letting it pass
is what we do when we excuse that an excusing response in abnormal
moral contexts has a sanctioning force. Drawing on experimental evidence, Sabini and Silver argue that
in cases where an individual sees others doing what she would not
do but doesnt voice her objections, moral drift occurs. The failure
to establish publicly the wrongness of a particular action gives it
an implicit legitimacy; even those who would be disposed to find
it wrong have a difficulty sustaining that view when others, presumably as competent on moral matters as they, give evidence by
their actions of finding it acceptable.18

In normal moral contexts, excusing responses to morally controversial actions are ambiguous. Not pointing out wrongdoing may show
18 John Sabini and Maury Silver, Moralities of Everyday Life (Oxford: Oxford University
Press, 1982), 83.

202

R ESPONSI BI LIT Y A N D R EPROACH

either that we excuse or that we sanction. In abnormal contexts, where

wrong actions are socially accepted, not pointing out wrongdoing is not
even ambiguous; it will automatically be interpreted as sanctioning.
One might object that it is possible to make clear that one excuses
while simultaneously offering moral correction. Parents do this all the
time: I know you didnt know any better and Im not blaming you.
Just remember in the future that you shouldnt.... While this works
in parent-child interactions (or similar cases where the interactors acknowledge their unequal moral knowledge), it does not work in most
adult-adult interactions, particularly not in abnormal moral contexts.
First, in parent-child cases, the presumably more knowledgeable adult
is, for that reason, entitled to make corrections. In adult-adult cases, excusing a presumably knowledgeable adult for moral ignorance and providing moral correction are likely to be viewed by the recipient either as
insulting, because it impugns his status as a normal adult, or arrogant,
because it claims privileged moral authority.19 Feminists are, of course,
claiming privileged moral authority because of the abnormal moral context. But from nonfeminists point of view, the context is normal and
they need neither excuses nor remedial education.
Second, mentioning the reason for excusing wrongdoing in abnormal contextsnamely because everyone else also (mistakenly) believes
that the behavior is acceptablemay well backfire into sanctioning.
This is because when appeals to what everyone else does or thinks work
meaningfully as justifications, they cannot be used as excuses. Consider
the use of he neutrally. Grammar textbooks spell out rules of proper
English, including use of he neutrally. Teachers correct and/or punish
students for not following the rules. The normalcy and social acceptability of this usage provides both the causal explanation and the justifying
reason for participating in this linguistic practice.20 Social determinants
19 Sabini and Silver, Moral Reproach, in Moralities of Everyday Life, discuss the way that
not having standing inhibits individuals willingness to offer moral reproach and, by
implication, others willingness to accept their reproach.
20 Since this chapter was written, social conventions with respect to using he neutrally
have changed significantly. The reader may wish to reflect on more contemporary examples such as the replacement of human and animal with human and nonhuman
animal in work on the ethics of our treatment of animals.

203

MOR A L A IMS

cause people to use he neutrally by providing the reason for doing so,
namely its propriety. Social conditioning that instilled only an automatic reflex to use he neutrally, and not the belief that this is proper
English, would be failed social conditioning. When the propriety of a
practice is socially accepted, citing as an excusing condition the social
determinants that created a belief in the practices propriety thus has a
sanctioning rather than an excusing force. Imagine telling Bleiers scientist that we excuse him for being morally unreflective because everyone else in science writes in a morally unreflective way. What will his
response be? Thats not an excuse, thats my reason! In short, citing the
social determinants behind participation in a practice can have excusing force only after that practice has been delegitimated. A feminist can
excuse Bleiers scientist, citing social determinants, to another feminist.
She cannot so excuse Bleiers scientist to his face or to another scientist,
because the excuse will appear to justify the behavior.
Citing excusing conditions also sanctions by making the practice
appear unalterable. Recall my earlier qualms about excusing the misuse
of cognitive authority. Yerkess excuse was that he could not have known
that there was anything wrong with his use of cognitive authority, because the scientific communitys acceptance of this use impeded his
moral reflection. While true in part, this excuse misrepresents the relationship between social practice and individual action by implying a one-way causal determination: social practice shapes individual
thought and action. In fact, causal determination proceeds dialectically.
Social practices can be sustained only through the concerted thought
and action of individual practitioners. Thus an excusing response to
individuals who participate in harmful social practices sanctions those
social practices by obscuring the individuals role in sustaining and, potentially, disrupting them.
How ought people to think of their relation to social practices?
Would encouraging a sense of responsibility for ones participation in
social practices facilitate moral and social progress better than encouraging an awareness of ones excusability? Does moral theorizing construct the wrong sorts of self-images by focusing (as I have done) on
excusing conditions? To answer these questions, let us return briefly to
the society where embezzling is a social practice.
204

R ESPONSI BI LIT Y A N D R EPROACH

Suppose, as moral theorists, we wish to criticize this practice while

excusing individual embezzlers. How would we carry this off? We would
need to explain what makes embezzling wrong: it results in company
owners suffering undeservedly and it violates owners rights to property.
But in order to make clear that we do not hold embezzlers responsible,
we would have to frame this discussion carefully. We might explicitly
state that individuals are not blameworthy. Alternatively or in addition,
we might draw attention to the social forces producing embezzling, to
the social practice of embezzling, to the social roles of embezzling employee and passively suffering employer, and to the ideologies that
mistakenly legitimate embezzling. Showing that we excuse would preclude describing embezzlers in such morally reproachful language as
greedy and depraved and as rights violators.
Drawing attention to excusing conditions and refraining from
reproaching embezzlers will, I think, have the net effect of sanctioning the practice of embezzling even when the reasons for embezzlings
wrongness have been fully articulated. This is because such theorizing
constructs self-images that are antithetical both to conveying the obligatoriness of respecting property rights and to motivating right action.
To see this, keep in mind the natural history of moral discoveries. In
the beginning, when the moral context is abnormal, an understanding
of embezzlings wrongness and of how social forces produce embezzlers will be confined to those in the know: moral theorists and those
who read them. Later, the fact that there are moral doubts being raised
about embezzling will filter down into popular consciousness without
those doubts yet being absorbed into popular consciousness. People
may then know that embezzling is thought wrong by some, without
themselves feeling or thinking that embezzling is wrong. Finally, the
wrongness of embezzling may be absorbed into popular consciousness,
normalizing the moral context. When the moral context is abnormal,
how will the person who refrains from embezzling because he is in the
know, and the embezzler, who merely knows that some people think
it wrong, regard themselves? And how are we (in the know) likely to
think about them?
If theorizing about embezzling presents it as a social ill about which
there is widespread moral ignorance, the nonembezzler will likely see
205

MOR A L A IMS

himself as a man of refined moral sensibilities, a morally enlightened

man. Because he refrains from doing what everyone else does, he
may see himself as someone to be admired and as deserving of employers gratitude for respecting their rights. Because he is exceptionally
enlightened, we may concur. Here is the first danger: in drawing attention to widespread but excusable moral ignorance, we construct a
conceptual scenario in which the nonembezzler can see his not embezzling as heroic, supererogatory, and hence deserving gratitude. Simultaneously, this scenario impedes his seeing that not embezzling is
simply what he ought to doand he is neither heroic nor deserving of
gratitude.
These consequences are far from hypothetical. Sensitive to the social
determinants of oppression, women often feel grateful when husbands
volunteer to babysit or when administrators show some minimal support for their feminist research interests. The driving force behind such
misplaced feelings of gratitude is the logic of moral language. X is obligatory means, Unless there are exceptional excusing conditions, you
are blameworthy and reproachable for not doing X. And X is supererogatory means, You are not blameworthy and reproachable for failing to do X, and deserve special praise for doing X. Unfortunately, this
logic breaks down in abnormal contexts where individuals are routinely
rather than exceptionally exempted from blameworthiness, and hence
from reproach, for failing to do the obligatory. No wonder, then, that
women have trouble sustaining their sense of what is owed them and
find themselves feeling grateful when given their due. The logic of moral
language dooms any attempt to sustain or convey the obligatoriness of X
while simultaneously excusing most failures to do X. Thus, in abnormal
contexts, we face a choice: either we can convey the obligatoriness of X
via moral reproach; or we can excuse, by withholding reproach, those
who deserve to be excused. But we cant do both.
Now for the embezzler. How are we likely to see him and how will he
see himself? If we believe that social forces have conspired to make him
both an embezzler and dim to the moral wrongness of what he does, we
will likely see him as a product of his times or just an old-fashioned guy.
If we are company owners, we might say, I dont let his little pilferings
get to me. Hes such a nice old man. He just doesnt understand (just
206

R ESPONSI BI LIT Y A N D R EPROACH

as women say, I dont let his little sexist remarks get to me. Hes such a
nice old man). If the embezzler is the least bit enlightenedhe realizes that others condemn embezzling but cant himself see what the fuss
is all abouthe may present himself as a product of his times or as an
old-fashioned guy, saying, Im sorry if my embezzling bothers you, but
I was brought up to take as take can, or I guess Im just old-fashioned. I
just dont feel right not dipping into the till now and then. (Women are
all too familiar with this old-fashioned man.) Here is the second danger:
in drawing attention to the social determinants of moral ignorance, we
construct a conceptual scenario in which we see the embezzler and he
sees himself as incapable of self-legislation. We thus refrain from doing
the one thing that might awaken him from his deterministic slumbers:
reproach him. Once again, we seem to face a choice: either we can
convey individuals self-legislative capacity to rise above social conditioning by reproaching failures to do so, or we can excuse, by withholding reproach, those who deserve to be excused. But we cant do both.
The point here is that theorizing is not just descriptive; it is also
reality constructing. In the process of describing the social forces producing and excusing dimness to certain kinds of wrongdoing, we are
also making up persons: the enlightened man and the old-fashioned
man (someone who is dim to wrongdoing as a result of social determinants). 21 We are making particular identities publicly available
for self-conscious wearing and labeling. Some identities, though, are
best kept out of the common market. The enlightened and the oldfashioned man are two. Both images encourage wearers as well as
labelers to focus on the obstacles to self-legislation and the heroic
effort it would take to learn a different way of thinking and action.
What is the alternative? Recall the first disanalogy I mentioned between Wolf s and my embezzler. My embezzler and the old-fashioned
man, unlike Wolf s embezzler, are capable of rising above their social
conditioning. Moreover, even though this might take heroic effort,
21. Making up people is borrowed from Ian Hackings article of the same title in Reconstructing Individualism: Autonomy, Individuality, and the Self in Western Thought, ed.
Thomas C. Heller, Morton Sosna, and David E. Wellbery (Stanford, CA: Stanford University Press, 1986). The idea of making up people comes primarily from social constructionists such as Erving Goffman, Peter L. Berger, and Thomas Luckmann.

207

MOR A L A IMS

refraining from participating in oppressive practices is obligatory,

not an elective, supererogatory act. Given this, we would be better off
making up a different set of identitiesidentities that draw attention
both to self-legislative capacities and to the moral obligatoriness of
not participating in oppressive practices. The identities rights violator and oppressor do the former, while not deserving gratitude
does the latter.
To summarize the moral of these stories, social vulnerability to moral
reproach is necessary for (1) publicizing moral standards, (2) conveying
the obligatory force of moral commands, and (3) sustaining our sense of
ourselves as self-legislators. In abnormal moral contexts, excusing excusable ignorance by withholding moral reproach inhibits the publicizing
and adopting of new moral standards. Thus, in abnormal contexts, it may
be reasonable to reproach moral failings even when individuals are not
blameworthy.

V. JUSTIFICATION V ER SUS POINT

Feminist confusion about moral responsibility is not the confusion of
muddled thinking but, rather, grows out of very real conflicts, in abnormal contexts, between being justified in assigning responsibility or excusing and there being a point in doing so. Justifications appeal to things
like rationality, the ability to take the moral point of view, having moral
knowledge and moral reasoning skills, having free will (at least in the
sense of not being overwhelmed by causal determinants), and so on. But
after we justify assignments of responsibility or excusability, we can still
ask, Whats the point? Why do we find it worthwhile to worry about
moral responsibility? Assigning responsibility licenses reproachful or
approving responses: anger, admiration, chastisement, praise, seeking
out, and snubbing. Moral reproach reminds or perhaps teaches us what
actions are morally unacceptable. So the first point is educational. The
second point is motivational: moral reproach motivates us to change
the way we act. The third point is conceptual: reproachful labelsfor
example, oppressor, exploiter, sexistconfirm our identities as
moral agents. By contrast, an excusing response has a limited point at
208

R ESPONSI BI LIT Y A N D R EPROACH

best: to recognize those who cannot reasonably be further educated, or

motivated, or made more alive to their agency.22
I have argued that, in normal contexts, decisions about responsibility can be both justified and have a point. Where everyone knows what
treatments of women are wrong, most people will be responsible for
their mistreatments of women, and thus women will be licensed to reproach mistreatments in ways that promote conformity to correct moral
standards. I have also argued that, in an abnormal context, decisions
about responsibility cannot be both justified and have a point. Where
moral ignorance is the norm, an excusing response to moral ignorance
precludes the social growth of moral knowledge.
Should women never be tolerant? Should we call the ordinary man
an oppressor? Should we reproach Bleiers and Addelsons scientists? I
do not know. But a commitment to moral improvement seems to require
sometimes going for the point.

22. The distinction between justification and point is not new. Determinists and consequentialists have frequently observed that even if there is no justification for assigning moral responsibility, there may nevertheless be a point to doing so, or at least to
acting as if people were responsible. Blaming, praising, punishing, and rewarding cause
people to conform to moral and legal standards; and that is a worthwhile goal. My own
view differs substantially from this one, since my claimthat reproaching those who
are not blameworthy is warranted by the benefits of doing sois strictly limited to
abnormal moral contexts.

209

PA R T I V

T E L LI NG MOR A L STOR I E S
FOR OT H ER S
All moral action takes place within a particular social practice of morality
where people share moral understandings, rely on those understandings to
make themselves intelligible, come to have moral identities for others, succeed in communicating morally valuable attitudes like respect to others, and
sometimes contest those shared moral understandings and resist the social
practices built on them. All the preceding essays have focused in one way or
another on the significancesometimes for the better, sometimes for the
worseof shared moral understandings.
This last pair of essays takes up a different dimension of practicing morality with others. Both focus on the interpretive moral stories we tell. Chapter8 highlights the various kinds of interpretive stories that we tell to other
people with the aim of making a positive contribution to their agency and to
their response to others agency. We offer those who are perplexed about what
to do, or who seem poised on the verge of doing the wrong thing, an interpretation of their situation that we think will aid their moral decision making.
We also offer those who are perplexed about what to feel, or who seem to
have the wrong reactive attitudes in response to others actions, interpretations that we hope will enable them to feel differently. In this latter case, we
do what I call emotional work on other peoples emotional lives. One might
read four of the essays in this volume as philosophical exercises in emotional
work; chapters1, 2, 7, and 9 offer interpretive stories for those who are perplexed about their feelings of having failed morally, of shame in response to

MOR A L A IMS

misguided criticism, of gratitude for receiving what is owed, and of resentment and inability to forgive. Indeed, I wrote those essays because I needed
better interpretive stories. Finally, we offer interpretive stories to those who
are having such a hard time in their lives that they find it difficult to act. These
are stories that comfort grief, boost esteem, and remind people of the good in
themselves and in their lives. A central aim of chapter8 is to draw attention
to an important moral rolethat of the moral mediator.
Chapter9 takes up the interpretive stories that enable the reactive attitude of forgiveness. Some of the interpretive stories we tell support only
what I call minimalist forgiving. These are stories about how forgiveness
is deserved, either because the wrongdoing was excusable or because it was
justified, or about the more important things for the sake of which it would
be better to give up resentment. The truly difficult stories to tell are stories
that permit forgiveness of culpable, unrepentant wrongdoers. These are stories of aspirational forgiveness, the kind of forgiveness that is truly a moral
gift. I tell such a story for a fictional Auntie Muriel. I have positioned this
chapter last in part because it appears more distant philosophically from the
other essays, in part because it returns us to thinking about the difficulty of
understanding what people are up to when they engage in what appears to
be wrongdoing (a difficulty explored in chapter 1), and in part because it
reminds us that we sometimes have nonmoral aimsin this case, the aim
of making biographical sense of our livesthat can be in tension with our
moral aims.

212

C ha pt e r8

Emotional Work
In thinking about our own moral experience, whether we do this in philosophic theories or sermons, in solitary reflection or gossip, we stylize
our experience. Each of us stretches the moral experiences occurring
in our own lives on a common frame of concepts (agency, personal responsibility), images (self-mastery), and stock of examples (keeping
promises, being a good Samaritan). This common frame creates a homogeneity in the moral narratives that get told; what we say or think
about morality repeatedly invokes these stock concepts, images, and
examples. The patterns of moral thinking thus stylize moral experience by determining what we notice or overlook, remember or forget,
and take as important or trivial about our moral life so that moral
thinking only partly captures the completeness of experience.1 Stylizing moral experience is not in itself objectionable. It becomes so when
the terms of moral thinkingits images and conceptsjar with those
that frame other significant experiences. For women, the terms of an
ethics of justiceautonomy, mastering self-interested impulses, rights
of noninterferencemay grate against the terms that frame their lives
as women. Thus the different terms of an ethics of care may more comfortably and coherently style the same moral activities that were once
jarringly thought of in terms of justice. But moral thinking may not
only objectionably mis-style moral activities; it may also render invisible, unspeakable, or trivial routine moral activities that we sense (even
if we cannot say) are central to goodness. My own sense is that this is
what has happened to emotional work. Emotional work names something that feels familiar, that my moral aspirations call me to do and to
1. I am thinking here of Peter L. Bergers and Thomas Luckmanns remarks on the power of
language in The Social Construction of Reality (New York: Doubleday, 1966).

213

MOR A L A IMS

demand from others. Yet I find myself speechless, unable to think the
activity in its moral dimensions. What is emotional work? Why does it
elude moral thinking? How might we rethink morality?
Arlie Russell Hochschild coined the term emotional work. 2 For
her, it named our efforts to conform our emotions to what we believe we
ought to feel. The party goer, she says, summons up a gaiety owed to
the host, the mourner summons up a proper sadness for a funeral. Each
offers up feeling as a momentary contribution to the collective good.3
Her idea that emotions are prescribedwe ought to feel one way rather
than anothercomes out of a social constructionist view of emotion.
Emotions do not lie beyond the pale of social life (being brute, perhaps
biological givens). They are tied to a societys meaning and value system,
and they are subject to social and moral regulation. Feeling rules prescribe when, where, how much, how long, about what, and toward whom
different emotions should be felt. Such rules prescribe not only emotional expression but also subjective experience. Hence the need to do
emotional work on ourselves.
This understanding of emotional workgiving others their emotional duedoes not elude moral thinking. For Aristotle, the virtuous
person has his emotional house so well ordered that feeling the right
emotions in the right ways is second nature. The religious possibility of
sinning in ones heart or being pure of heart presupposes emotional obligations. Marcia Baron has argued that Kantian moral duties include
emotional duties, because merely going through the motions is often
less than what duty requires.4 And in daily life, we pass moral judgment
on heartlessness, selfish jealousy, sullenness, and self-pity. Moral thinking accustoms us to measuring moral track records not only by the yardstick of correct action but also by that of correct emotion.

2. Arlie Russell Hochschild, The Sociology of Feeling and Emotion: Selected Possibilities, in Another Voice: Feminist Perspectives on Social Life and Social Science, ed. Marcia
Millman and Rosabeth Moss Kanter (New York: Anchor, 1975); and The Managed Heart:
Commercialization of Human Feeling (Berkeley: University of California Press,1983).
3. Hochschild, The Managed Heart, 18.
4. Marcia Baron, The Alleged Moral Repugnance of Acting from Duty, Journal of Philosophy 81 (1984): 197220.

214

EMOTIONA L WOR K

But what attracted some feminists, including myself, to the term

emotional work was not this ordering of our emotional households.
Emotional work names something else, a work women do and are expected to do, especially in managing the domestic household. It names
a familiar moral activity that nevertheless escapes moral recollection
and reflection. Emotional work names the management of others
emotionssoothing tempers, boosting confidence, fueling pride, preventing frictions, and mending ego wounds. Taking care of others, creating domestic harmony, and caring about how others fare morally calls
for work on others emotions. This emotional work eludes moral thinking. It falls outside our paradigms for moral activity.
Why does emotional work done on ourselves show up in moral
thinking while emotional work done on others does not? Moral thinking tends to be dyadic and agent centered. We think in terms of I, the
agent, acting toward or on you, the patient. Given this emphasis on
agency, we think that there are two morally assessable roles open to us.
We can be (good or bad) moral agents, worrying about what we ought to
do or give to others. We can also be (good or bad) moral judges, witnessing an agent-patient dyad and critically assessing what the agent ought
to have done or given. In either case, we locate moral activity, responsibility, and praise- or blameworthiness exclusively in the agent. The patients are merely passive beneficiaries or victims of our agency, and the
judges merely spectators. The moral individualism of this style of thinking makes it natural for us, when we think about moral activity, to think
about our own moral activity. We understand our moral task to be putting our own moral households in order. We may turn a judicial eye on
others moral households, but governing their households is their affair,
not ours. The link forged between moral activity and self-management
is particularly strong in the more Kantian strain of our thinking. That
strain equates managing others moral activity with immoral disrespect
for others agency.
This picture of moral thinking explains the visibility of emotional
work done on the self and the invisibility of emotional work done on
others. In evoking morally prescribed emotions in ourselves, we worry
about managing the self, tidying our own moral households, looking
outward only to judge others emotional work. Emotional work on the
215

MOR A L A IMS

self is required lest others become victims of our mismanaged emotions.

This fits a dyadic, agent-centered picture of moral activity. By contrast,
managing others emotions smacks of disrespectful manipulation of
others agency. In taking on the burden of managing others emotions,
we step beyond the moral roles agent and judge. Thus emotional work on
others becomes invisible to moral thinking.
Dyadic, agent-centered moral thinking may adequately trace the
contours of some of our moral experience, but it lacks universal applicability. Emotional work on others, moral education, and moral counseling
elude dyadic, agent-centered moral thinking. Because moral education
and counseling often are emotional work, I want to start by exploring
this more familiar territory and its proper style of moral thinking.
Parents assume the task of morally educating their children. In
times of changing morals, people may also assume the task of educating fellow adults. Throughout our lives, we find ourselves called on to
provide moral counseling to those who ask, What should I do? Moral
education and counseling both involve more than teaching or applying
moral standards. They involve teaching others how to interpret moral
situations. A friend who asks What should I do? asks neither for a pronouncement nor a list of rules. She asks for help in putting her situation
into perspective, finding some interpretation of her circumstances that
will make one course of action clearly preferable.
Moral education and counseling elude dyadic agent-centered thinking because the moral educator or counselor is neither agent (her actions
are not the central issue) nor specular judge. Instead, she mediates between agent and patient, thus operating within an essentially triadic
and mediator-centered moral relation. When, for example, I counsel a
friend dissatisfied with her marriage and considering an affair, I mediate between her prospective agency and the prospective patients of her
action (her husband, child, potential lover, and also herself). It is not our
own moral household, but anothers that needs ordering and for whose
eventual order or disorder mediators assume partial responsibility. This
participation in anothers agency is partly what makes moral individualism strikingly out of place in mediator-centered moral thinking. Partly,
too, in recognizing how our own moral households were constructed
through moral education and are continuously refurbished through
216

EMOTIONA L WOR K

moral counseling, our moral debt to the mediating work of others becomes visible.
Placing moral mediation, not moral agency, at the center of moral
thinking shifts our understanding of morality away from individual task
and toward cooperative venture. This is not, however, the contractarians cooperative morality where each agrees to play by the rules and
properly manage the self. It is a deeply cooperative morality in which
agency itself is doubly open to management by others. First, moral reasoning preceding action paradigmatically occurs between people, not
in private deliberation. Even if we sometimes reason in private, we do so
only by interiorizing the dialogue through which we first learned morality and through which we update and correct our moral thinking. In
daily life, we rely on others not merely to manage their own agency but
also to help us manage ours by listening and dialogue. Listening invites
us to bring to mind and speech our interpretations of moral situations
and our judgments about what would be best to do. 5 Dialogue with
others shows us alternative interpretations and judgments.
That agency is open to management by others means, second, that
immoral action (the abuse of agency) is open to intervention by all those
positioned to mediate. In daily life we do not see ourselves as alone in a
moral wilderness. We rely on others to intervene and protect us when
we are about to be victimized; we rely on others to call offenders to task
and to help us recuperate from or to mitigate the harm done by others
misuse of their agency.
Emotional work is the work of a moral mediator. It is part and parcel
of moral counseling and educative moral dialogue. Because what we
feel is tied to how we interpret situations, helping others get the right
moral perspective cannot be detached from working to correct their
emotional attitudes. I think of times, for example, when after a class
on sexual harassment, a student has come to me with her story about
5. I have in mind Nell Mortons notion of hearing to speech in The Journey Is Home
(Boston: Beacon, 1985). The presupposition underlying hearing to speech is that
people often come to know their own minds first through dialogue. It is only the mistaken assumption that people are fully capable of knowing their own thoughts by themselves that leads to the view that helping people formulate their ideas is likely to be
manipulative.

217

MOR A L A IMS

a teacher or employer. Her emotional reaction to harassment concerns

her. Even though intellectually she condemns harassment, she doesnt
know how to feel its wrongness in her experience. She doesnt know
how to feel angry, or if she does, she doesnt know how to accept the
legitimacy of her anger. Self-doubts, self-criticisms, and respect for authority clutter her perceptions. So we tell a story about her experience
that enables and legitimates the anger. We talk about the unfairness
of what happened to her, about how it undermined her achievement,
about the contempt shown her as a woman, about her worthiness. This
is emotional workenabling someone to feel what, intellectually, she
knows to be true. This is also emotional work on a grand scale, the kind
that transforms emotional structure, making anger at harassment possible for the first time. This transformation of emotional structure goes
hand in hand with a transformation in the moral belief system that defines the proper objects of anger, guilt, contempt, and so on. Emotional
work is critical to successfully transforming a moral belief system, because the emotional enlivening of moral beliefs allows them to have
moral force.
Some emotional work more simply aims to correct inappropriate
emotions that are grounded in misinterpretations. In feeling emotions of
self-assessment (guilt, shame), other assessment (anger, contempt), and
wounding (feeling betrayed, let down, excluded, hurt), we sometimes
get it wrong, laying blame or fault or bad intentions where they dont
belong. Doing emotional work, then, is a matter of telling the story differently: She didnt mean to dismiss your point. She thinks very highly
of you. Remember how she usually appeals to you for confirmation. Or
You know your father wasnt really angry at you. He just wants to get his
project finished without interruption. Maybe hell do something with
you later. A lot, though not all, of this emotional work not only mediates by shaping the persons moral interpretation of others, but it also
mediates between persons by creating the understanding between them
that at least permits forgiveness, if not a retraction of blame. Telling stories about other people correctly, and thus doing emotional work well,
requires psychological acumen: the ability to hear between the lines, to
read body language, to interpret inflection and tone, to add up behaviors
into a psychological profile. It also draws on a general understanding of
218

EMOTIONA L WOR K

human psychology, as well as detailed understandings of particular persons and personality types.
Emotional work is also part and parcel of mediating intervention in
the immoral exercise of agency. That is, caring about how others fare
morally, and being willing to do emotional work, is not just a matter
of caring that others get the right moral interpretations and thus that
their emotional attitudes and moral actions are well informed. In caring
about how others fare morally, we also care about how they fare as patients of others moral agency. The moral wound inflicted by abused
agency goes beyond mere violation of rights or disrespect. It is also
an emotional wound: feeling hurt, resentful, humiliated, betrayed, let
down, disregarded, shamed, isolated, counted for less. Although helpless to prevent emotional wounding, we may still be called on to do
emotional work as a way of making the best of a morally bad situation.
I recall being a new member in a department and unwittingly violating
department policy for handling course overrides. When I was tonguelashed in from of my office mate, he intervened with humor, quipping,
Dont we wish there were hordes of students beating down the doors to
our classes! He turned what would have been, without moral support, a
wounding and alienating situation into one of mild affront. This, too, is
emotional worktaking the emotional sting out of abuse with humor,
commiseration, compensation, or psychological explanations that make
abuse forgivable.
But we are not always helplessly positioned where remedial emotional work is our only option. Emotions motivate action, and we may
sometimes be morally called on to reroute others actions by managing
their emotions. Women who live with abusive men learn well the strategies that deflect anger from its path toward themselves or their children.
They live walking on eggs, finely tuned to their mens emotions. But
daily life is also strewn with such eggs that require managing both individuals emotions and emotional atmospheres in ways that will bring
out the best in others and prevent the worst.
Both rerouting others actions by managing their emotions and
taking the motional sting out of abuse are morally risky forms of emotional work. In protecting people from moral abuse or from its emotional sting once it happens, we may simultaneously be protecting from
219

MOR A L A IMS

moral reproach and from being taken to moral task those who perhaps
most need being reproached and taken to task. Those doing virtuous
emotional work must tread the fine line between protecting the deserving from harm and refusing to protect the underserving from reproach.
I want to close with a note about benevolence. Fearing that managing others interior (their emotions, needs, patterns of interpretation) would undermine or intrude on autonomy, agent-centered moral
thought assumes that benevolence should provide instrumental goods.
Out of benevolence, I offer a hand, directions, food, housing. The good
Samaritan offers others the instrumental goods and services needed for
carrying out their preformed plans. But there is another kind of benevolence we may feel is important, one that moves into the interior and
mediates between a person and herself by managing her emotions. The
emotional blows we suffer in daily life may stop us in our tracks, or make
us hesitate, or set us off our path. It is not for lack of the right tools that
we fail to go forward, but for lack of desire. Grief, disappointment, lack
of confidence, lost pride, or failed trust undermine our agency as surely
as malnourishment or homelessness. What we may need most is a story,
one that reconciles us to the past and reopens the doors to the future.
The ear that listens to grief or the voice that reminds us of the good in us
allows that story to be told. It can be told poorly, dismissing grief, creating false confidence, or mending trusts that are better left broken. But
done well, emotional work is part of the virtue of benevolence.

220

C ha pt e r9

Changing Ones Heart

Why should I forgive him?A perfectly normal question, but what
kind of answer does one expect? Presumably, a reason good enough
morally to justify forgiving. That reason must also make forgiving psychologically possible by providing a description of the wrongdoer under
which he becomes an appropriate object for a changed heart. Being
shown that he deserves forgiveness would do this. So would realizing
that continued resentment is causing too much harm. The problem is
that forgiveness offered because it is deserved or to avoid the costs of
resentment is disappointing. Such forgiveness is not, perhaps, what one
aspired to get or to become able to give.
This is an essay about stories of forgiveness. The person who succeeds in forgiving us by appealing to good reasons typically tells a story
that distances our misdeed from the biography of the true self. In some
stories, we are excused or we repented. Sometimes the story forgetfully
omits the misdeed altogether, dwelling instead on our better side. But
aspiringly, we hope for a different story, one that shows some understanding of our whole self, complete with nasty, unrepentant, knowing
choices to hurt others. We dont want to be pared down to some pure,
good core.
What puzzles me about forgiveness is that, on the one hand, insisting that a person must deserve forgiveness or, at the very least, that there
is some justification for overlooking an injury seems cheap. It dodges
the hard task of forgiving while keeping the injurys inescapable connection to the agent in full view. On the other hand, forgiving unrepentant
people for inexcusable injuries seems repugnant, if not impossible.
In the first two sections of this chapter I try to explain the inadequacy and attraction of insisting that forgiveness be deserved or

221

MOR A L A IMS

consequentially justified. When looked at carefully, forgiving for these

reasons does not, really, look like forgiveness at all, because appealing
to good reasons has (and I will argue this later) the unwelcome consequence of making forgiveness nonelective. To some this will seem painfully obvious. What is curious is that when philosophers try to envision
some more robustly generous kind of forgiveness, the project almost
invariably fails. In section I, I focus on getting that failure out into the
open. In section II, I work through what I think is the source of the failure: we cannot make sense of the idea that persons could choose evil.
So we find ourselves driven to explain away or refuse to think about
those choices. In doing so, we may undergo a change of heart, but not
one that forgives the choice of evil itself. In the final section, I suggest a
way of telling stories that renders the choice of evil intelligible and permits a forgiving change of heart toward persons who make that choice.
Throughout, my concern is not just with locating the proper moral justification for changing ones heart but, more centrally, also with locating
the descriptions of wrongdoers under which genuine forgiveness could
be psychologically possible.

I. THE PROBLEM W ITH FORGI V ING

As part of emotion language, I forgive you works like I have stopped
being angry with you. It conveys information about what responses
you can expect from me. There will not, for example, be later recriminations, cold shoulders, and paybacks. I forgive you also tells you
that I have undergone a change of hearthave foresworn resentment,
anger, or other hard feelings. But, at least as used casually in everyday
life, I forgive you does not tell you why my original conviction that
you injured me no longer supports resentment. It does not tell you how
I manage to change my heart. The change may happen through a realization that resentment was a mistake. Really, no wrong was done.
Or the harm turns out to be excusable. Alternatively, the change may
happen because feeding resentment by dwelling on injuries, rather
than overlooking them, seems a bad idea. It is bad for the health or
is ruining a valued relationship. Finally, the change may result from
222

CH A NGING ONES HEA RT

discovering that the choice to harm, however culpable, is nevertheless

intelligible.
Among changes of heart achieved by various routes, some seem genuine cases of forgivenesswhat I call aspirational forgiveness. Here,
one manages to change ones heart while retaining a clear sense of the
others culpability and ones own entitlement to resentment. Other
changes of heart, though commonly called forgiveness, look on closer
inspection like something else: excusings, or overlookings, or givings
of what is due. Here one manages to change ones heart only by losing
clear sight of the others culpability and ones continued entitlement to
resentment. Because these latter are real changes of heart, and because
the linguistic use of forgiveness is not as finely tuned as philosophers
might like, I will call such changes of heart forgiveness, but of a minimalist variety. My central worry is that the decision that one ought to
forgive seems to yield only minimalist forgiving. Thus, the pursuit of a
moral justification for changing ones heart appears incompatible with
the attempt genuinely to forgive.

Minimalist Forgiving
Ought I to forgive? first comes up when we become discriminating
about forgiving. One way of asking this ought-question is, Is there
anything for the sake of which I ought to forgive even though I am clearly
entitled to my resentment? Sometimes the answer is yes. Satisfying the
resentful desire for retribution may not be worth risking the loss of an
otherwise good or needed relationship. In that case, we may try to circumvent resentment by replacing thoughts of injury with thoughts of
the persons better traits. Joanna North describes a man whose friend
has abandoned him for more amusing things. For the sake of friendships memory, he tries to forgive.
[H]e may refuse to dwell upon the thought of his friends rude
departure, and the unanswered letters. This is not to say he forgets them, but when the thought of them occurs he does not dwell
upon them. He will turn his mind to other things.... If the man
tries to think of his friends good points, of his amusing wit and
223

MOR A L A IMS

charm, of his courage, strength and adventurous spirit, he may

eventually manage to expel his resentment and anger.1

The friend gets a generous forgiveness, more than his due. But the injury
itself remains unaddressed. Because it is unaddressed, there remains a
real danger that efforts not to dwell on it will ultimately fail. Such forgiveness seems minimalist, partly because of its potentially temporary
nature, but it is more so because the injury is not so much forgiven as
ignored.2
Philosophers tend to focus on a second ought-question: Does she
deserve forgiveness? Concern with what is deserved is connected to
the thought that emotions ought to be warranted. Anger ought to be
well founded; if it is not, forgiveness is in order. Similarly, the absence
of anger should be well founded; otherwise, one should refuse to forgive. (Or possibly one should strive to become unforgiving. You should
resent that. Wheres your self-respect?)
Does she deserve forgiveness? asks: are there any facts about her
that make continued hard feelings unwarranted and inappropriate? In
particular, do any facts show that she should not (or should no longer)
be held to account (or as much to account) for wrongdoing? Such facts
typically show that the beliefs which grounded ones anger were false.
The wrongdoing was in fact excusable or justified, although one mistakenly thought it was not. Or some change occurring subsequent to
the wrongdoingfor example, repentance, restitution, or receipt of
punishmentmakes it unreasonable to continue holding the person to
account.
The distinction between deserved and undeserved forgiveness is
ultimately a distinction between changes of emotional attitude that
are warranted by their objects and ones that are not. A commitment to
having only warranted changes of heart is, obviously, a commitment to
a minimalist kind of forgiveness. To give up my hard response upon discovering your action was excusable or justified is to grant you no more
1. Joanna North, Wrongdoing and Forgiveness, Philosophy 62 (1987): 499508, 506.
2. Peter French pointed this out to me. Other writers on forgiveness have made a similar
point.

224

CH A NGING ONES HEA RT

than your due. It is to admit that there was nothing to be angry about,
nothing to forgive. Similarly, to stop my hard response after you have repented, or made up for things, or suffered enough is to do nothing more
than I ought. Once the desert question is raised, there will be no choice
to forgive, no generous granting of forgiveness. Reason requires changing ones heart, and forgiveness thus ceases to be elective.
One might, of course, try to reintroduce some element of electivity
by pointing out that, in a wide range of cases, the desert question simply
cannot be decisively answered, and the choice to forgive or not to forgive will have to be made. Desert questions may lack decisive answers,
for at least two reasons. First, the forgiver may find it difficult to know
whether the wrong is serious enough to merit resentment, whether an
excuse or justification is acceptable, whether the repentance is genuine,
or whether one has made things up or suffered enough. Under uncertainty, forgiveness cannot be required, and one remains free to elect to
grant the benefit of the doubt. Ultimately, however, such appeals to uncertainty fail to make forgiveness genuinely elective, because they presuppose that if we had sufficient knowledge, forgiving (or the refusal to
forgive) would be rationally required.
Second, desert questions may sometimes lack a decisive answer because of vagueness in the standards for what counts as a passable excuse,
an adequate justification, sufficient repentance, or full restitution for
harming. Often we must exercise our own discretionary judgment in deciding, for example, which excuses are passable and thus whether continued resentment or forgiveness is warranted. The need on occasion to
exercise discretionary judgment in deciding who deserves forgiveness,
however, makes forgiveness elective in only a weak sense. Once the decision in favor of desert has been made, forgiveness becomes required.
Thus, although people may exercise their discretionary judgment differently, with some adopting lenient and others strict standards of desert,
the more forgiving sorts of persons on this account cannot credit themselves with greater generosity. They simply have lower standardsfor
example, lower standards for what counts as a passable excuse.
The tension between wanting to forgive only the deserving and at
the same time recognizing the minimalism of restricting forgiveness
this way yields a variant of what Aurel Kolnai calls the paradox of
225

MOR A L A IMS

forgiveness: if only the deserving ought to be forgiven, then forgiveness is either unjustified [in the case of the undeserving] or pointless [in
the case of the deserving, because there is nothing to forgive].3

Aspirational Forgiveness
The minimalist notion of forgiveness seems unsatisfying as an account
of full-blooded forgiveness. First, it places the object of forgiveness
under the wrong sort of description. In forgiving for the sake of... , one
refuses to think about the wrongdoer as a wrongdoer. In forgiving the
deserving, one places the person under a description that implies there
is nothing to forgive. In neither case does forgiveness for a wrong take
place. Second, forgiveness becomes morally required rather than elective. It would be irrational not to forgive when forgiveness would preserve something more important than retributive satisfaction. It would
be irrational not to forgive those who deserve it.
Unlike minimalist forgiving, the forgiveness we aspire to get (and
give) is forgiveness for culpable, unrepentant, unpunished, and unrestituted wrongdoing, whose existence is not dismissed by refusing to think
about it. Or, more weakly stated, we want forgiveness for the culpability
that remains after all excuses, justifications, restitutions, and repentant
reforms have been made and accepteda culpability that warrants our
continuing to be resented. When I ask aspiringly for forgiveness, I ask
you to forgive me for something that renders me undeserving and entitles you to hard feelings toward me.
Some caution is needed, however, in understanding what forgiving
me for something means. It does not simply mean that I want you to recognize the wrongness of the act but not resent me for it. The latter can be
done by simply excusing me. In excusing me, you continue to regard the
act as wrong. The excuse simply enables you to separate your assessment
of the act from your assessment of me. As a result, I, as the object of a possible emotional attitude, do not fall under the description accountable
agent of wrongdoing. Hence I do not become an appropriate object of
3. Aurel Kolnai, Forgiveness, Proceedings of the Aristotelian Society 74 (19731974):
91106, 99.

226

CH A NGING ONES HEA RT

resentment. Of course, not all excuses completely exonerate; typically,

they only mitigate. My point here is that insofar as ones change of heart
is grounded in features of a person that make her non- or less culpable,
forgiveness is being offered for the absence of full culpability and not for
the culpability itself. Thus forgiveness based on accepting excuses represents only a readjustment of ones emotional attitude away from levels
of resentment that were in fact unwarranted. Although you may succeed
in changing your heart toward me by divorcing me from my act in this
or other waysfor example, by minimizing or overlooking the harm,
or by seeing my act as committed by a former, different selfyou will
fail to forgive me in the way I aspirationally want. What distinguishes a
minimalist forgiving of me from aspirational forgiving is the story you
tell about me. Minimalist forgiving tells a story that places me under
some description that makes me an inappropriate object of resentment
(or of as much resentment). In aspirational forgiveness, I, as the object of
forgiveness, remain under an intentional description of culpability that
warrants continued, unreduced resentment.
In addition to having as its object a person under a damning description, the element of moral choice is critical to this aspirational notion
of forgiveness, and it is choice of a specific kind. Unlike minimalist forgiving, where because of insufficient data or vague standards we must
choose whom to count as deserving of forgiveness, the moral choice
opened to us by aspirational forgiveness is one of how to respond to the
decidedly undeserving. Because forgiveness is an elective response to
culpable wrongdoing, it is conceptually connected with supererogatory
acts of generosity and charity. It is something we ask or hope, rather than
demand, for ourselves and that we grant, rather than owe, to others.
Forgiveness is a gift, not the paying of a debt or the remission of a debt
whose collection would prove too costly.

Repentance and Double Vision

Although forgiving by forgetting or by finding excuses or justifications for harm plainly falls short of aspirational forgiveness, many have
thought that forgiving the repentant is aspirational. The repentant
person finds her own actions neither justified nor excusable. She admits
227

MOR A L A IMS

our entitlement to resent her, but she hopes we will choose to base our
feelings on who she now promises to be for the future, rather than who
she was in the past. To forgive her appears to involve both seeing the
culpability of her wrong and electing to give up resentment. It appears
aspirational.
I think it is not. Those who find repentance important do so because
sincere repentance makes forgiveness both risk free and rational. In
undergoing a repentant change of heart, the wrongdoer makes herself
someone who will not injure us this way again. In breaking the connection between her wrongdoing and her true self, the reformed person
ceases to be an appropriate object of resentment. Only by refusing to
accept either the sincerity or the sufficiency of repentance can resentment retain a legitimate foothold. The point is this: to the extent that
repentance is allowed to count in favor of forgiving, so to that extent the
wrongdoer ceases to be viewed under a damning description and forgiving ceases to be elective.
Oddly, literature on forgiving the repentant fails to confront the
minimalism of forgiving in exchange for repentance. Instead, one finds
a kind of double vision. Forgiving the repentant is owed, rationally required, minimalalso elective, generous, aspirational. Such double
vision seduces because it satisfies the desire for justification and freedom from the charge of being too forgiving. It also satisfies the aspirational desire to exceed the bounds of justification in matters of the heart.
The failure to get beyond minimalist forgiving is evident in discussions by Richards, Kolnai, and Murphy of forgiving the repentant.
Norvin Richards claims that repentance does not mandate forgiveness.4 Like heroism, forgiving is admirable to do but not wrong to
omit.5 But his explanation of the electivity of forgiveness suggests the
contrary. Refusal to forgive the repentant is acceptable, on his view, if
one is unconvinced that the wrongdoer has in fact reformed.6 But this

4. Norvin Richards, Forgiveness, Ethics 99 (1988): 7797, 87.

5. Ibid., 80.
6. After giving reasons for forgiving the repentant, Richards says, On the other hand, neither would there be anything amiss in the hard feelings continuing.... For one thing, it
might not be so clear that the flaw has been repaired (ibid., 88).

228

CH A NGING ONES HEA RT

only shows that claims to repentant reform do not mandate forgiveness. By implication, true reform would mandate forgiveness. Richards
also argues that, sometimes, forgiving the repentant is psychologically
impossible. Then refusal to forgive is permissible.7 But this only shows
the excusability of failing to forgive those one is rationally required to
forgive.
Aurel Kolnai exercises the same double vision. He describes forgiving the repentant as an exquisite act of charity or benevolence.8 Yet he
also explicitly states that, because the repentant appear to deserve forgiveness, forgiving them is duty-like and a quasi-obligation. 9 How
can something that I am quasi, albeit not strictly, obligated to render be
an exquisite act of charity? Surely a quasi-obligation to forgive the repentant would make forgiveness at best only a minor act of charity, not
an aspirational achievement.
Although I think that Jeffrie Murphy is also guilty of double vision,
his explanation of how repentance works as a reason to forgive is tremendously helpful. It brings out the similarity between accepting repentance and excusing wrongdoing. As a result, it becomes clear why
forgiving because of repentance can only be minimalist forgiving. (This,
of course, was not Murphys intention.)
On Murphys view, the only acceptable grounds for forgiveness are
ones that enable us to draw a distinction between the immoral act and
the immoral agent; for then we can follow Saint Augustines counsel and
hate the sin but not the sinner.10 Most notably, proof of good intentions, of repentance, and of former good character enable this divorce.11
7. As a reason for finding continued resentment toward the repentant permissible, he
says, ... it might be that the harm was too serious for the change to relieve ones other
associations between this person (that face, those hands, that smirk) and what she did
to you, despite those features having not even partly caused the harm (ibid.).
8. Kolnai, Forgiveness, 104.
9. Ibid., 101, 105.
10. Jeffrie G. Murphy, Forgiveness and Resentment, in Forgiveness and Mercy, ed. Jeffrie
G. Murphy and Jean Hampton (Cambridge: Cambridge University Press, 1988), 24.
11. Murphy speaks of old times sake rather than former good character. But since he says
that old times enable forgiveness for what you once were similar to the way repentance
enables forgiveness for what you now are (ibid., 29), I assume the old times must include the agents good character.

229

MOR A L A IMS

But the divorce of act from agent is exactly what all excuses do. To
excuse, Murphy explains, is to say this: What was done was morally
wrong; but, because of certain factors about the agent (e.g., insanity),
it would be unfair to hold the wrongdoer responsible or blame him
for the wrong action.12 Although good intentions, repentance, and
former good character do not completely excuse the way insanity does,
all three work as partial excuses. They work as reasons for thinking that
it would be unfair to hold the wrongdoer fully responsible and blameworthy. Thus, as Murphy himself says, resentment would be inappropriate and irrational.13 All excuses work by driving a larger or smaller
wedge between act and agent. Thus in insisting that forgiveness be
given only when act can be separated from agent, Murphy in essence
is requiring that forgiveness be given only for excused wrongdoing.
Whatever elective appearance forgiveness may have reduces to the fact
that, in most cases, we will have to deliberate about the repentances
authenticity (Is he genuinely repentant?) and sufficiency (Is repentance enough?).

The Problem of Forgiving without Condoning

What motivates the quest to find individuals deserving of forgiveness? What underlies reluctance to go beyond minimalist forgiving to
a clear-eyed forgiveness of the unexcused and unrepentant? The fear of
condoning wrongdoing is at least partly responsible. But the danger of
condoning is overrated. What I will begin to argue here and continue
in section II is that getting beyond minimalist forgiving is hard, but not
because one cannot morally justify doing so. Rather, the stories that
permit us to give up resentment all seem to be ones of desert.
The concern that one might, by forgiving, condone wrongdoing
arises because moving from resentment to some more positive emotion
is not simply a matter of changing how one feels about wrongdoers. No
12. Ibid., 20.
13. Ibid., 29. That I repent is one kind of excuse and not something in a category of its own
is more obvious if one keeps in mind that to repent is to admit having made a mistake.
It was a mistake is clearly an excuse. J. L. Austins A Pleas for Excuses, in Philosophical Papers (Oxford: Oxford University Press, 1979), is useful to keep in mind.

230

CH A NGING ONES HEA RT

emotion is simply a feeling; emotions include thoughts and behavior in

such a way that emotion-inappropriate thoughts and actions provide evidence against ones emotion claim.14 A forgiving change of heart thus
commits us to changing how we think about and treat the wrongdoer.
In particular, forgiveness appears to commit us to some more positive
moral assessment of the wrongdoer and to actions consonant with that
assessment. This seems to rule out retribution, moral reproach, nonreconciliation, a demand for restitution, and in short, any act of holding
the wrongdoer to account. This, of course, is not a problem if the person
ought not to be held to account (or held to account any longer)if, that
is, she deserves forgiveness because the injury was excused, justified, or
repented.
Culpable wrongdoers are the problem. For them, not responding
resentfully may seem to send a condoning message: the act was not
wrong, or wrongdoing will not be penalized, or I am not the others
moral equal, so that in this case the act does not count as moral mistreatment.15 In sending any one of these messages, one fails to prevent
repetitions and may well encourage them. As a result of doing nothing
to improve the wrongdoers character, both oneself and others remain at
risk of future mistreatment. Worse yet, the condoning messages might
have a wider audience. All who witness the lack of resentment may take
it as a green light for misbehavior. In any event, fear of condonation is
here largely a fear of undermining the social practice of morality by not
conveying and enforcing the rules and by not improving improvable
characters.
14. This point is so commonplace in theory of emotion that it is hard to know whom to cite.
Particularly good (and very similar) arguments for the connection between emotion,
thought, and action are given in Frithjof Bergmann, A Monologue on the Emotions,
in Understanding Human Emotions, ed. Fred D. Miller Jr. and Thomas W. Attig (Bowling Green, OH: Bowling Green State University, 1979); and Errol Bedford, Emotions, in What Is an Emotion? ed. Cheshire Calhoun and Robert C. Solomon (Oxford:
Oxford University Press, 1984).
15. Paul Lauritzen, Forgiveness: Moral Prerogative or Religious Duty? Journal of Religious Ethics 15 (1987): 141154, stresses the consequences of not penalizing (the
wrongdoer may be emboldened by the mildness of the penalty [145]); and Jean Hampton and Jeffrie G. Murphy, Forgiveness and Mercy (Cambridge: Cambridge University
Press, 1988), stress failure to assert ones own moral value.

231

MOR A L A IMS

Some have tried to argue that forgiving is compatible with a stern

or retributive response. Thus one can forgive the undeserving and still
convey the correct moral messages.16 Some harsh treatment, Murphy
observes, would, of course, be incompatible with forgivenessnamely
harsh treatment the very point of which would be to show you how
much I hate and resent you. But when the harsh treatment is based on
other factors (e.g., concern with legal justice), forgiveness need pose no
obstacle to such treatment.17 Depersonalizing the retributive demand
(It is not I who require, but the law or justice) makes forgiving compatible with retribution. But where forgiveness matters mostin intimate
relations where we are vulnerable to betrayalssuch depersonalizing
strategies are unlikely. I, not justice, require retribution for betraying
my love, trust, or friendship. Here forgiving requires resisting retributive responses and risking condonation.
But the belief that, without retribution, forgiving the undeserving
always risks condonation rests on two assumptions: (1) that in every unrepentant case, failure to protest sends a condoning message; and (2)
that sending this message will always have some significant and morally
objectionable consequence. Kolnai appears to accept the idea that unrepentant wrongdoers will always interpret lack of protest as condonation
and will consequently always repeat their offense. He remarks that if the
offense is still subsisting, then by forgiving you accept it and thus confirm it [for, presumably, the wrongdoer] and make it worse.18
As empirical statements about what our actions mean to others,
and about the consequences of those meanings, both are false. In social
contexts where everyone knows what the wrong acts are and who
the moral equals are, and where wrongdoing frequently though not
invariably meets with protests or penalties, the average person would
not likely interpret failure to protest as condonation. If I catch a neighborhood adolescent bashing my mailbox and forgive him, he might
think me exceptionally nice or wimpish. He surely would not infer that
16. Murphy and Hampton, Forgiveness and Mercy; Kathleen Dean Moore, Pardons: Justice,
Mercy and the Public Interest (Oxford: Oxford University Press, 1989); R. S. Downie,
Forgiveness, Philosophical Quarterly 15 (1965): 128134.
17. Murphy, Forgiveness and Mercy, 22 (emphasis mine).
18. Kolnai, Forgiveness, 98 (emphasis mine).

232

CH A NGING ONES HEA RT

bashing mailboxes is morally permissible, or will not be penalized, or

that I am not his moral equal.19 Nor do I necessarily worsen his behavior, because by forgiving him he may come to see that he is harming real
people, people he might like. As a point about human moral psychology,
the idea that resentment, protest, and punishment best effect moral improvement is surely misguided. The last thing some people need is yet
more resentment and punishment. 20 Furthermore, as a point about ordinarily decent but flawed persons moral psychology, the idea that moral
improvement requires others tutelage or punitive sticks is equally misguided. To be a moral agent is to be capable of self-correction. In short,
because forgiving the undeserving is not synonymous with condoning,
there can be no blanket moral objection to aspirational forgiveness.
There is, however, a different, more serious, concern about condonation. It is this: when the audience of condonation is oneself, not
others, condoning is not just a matter of what one does in response to
wrongdoing, and hence of what messages one sends. It is also a way of
proceeding to think about being injured. One refuses to look at this fact
straight on. Instead, one begins making up or tracking down possible
excuses, possible justifications. Or one might just pretend to oneself that
it never happened. In forgiving the undeserving by minimizing, rationalizing, or ignoring injuries, one effects a change of heart by telling a
story that unrealistically portrays the wrongdoer as more deserving of
benevolent attitudes than in fact he is. In doing so, not only does one
end up offering only minimalist forgiveness (a forgiveness mistakenly
taken to be owed), but one also risks becoming blind to ones own moral
entitlements.
Is it possible to forgive the undeserving without condoning wrongdoing for oneself (as it does seem possible to forgive them without
necessarily condoning wrongdoing for others)? Is it possible to follow
Norths injunction that If we are to forgive, our resentment is to be
overcome not by denying ourselves the right to that resentment, but
19. For an extended discussion of the contexts in which letting wrongdoing pass does
amount to condonation, see chapter7 this volume.
20. Jean Hampton, Forgiveness Resentment and Hatred, in Forgiveness and Mercy, 87, is
one who points out that forgiveness may effect moral improvement.

233

MOR A L A IMS

by endeavouring to view the wrongdoer with compassion, benevolence

and love while recognizing that he has wilfully abandoned his right to
them.21 In order to forgive the undeserving without condoning their
wrongdoing for ourselves, we would have to be capable of aspirational
forgiveness. That is, we would have to be capable of telling a story about
wrongdoers that simultaneously portrays them as undeserving and unlocks the doors to a change of heart. The problem is that there appears to
be no aspirational story, only stories of desert.

II. TELLING STOR IES

An aspirational story would begin by connecting misdeeds to the
agents true self. The protagonists of aspirational stories are thus always
unclean, unworthy. The object of aspirational stories is to get us to
accept this fact and go on without resentment. By contrast, stories of
desert separate misdeeds from the agents true self. His true self appears
worthy in spite of the injury he inflicted. We can go on without resentment because he is worth it.
I have said that aspirational stories appear impossible ones to tell
successfully. P. F. Strawsons discussion of resentment in Freedom and
Resentment suggests why. 22 On Strawsons view, to regard someone
as a person is to regard her as someone with whom one might conduct
an interpersonal relationship. Although we might relate to individuals
in other ways, we could not relate to them interpersonally if it did not
matter to us whether the actions of other peopleand particularly of
some other peoplereflect attitudes towards us of goodwill, affection,
or esteem on the one hand or contempt, indifference, or malevolence on
the other.23 Thus life with other persons is characterized by an expectation and a demand for the manifestation of a certain degree of goodwill
or regard on the part of other human beings towards ourselves; or at
21. North, Wrongdoing and Forgiveness, 502.
22. P. F. Strawson, Freedom and Resentment, in Freedom and Resentment and Other
Essays (London: Methuen, 1974).
23. Ibid., 5.

234

CH A NGING ONES HEA RT

least... the absence of the manifestation of active ill will or indifferent

disregard.24
Resentment is one possible reactive attitude toward our expectations being disappointed; hurt and moral indignation are other possibilities. But even if others may disappoint us by harming, resentment
nevertheless presupposes that they have the capacity to bear us goodwill
and to care about their moral mistreatment of us. Should we come to
believe that our expectations are misplaced because we are dealing with
individuals who are incapable of caring about our well-being, we will simultaneously become invulnerable to resentment and cease seeing ourselves as dealing with individuals who are persons.
The forgiving abandonment of resentment is also a reactive attitude.
It is something offered to disappointing persons. Thus any story enabling
us to forgivingly overcome resentment must portray the wrongdoer as
the sort of individual who continues to be an appropriate object of reactive attitudes because she is capable of caring about our well-being.
Strawson tells two sorts of stories. The first sort employs such expressions as He didnt mean to, He hadnt realized, He didnt know;
and also all those which might give occasion for the use of the phrase
He couldnt help it, when this is supported by such phrases as He was
pushed, He had to do it, It was the only way, They left him no alternative, etc.25 What Strawson finds significant about these stories of mistake, physical compulsion, and tragically forced choice is that none of
them invites us to suspend towards the agent, either at the time of his
action or in general, our ordinary reactive attitudes. 26 Instead, we understand how things go wrong and situations are complicated, and as
a result, how agents of goodwill can, without wishing to, end up disappointing us. 27
The second sort of story is a variation of the first. It employs such
expressions as He wasnt himself, He has been under very great
strain recently, He was acting under post-hypnotic suggestion.28 The
24. Ibid., 24.
25. Ibid., 7.
26. Ibid.
27. Ibid., 8.
28. Ibid.

235

MOR A L A IMS

difference is that here we are invited to suspend our ordinary reactive attitudes so long as the factors impairing the capacity for goodwill persist.
But because such stories also invite us to identify the real self with the
self he would be if... , and that self is one of whom we can expect goodwill, the agent remains a potential terminus of interpersonal relations
and reactive attitudes. Stories of repentance belong in this category,
because the repentant person claims now to be the person she would
have been if.... Again, we understand how things go wrong and situations are complicated in such a way that agents of goodwill may not be
themselves.
There are other obvious stories that enable the overcoming of resentment while preserving the personhood of agents and the appropriateness of reactive attitudes. One is a story about how things did
not go wrongthe injury was, after all, justified. Another is a forgetful story. But all the stories are recognizably variations on a single
theme: by divorcing a wrong act or vicious character trait from the
agents true self, they make the true self deserving of forgiveness. The
agent continues to be the sort of individual who bears us goodwill
(or would do so if . . . ) even though her action or character trait is
disappointing.
Such stories also sustain our faith in others capacity for goodwill
and in the possibility of interpersonal relationships with them. They
enable us to make sense of how persons could do harmthat is, how
someone with the capacity to care about our well-being could nevertheless do harm.
When explanatory stories distancing wrongdoing from the agents
true self fail, we must find some other way of making sense of our being
injured. Why does Strawson omit an aspirational story that would do
this job? Why not tell a story about the deliberate, unrepentant, unexcused, and unjustified choice to injure that at the same time depicts the
agent as a candidate for interpersonal relations and reactive attitudes
such as forgiveness? The point of such a story would again be to make
sense of how persons could injure us and to do so in a way that makes
forgiveness possible. The reason Strawson omits an aspirational story is
simply this: whatever shows that the agent really meant to do harm, and
thus cannot be divorced from his act, also tends to show that he is not an
236

CH A NGING ONES HEA RT

appropriate object of reactive attitudes, that he is not a person. As Daniel

Dennett puts it so clearly,
our assumption that an entity is a person is shaken precisely in
those cases where it matters: when wrong has been done and the
question of responsibility arises. For in these cases the grounds
for saying that the person is culpable (the evidence that he did
wrong, was aware he was doing wrong, and did wrong of his own
free will) are in themselves grounds for doubting that it is a person
we are dealing with at all. 29

In order to make sense of the knowing, unjustified choice to harm,

we find ourselves driven to tell stories denying that the agent is fully a
person. She is in some ways more like a bee or a volcanosomeone of
whom one should not have expected goodwill in the first place. Because
this sort of story eliminates some individuals from the field of candidates for interpersonal relations, it is a story told only with reluctance
and only when other sense-making strategies have failed.
Strawson describes one variant of this latter story: the attribution of psychological defect. Some individuals mean to harm because
their ways of thinking and feeling have gone terribly wrong or are
dramatically undeveloped. As a result, they have ended up with a bizarre worldview, or an inability to experience reactive attitudes such
as remorse, or without conscious purposes for acting. Only some deterministic description of the psychological machinery at work in such
individuals can make any sense at all of how they could think, feel, and
desire what they do. Here, we concentrate on understanding how he
works, with a view to determining our policy accordingly.30 Even so,
such individuals remain partly opaque. They are like beetles or kudzu.
We understand the machinery of behavior (e.g., reflex response, tropism, psychopathology) but are unable to enter sympathetically into
a life like that.
29. Daniel Dennett, Conditions of Personhood, in Identities of Persons, ed. Amelie O.
Rorty (Berkeley: University of California Press, 1976).
30. Strawson, Freedom and Resentment, 12.

237

MOR A L A IMS

Strawson calls the attitude adopted toward such individuals the

objective attitude. To adopt the objective attitude to another human
being is to see him, perhaps, as an object of social policy; as a subject
for what, in a wide range of sense, might be called treatment; as something certainly to be taken account, perhaps precautionary account,
of; to be managed or handled or cured or trained; perhaps simply to be
avoided.31 What the objective attitude rules out, however, is the possibility of engaging in interpersonal relationships. Not only is resentment
toward these individuals misplaced but so is forgiveness.
A second way, not mentioned by Strawson, of making sense of intentional harm by denying full personhood is to attribute malignancy
or moral indifference. Unlike the previous sense-making story, here we
imagine that malignant and indifferent individuals are still constitutionally able to feel goodwill, to care that others resent or hate them, and
to listen to moral suasion. Thus, we continue demanding goodwill from
them and resenting their failure to display it. But we also imagine that
they have made some fundamental and fundamentally perverse choice
not to care or listen. They are permanently disappointing individuals.
Only by attributing some fundamental choice of evil or indifference can
we make any sense at all of how they could do, desire, and feel what they
do. Even so, they remain partly opaque. We cannot enter sympathetically into the life of the frozen hearted. 32
In Life before Man, Margaret Atwood illustrates the struggle to
find some way of making sense of harm-causing actions. Elizabeths
Auntie Muriel is a dreadful individual. She in essence buys Elizabeth
and her sister from their alcoholic mother. Out of respect for what she
calls decency, Auntie Muriel is willing to pay Elizabeths husband to
stay in their failed marriage. She is unbearably self-righteous and stingy
beyond belief. After the death of Elizabeths mother, Auntie Muriel
draws up a list of every item shed ever lent, given or donated. One
light bulb, 60 watt over the sink. One blue plastic shower curtain. One
31. Ibid., 9.
32. Frozen hearted is Gary Watsons term. His discussion of the problem of making sense
of choices of evil has shaped my own. Gary Watson, Responsibility and the Limits of
Evil: Variations on a Strawsonian Theme, in Responsibility, Character, and the Emotions, ed. Ferdinand Schoeman (Cambridge: Cambridge University Press, 1987).

238

CH A NGING ONES HEA RT

paisley Viyella housecoat. One Wedgewood sugar bowl. Miserly gifts

and chipped discards. Auntie Muriel wanted them back.33 The magnitude of Auntie Muriels unrepented faults rules out the possibility of
excuse. In order to explain how anyone could be like that, Elizabeth
finds herself driven to imagine that Auntie Muriel either has been psychologically stunted or is malignant. Trying out the objective attitude,
Elizabeth asks,
Why is she friendless? Elizabeth is aware of the way she ought
to be thinking. Shes read magazines and books, she knows the
lines. Auntie Muriel was thwarted in youth. She had a domineering father who stunted her and wouldnt let her go to college because college was for boys. She was forced to embroider.... Auntie
Muriel had a strong personality and a good mind and she was not
pretty, and patriarchal society punished her. These things are
alltrue.
Nevertheless, Elizabeth can forgive Auntie Muriel only in
theory. Given her own sufferings, why has Auntie Muriel chosen
to transfer them, whenever possible, to others? Elizabeth can still
see herself, at the age of twelve, writhing on her bed with her first
menstrual cramps, nauseated with pain. Auntie Muriel standing
over her holding the bottle of aspirin out of reach. This is Gods
punishment. 34

Elizabeth cannot make the objective attitude work. Auntie Muriel

looks too much like someone who makes choices, but they are not the
choices of someone of whom one could expect good will. Auntie Muriel
is an old reptile, a malevolent vitality; and such malevolent vitality
cannot die. Hitler lived on after the discovery of his smoldering teeth,
and Auntie Muriel too is one of the immortals.35 Elizabeth has never
thought of her aunt as compounded of mortal flesh like other people;
rather as being, from neck to knees, built of a warty growth, something
33. Margaret Atwood, Life before Man (New York: Fawcett Crest, 1979), 133.
34. Ibid., 132.
35. Ibid., 345, 324.

239

MOR A L A IMS

like gum rubber, impermeable and indestructible.36 She is more and

less than human, impermeable and indestructible by others hatred.
Auntie Muriel has removed herself from the world of interpersonal relationships. Forgiveness is out of the question.
It will not do to object that Auntie Muriel is exceptionally flawed
and that it is unfair to want an aspirational story told for her. Lesser flaws
enable us to drive the wedge between action or singular character trait
and self. They enable us to offer minimalist, but not aspirational, forgiveness. Auntie Muriel needs a story.
Auntie Muriel needs a story, and in general we need stories of aspirational forgiveness because, at some point, we are likely to come to
the same realization that Lesje does in Life before Man: She believed
that if people could see how they were acting, they would act some other
way. Now she knows this isnt true.37 At the point where one loses the
innocent faith that there will always be some passable excuse or repentant apology when others disappoint us, one of two things must happen:
either one despairingly concludes that some (many) humans are not
really persons (but malignant vitalities or psychologically stunted individuals) or one finds some different way of making sense of how they
could mean to harm.

III. TELLING THE ASPIR ATIONA L STORY

What underlies the assumption that only the psychologically stunted,
the malignant, or the indifferent could really mean to harm or could
resist demands for reform even after they clearly see what they are
doing? It is, I think, the belief that normal persons, with normal capacities for goodwill, need above all to be able to make moral sense of their
choices and actions. To be a normal person is to identify the most sensible (or, if you like, rational) thing to do with the morally justified thing
to do. It is, thus, to be continually open to demands for better behavior.
Only those who are seriously defective could refuse to prioritize making
moral sense of their lives and thus could resist demands for reform.
36. Ibid., 321.
37. Ibid., 341.

240

CH A NGING ONES HEA RT

The insistence that individuals must deserve forgiveness also rests on

this same assumption. The insistence on desert is motivated by the belief
that to treat someone as a person is just to hold them up to moral standards
by judging them within the moral court. It is also to hold them to those
standards by demanding that disappointing individuals either provide a
passable account of themselves or, failing that, reform. Requiring that forgiveness be deserved is a way of enforcing the expectation that persons be
able to make moral sense of their actions. Those who can (by offering excuses or justifications for harming) get forgiven; those who cannot get resentfully requested to repent and reform. Should they refuse to repent or
should their promises of reform be proved false, they will have confirmed
their diminished personhoodthey must be psychologically stunted,
malignant, or indifferent to be so resistant to moral demands.
By contrast, aspirational forgiveness presupposes a quite different
picture of what it means to be a normal person with normal capacities for
goodwill. To the extent that persons are the sorts of individuals toward
whom reactive attitudes are appropriate and interpersonal relations are
possible, they must also be the sorts of individuals who are responsive
to moral demands and who need to make moral sense of their choices.
But normal persons also live through time, serially confronting different
configurations of events, obstacles, unasked-for responsibilities, internal needs and motivations, others sometimes irrational demands and
needs, and so on. In living through time, normal persons need to make
the sorts of choices that will add up to and sustain an integrated, rather
than fragmented, biography. They need their actions to make sense
within, or to make sense of, their past and projected future lives. What
I suggest is that aspirational forgiveness is achieved by seeing that, although an agents wrongdoing fails to make moral sense, it does make
biographical sense. I also suggest that a commitment to going beyond
a merely minimalist forgiveness is a commitment to deprioritizing the
moral and to seeing that there may be equally important ways that
normal persons of goodwill need to make sense of their lives. 38
38. In thinking about self-integration, I had in the back of my mind Bernard Williamss
Persons, Character, and Morality, in Moral Luck (Cambridge: Cambridge University
Press, 1981).

241

MOR A L A IMS

Telling the aspirational story requires hunting for a deeper reading

of unexcused and unrepented wrongdoing, one that tries to understand
how culpable wrongdoing fits into the larger pattern of a persons life.
This perspective does not assume that the surface messages of injurious
actions (about what the agent thinks of our worth, of what is permissible,
and the like) are all there is to the meaning of causing harm. Instead, this
perspective assumes that what gets said by persons through their actions
is often something about themselves and their lives, not just about their
moral views. It thus charitably assumes that wrongdoing is less likely to
be a blow directly aimed at us than to be shrapnel from something else
more complicated and more interesting in the personslife.
What was Auntie Muriel trying to say? At the end of Life before Man,
Auntie Muriel is dying of cancer. In her last conversation with Elizabeth,
Auntie Muriel speaks of her relation to Elizabeths alcoholic mother.
you didnt know. You think I was hard on her but I gave her money,
all those years. It wasnt your Uncle Teddy.... I left instructions
at my bank.... She hated me. She wouldnt see me, she used to
call me on the phone when she was drunk and say.... But I did
my duty. It was what Father would have wanted. Your mother was
always the favorite.
To Elizabeths horror, Auntie Muriel is beginning to cry. Tears
seep from her puckered eyes; a reversal of nature, a bleeding
statue, a miracle. 39

In her constantly and self-righteously harping on the sacrifices she

has made for Elizabeth, in her petty wish to have her miserly gifts and
chipped discards back after her sisters death, in her uncompromising
insistence on decency to the point of being willing to pay Elizabeths
husband to stay in a miserable marriage, there is protest. It is not fair
that she was not her fathers favorite daughter. It is not fair that her sister
has stolen her fathers love, saddled her with her children, lived and
boozed off her charity, and has hated her for it. And what does Auntie
Muriel get in return? No ones love, not her fathers, not her sisters, not
39. Atwood, Life Before Man, 327.

242

CH A NGING ONES HEA RT

Elizabethsonly the small consolation that she has done her duty and
tried to live a decent life. And she will not allow Elizabeth to take this
from her by approving of Elizabeths indecent lifestyle. Nor will she
allow her sister to keep her gifts after deathchipped discards they may
be, but her sister has taken enough.
Auntie Muriel is not a malignant vitality, one of the frozen hearted.
It is not that she does not care about others resentment, hatred, and dislike of her; and she has, perhaps, cared too much about what some others
felt toward herher father, for instance. Given what she has suffered
undeservedly within her family, the price of others goodwill seems to
her unreasonably high.
It is possible to enter sympathetically into a life like Auntie Muriels.
In doing so, Auntie Muriels meanness looks less like meanness pure
and simple and more like meanness expressing protest against the unfairness and lovelessness of her own life. The cold stinginess displayed
toward her sister, for instance, makes sense against the backdrop of her
fathers unwarranted favoritism and her sisters ungrateful hatred. To be
generous and forgiving toward her sister, to take loving care of her sisters daughter, would only compound the injustice of her sisters having
received so much without deserving it. Similarly, her rigid insistence
to Elizabeth on the importance of moral decency and the necessity of
making sacrifices makes sense against the backdrop of her having spent
her life compensating for her sisters self-indulgent lack of moral decency. It is Auntie Muriels way of ensuring that Elizabeth will not do to
others what Auntie Muriels sister has done to her: unfairly make them
pay for her own pleasures.
One might argue that in entering sympathetically into Auntie Muriels life and placing her meanness, rigidity, and stinginess in biographical context, all one is doing is adopting a broader perspective from which
to assess whether Auntie Muriels actions make moral sense. That is, one
might claim that the inclusion of biographical context does not represent a different perspective from which Auntie Muriels actions might
make a different, nonmoral kind of sense. Instead it represents the same
moral perspective (from which the question of Auntie Muriels deserving forgiveness might be answered), but broadened now to catch a fuller
range of possible moral justifications or excuses.
243

MOR A L A IMS

It is true that one might appeal to biographical narratives to try to

make moral sense of persons wrongdoing. Although moral assessments
of persons typically are made by slicing their actions and character traits
out of biographical context, there is in principle no reason moral assessments cannot be finely sensitive to contextual details. In Auntie Muriels
case, however, whatever sense appealing to her biographical narrative
helps us make of her wrongdoing, it is not moral sense.
Even after the telling of Auntie Muriels story, Elizabeths question
retains its full force: Given her own sufferings, why has she chosen to
transfer them, whenever possible, to others? She certainly is not justified in doing so. Her fathers and sisters inadequate love hardly justifies
her loveless treatment of Elizabeth. Nor does the absence of gratitude
for doing her duty and making sacrifices justify imposing a painful and
rigid conception of the duties of decency on Elizabeth. Similarly, the
story of Auntie Muriels life does not excuse her moral failings. One can
imagine even Auntie Muriel having made a more praiseworthy response
to the unfairness in her luck. She might instead have vowed to become
a more loving mother to Elizabeth than her father was to her or than her
sister was to Elizabeth. Although given her story, one can understand
how Auntie Muriel has become the person she is and why she might
resist repentance and reform, one is still left with the thought that she
could have done otherwise. Her meanness is still culpable meanness.
Because the aspirational story offers neither justifications nor excuses for Auntie Muriel, it fails to make her someone of whom we might
approve. She will still fix Elizabeth with her disapproving glare. She will
meddle. She will continue having those stiff family dinners at Christmas
that make everyone miserable. And all of this will seem too clearly part
of Auntie Muriel. That is just the way she is and how she wants to be. And
it is not pleasant. Auntie Muriel is substandard as a person, and Elizabeth will still probably feel the urge to throw a bowl at her.
Moreover, because the aim of the aspirational story is to make sense
of culpable wrongdoing by understanding how it might be a sensible
way of continuing a biography, such stories cannot wipe away sins.
As Elizabeth says, she cannot give absolution.40 The minimalist story
40. Ibid., 328.

24 4

CH A NGING ONES HEA RT

does have this effect. Its aim is to show that the true self did not mean to
injure. There was not, really, something to feel hurt about, something to
resent. The minimalist story retracts our perception of the past; an expression of ill will did not really happen. The aspirational story has quite
the opposite effect. It confirms our perception of the past; the injury is
here to stay. It cannot be wiped away because the agents true self meant
it and will not retract what she did. She would do it again.
So where does forgiveness enter in? What does aspirational forgiveness mean if it does not mean wiping the slate clean, reapproving the
other as someone basically good, wanting to go on with this person who
has, contrary to appearances, turned out to be the sort of person one
might really want to go on with? I think it means simply this: that one
stops demanding that the person be different from what she is. Having
come to the point of understanding that an indecent flawlike Auntie
Muriels rigid insistence on decency, with all the harm that causesis
the persons way of holding her life together, one also sees the cruelty
and disrespect for sense-making choices involved in demanding change.
One may still put the person on moral trial and find her wanting. But aspirational forgiveness is the choice not to demand that she improve. It is
the choice to place respecting anothers way of making sense of her life
before resentfully enforcing moral standards.
Such forgiveness cannot be obligatory, for at least two reasons. First,
although intimates may have a duty to try to understand how choices
make sense within each others lives, the sort of sympathetic entrance
into anothers life necessary for telling an aspirational story is sufficiently burdensome that it would be unreasonable to require that we do
this for everyone. Second, and more important, we cannot be obligated
to refrain from demanding that persons make moral sense of their actions. To treat someone as a person is, in part, to see them as continuously subject to moral requirements. It simply is not all that is involved
in treating someone as a person. (In addition, to treat someone as a
person is to treat her as someone for whom sustaining an integrated biography matters.) Thus, although we may recognize that persons also
need to act in ways that are responsive to their own histories, we cannot
make this need an overriding consideration without jeopardizing the
personhood of persons. What one hopes is that people will live up to
245

MOR A L A IMS

moral requirements while at the same time acting in ways that make
sense within the biography of their lives. The Auntie Muriels of the
world pose a problem, because the sorts of actions that make sense to
them, given their histories, are not ones that make moral sense. Morality
and self-integration have come apart for them, so that we must choose
between treating them as persons by resentfully making moral demands
or treating them as persons by forgivingly understanding how they have
made sense of their lives.
The choice to forgive under these circumstances forces upon us a
second choice, one that we might prefer never to have to make: either we
go on with her, accepting that she cannot be who we want her to be, and
knowing what going on will cost; or we disengage, removing ourselves
from harms way.

246

BIBLIOGR A PH Y

Addelson, Kathryn Pyne. The Man of Professional Wisdom. In Discovering Reality, edited by Sandra Harding and Merrill B. Hintikka, 165186. Boston:
D. Reidel, 1983.
Addelson, Kathryn Pyne. Moral Passages: Toward a Collectivist Moral Theory.
New York: Routledge, 1994.
Aldrich, Virgil C. An Ethics of Shame. Ethics 50 (1939): 5777.
Anderson, Elizabeth. The Imperative of Integration. Princeton: Princeton University Press, 2010.
Atwood, Margaret. Life Before Man. New York: Fawcett Crest, 1979.
Austin, J. L. Philosophical Papers, 3rd ed. Edited by J. O. Urmson and G. J. Warnock, 175204. Oxford: Oxford University Press, 1979.
Baier, Annette C. What Do Women Want in a Moral Theory? Nous 19, no. 1
(1985): 5363.
Baier, Annette C. Hume, the Womens Moral Theorist. In Women and Moral
Theory, edited by Eva Feder Kittay and Diana T. Meyers, 3755. Totowa, NJ:
Rowman & Littlefield, 1987.
Baron, Marcia. The Alleged Moral Repugnance of Acting from Duty. Journal of
Philosophy 81 (1984): 197220.
Bartky, Sandra Lee. Toward a Phenomenology of Feminist Consciousness. In
Feminism and Philosophy, edited by Mary Vetterling-Braggin, Frederick A.
Elliston, and Jane English. Totowa, NJ.: Littlefield, Adams, 1981.
Bartky, Sandra Lee. Femininity and Domination: Studies in the Phenomenology of
Oppression, 83119. New York: Routledge, 1990.
Becker, Marvin. Civility and Society in Western Europe, 13001600. Bloomington:
Indiana University Press, 1988.

247

BIBLIOGR A PH Y

Bedford, Errol. Emotions. In What Is an Emotion? edited by Cheshire Calhoun

and Robert C. Solomon, 265278. Oxford: Oxford University Press, 1984.
Behabib, Seyla. Situating the Self: Gender, Community and Postmodernism in Contemporary Ethics. New York: Routledge, 1992.
Berger, Peter L., and Thomas Luckmann. The Social Construction of Reality. New
York: Doubleday, 1966.
Bergmann, Frithjof. A Monologue on the Emotions. In Understanding Human
Emotions, edited by Fred D. Miller Jr. and Thomas W. Attig, 117. Bowling
Green Studies in Applied Philosophy, vol. 1. Bowling Green, OH: Bowling
Green State University, 1979.
Bleier, Ruth. Lab Coat: Robe of Innocence or Klansmans Sheet? In Feminist
Studies/Critical Studies, edited by Teresa de Lauretis, 5566. Bloomington:
Indiana University Press, 1986.
Blum, Lawrence. Community and Virtue. In How Should One Live: Essays on the
Virtues, edited by Roger Crisp, 231250. Oxford: Clarendon, 1996.
Blustein, Jeffrey. Care and Commitment: Taking the Personal Point of View. New
York: Oxford University Press, 1991.
Buss, Sarah. Appearing Respectful: The Moral Significance of Manners. Ethics
109 (1999): 795826.
Cahoone, Lawrence. Response to Alan Wolfe: Are We Losing Our Virtue? The
Case of Civility. In Civility, edited by Leroy Rouner, 142150. Notre Dame,
IN: University of Notre Dame, 2000.
Calhoun, Cheshire. Family Outlaws: Rethinking the Connections between
Feminism, Lesbianism, and the Family. In Feminism and Families, edited by
Hilde Lindemann Nelson, 131150. New York: Routledge, 1996.
Card, Claudia. Lesbian Choices. New York: Columbia University Press, 1995.
Card, Claudia. The Unnatural Lottery: Character and Moral Luck. Philadelphia:
Temple University Press, 1996.
Chisholm, Roderick M. Supererogation and Offence: A Conceptual Scheme for
Ethics. Ratio 5 (1963): 114.
Cummiskey, David. Kantian Revolutions. Unpublished manuscript.
Daly, Mary. Gyn/Ecology: The Metaethics of Radical Feminism. Boston: Beacon,
1978.
Davis, Kevin R. Kants Different Publics and the Justice of Publicity. KantStudien 83 (1992): 170184.
Deigh, John. Shame and Self-Esteem: A Critique. Ethics 93 (1983): 225245.
Dennett, Daniel. Conditions of Personhood. In Identities of Persons, edited by
Amelie O. Rorty, 175196. Berkeley: University of California Press, 1976.
Dickens, Charles. A Christmas Carol and Other Haunting Tales. New York: Doubleday, 1998.
Downie, R. S. Forgiveness. Philosophical Quarterly 15 (1965): 128134.
Driver, Julia. The Suberogatory. Australasian Journal of Philosophy 70 (1992):
286295.
248

BIBLIOGR A PH Y

Duberman, Martin. Cures: A Gay Mans Odyssey. New York: Plume, 1992.
Feinberg, Joel. Supererogation and Rules. Ethics 71 (1961): 276288.
Ferguson, Ann. Is There a Lesbian Culture? In Lesbian Philosophies and Cultures, edited by Jeffner Allen, 6388. Albany, NY: SUNY Press, 1990.
Frankfurt, Harry. Freedom of the Will and the Concept of a Person, Journal of
Philosophy 68 (1971): 520.
Frankfurt, Harry. Identification and Wholeheartedness. In Responsibility, Character, and the Emotions: New Essays in Moral Psychology, edited by Ferdinand
Schoeman, 2745. Cambridge: Cambridge University Press, 1987.
Friedman, Marilyn. What Are Friends For? Feminist Perspectives on Personal Relationships and Moral Theory. Ithaca, NY: Cornell University Press, 1993.
Frye, Marilyn. The Politics of Reality: Essays in Feminist Theory. Berkeley, CA:
Crossing Press, 1983.
Frye, Marilyn. Willful Virgin: Essays in Feminism 19761992. Berkeley, CA:
Crossing Press, 1992.
Gilligan, Carol. In a Different Voice. Cambridge, MA: Harvard University Press,
1982.
Green, Ronald M. The First Formula of the Categorical Imperative as Literally
a Legislative Metaphor. History of Philosophy Quarterly 8 (1991): 163179.
Grimshaw, Jean. Philosophy and Feminist Thinking. Minneapolis: University of
Minnesota Press, 1986.
Gutmann, Amy, and Dennis Thompson. Moral Conflict and Political Consensus. In Liberalism and the Good, edited by R. Bruce Douglass, Gerald
M.Mara, and Henry S. Richardson, 125147. New York: Routledge, 1990.
Hacking, Ian. Making Up People. In Reconstructing Individualism: Autonomy,
Individuality, and the Self in Western Thought, edited by Thomas C. Heller,
Morton Sosna, and David E. Wellbery, 222236. Stanford, CA: Stanford
University Press, 1986.
Hampton, Jean. Forgiveness, Resentment and Hatred. In Forgiveness and Mercy,
edited by Jeffrie G. Murphy and Jean Hampton, 3587. Cambridge: Cambridge University Press, 1988.
Held, Virginia. Feminism in Moral Theory. In Women and Moral Theory, edited
by Eva Feder Kittay and Diana T. Meyers, 111128. Totowa, NJ: Rowman &
Littlefield, 1987.
Herman, Barbara. The Practice of Moral Judgment. Cambridge, MA: Harvard University Press, 1993.
Herman, Barbara. Moral Literacy. The Tanner Lectures on Human Values. Delivered at Stanford University, April 2324, 1997, 372. <http://tannerlectures.
utah.edu/lecture-library.php#g>.
Herman, Barbara. Morality and Everyday Life. Proceedings and Addresses of the
American Philosophical Association 74, no. 2 (2000): 2945.
Heyd, David. Supererogation: Its Status in Ethical Theory. Cambridge: Cambridge
University Press, 1982.
249

BIBLIOGR A PH Y

Hill Jr., Thomas E. Autonomy and Self-Respect. Cambridge: Cambridge University

Press, 1991.
Hill Jr., Thomas E. Dignity and Practical Reason in Kants Moral Theory. Ithaca,
NY: Cornell University Press, 1992.
Hoagland, Sarah Lucia. Lesbian Ethics: Toward New Value. Palo Alto, CA: Institute of Lesbian Studies, 1988.
Hochschild, Arlie Russell. The Sociology of Feeling and Emotion: Selected Possibilities. In Another Voice: Feminist Perspectives on Social Life and Social Science. Edited by Marcia Millman and Rosabeth Moss Kanter, 280307. New
York: Anchor, 1975.
Hochschild, Arlie Russell. The Managed Heart: Commercialization of Human Feeling. Berkeley: University of California Press, 1983.
Hunting, Claudine. The Philosophes and the Question of Black Slavery. Journal of the History of Ideas 39 (1978): 405418.
Isaacs, Tracy. Cultural Context and Moral Responsibility. Ethics 107, no. 4
(1997): 670684.
Kahan, Dan M. What Do Alternative Sanctions Mean? University of Chicago
Law Review 63 (1996): 591653.
Kant, Immanuel. Grounding for the Metaphysics of Morals. Translated by James W.
Ellington. Indianapolis, IN: Hackett, 1981.
Kant, Immanuel. Perpetual Peace and Other Essays on Politics, History, and Morals.
Translated with an introduction by Ted Humphrey. Indianapolis, IN: Hackett, 1983.
Kant, Immanuel. The Metaphysics of Morals. Translated by Mary J. Gregor. Cambridge: Cambridge University Press, 1991.
Kasson, John F. Rudeness and Civility: Manners in Nineteenth Century Urban
America. New York: Hill and Wang, 1990.
Kekes, John. The Great Guide of Human Life. Philosophy and Literature 8
(1984): 236249.
Kekes, John. Shame and Moral Progress. In Ethical Theory: Character & Virtue,
edited by Peter A. French, Theodore E. Uehling Jr., and Howard K. Wettstein, 282296. Midwestern Studies in Philosophy, vol. 13. Notre Dame, IN:
University of Notre Dame Press, 1988.
Kingwell, Mark. A Civil Tongue: Justice, Dialogue and the Politics of Pluralism. University Park: Pennsylvania State University Press, 1995.
Kolnai, Aurel. Forgiveness. Proceedings of the Aristotelian Society 74 (1973
1974): 91106.
Korsgaard, Christine M. Kants Formula of Universal Law, Pacific Philosophical
Quarterly 66 (1985): 2447.
Lauritzen, Paul. Forgiveness: Moral Prerogative or Religious Duty? Journal of
Religious Ethics 15 (1987): 141154.
Lehtinen, Ullaliina. How Does One Know What Shame Is? Epistemology, Emotions, and Forms of Life in Juxtaposition. Hypatia 13 (1998): 5677.
250

BIBLIOGR A PH Y

Lugones, Maria. Playfulness, World-Traveling, and Loving Perception. Hypatia 2 (1987): 319.
Lugones, Maria. Hispaneando y Lesbiando: On Sarah Hoaglands Lesbian
Ethics. Hypatia 5 (1990): 138146.
Lugones, Maria. On the Logic of Pluralist Feminism. In Feminist Ethics, edited
by Claudia Card, 3544. Lawrence: University Press of Kansas, 1991.
Margalit, Avishai. The Decent Society. Cambridge, MA: Harvard University Press,
1996.
Martin, Judith. Miss Manners Rescues Civilization from Sexual Harassment, Frivolous Lawsuits, Dissing and Other Lapses in Civility. New York: Crown, 1996.
McConnell, Terrance C. Moral Blackmail. Ethics 91 (1981): 544567.
McFall, Lynne. Integrity. Ethics 98 (1987): 520.
Moody-Adams, Michelle M. The Idea of Moral Progress. Metaphilosophy 30,
no. 3 (1999): 168185.
Moore, Kathleen Dean. Pardons: Justice, Mercy and the Public Interest. Oxford:
Oxford University Press, 1989.
Morgan, Kathryn Pauly. Women and Moral Madness. In Science, Morality, and
Feminist Theory, edited by Marsha Hanen and Kai Nielsen, 201226. Calgary: University of Calgary Press, 1979.
Morton, Nell. The Journey Is Home. Boston: Beacon Press, 1985.
Mulholland, Leslie. Kant: On Willing Maxims to Become Laws of Nature. Dialogue 18 (1978): 92105.
Murphy, Jeffrie G. Kant: The Philosophy of Right. New York: St. Martins, 1970.
Murphy, Jeffrie G. Moral Death: A Kantian Essay on Psychopathy. In Ethics and
Personality: Essays in Moral Psychology, edited by John Deigh, 207222. Chicago: University of Chicago Press, 1992.
Murphy, Jeffrie G., and Jean Hampton, Forgiveness and Mercy. Cambridge: Cambridge University Press, 1988.
Nagel, Thomas. Mortal Questions. Cambridge: Cambridge University Press, 1979.
Nagel, Thomas. Concealment and Exposure. Philosophy & Public Affairs 27
(1998): 330.
North, Joanna. Wrongdoing and Forgiveness. Philosophy 62 (1987): 499508.
Nussbaum, Martha C. The Fragility of Goodness: Luck and Ethics in Greek Tragedy
and Philosophy. Cambridge: Cambridge University Press, 1986.
OHear, Anthony. Guilt and Shame as Moral Concepts. Proceedings of the Aristotelian Society 77 (19761977): 7386.
ONeill, Onora. Acting on Principle: An Essay on Kantian Ethics. New York: Columbia University Press, 1975.
ONeill, Onora. Constructions of Reason: Explorations of Kants Practical Philosophy. Cambridge: Cambridge University Press, 1989.
Orwin, Clifford. Civility. American Scholar 60 (1991): 553564.
Piers, Gerhart, and Milton B. Singer, eds. Shame and Guilt: A Psychoanalytic and a
Cultural Study. Springfield, IL: Charles C. Thomas, 1953.
251

BIBLIOGR A PH Y

Piper, Adrian. Passing for White, Passing for Black. In Passing and the Fictions of
Identity, edited by Elaine Ginsberg, 234270. Durham, NC: Duke University
Press, 1996.
Rawls, John. A Theory of Justice. Cambridge, MA: Harvard University Press, 1971.
Rawls, John. Political Liberalism. New York: Columbia University Press, 1993.
Richards, Norvin. Forgiveness. Ethics 99 (1988): 7797.
Sabini, John, and Maury Silver. Moralities of Everyday Life. Oxford: Oxford University Press, 1982.
Shoeman, Ferdinand. Statistical Norms and Moral Attributions. In Responsibility, Character, and the Emotions: New Essays in Moral Psychology, edited by
Ferdinand Shoeman, 287315. Cambridge: Cambridge University Press,
1987.
Sinopoli, Richard C. Thick-Skinned Liberalism: Redefining Civility. American
Political Science Review 89 (1995): 612620.
Spelman, Elizabeth V. Anger and Insubordination. In Women, Knowledge, and
Reality: Explorations in Feminist Philosophy, edited by Ann Garry and Marilyn Pearsall, 263273. Boston: Unwin Hyman, 1989.
Strawson, P. F. Freedom and Resentment and Other Essays. London: Methuen,
1974.
Tangney, June Price. Shame & Guilt in Interpersonal Relationships. In SelfConscious Emotions: The Psychology of Shame, Guilt, Embarrassment and Pride,
edited by June Price Tangney and Kurt W. Fischer, 113139. New York: Guilford, 1995.
Tangney, June Price, and Kurt W. Fischer, eds. Self-Conscious Emotions: The Psychology of Shame, Guilt, Embarrassment and Pride. New York: Guilford, 1995.
Taylor, Gabriel. Pride, Shame, and Guilt: Emotions of Self-Assessment. Oxford:
Clarendon Press, 1985.
Urmson, J. O. Saints and Heroes. In Moral Concepts, edited by Joel Feinberg,
6073. London: Oxford University Press, 1969.
Walker, Margaret Urban. Moral Understandings: A Feminist Study in Ethics. New
York: Routledge, 1998; 2nd ed. (New York: Oxford University Press, 2007).
Walzer, Michael. Civility and Civic Virtue. Social Research 41 (1974): 593611.
Washington, George. George Washingtons Rules of Civility and Decent Behaviour in Company and Conversation. Edited, with an introduction by Charles
Moore. Boston: Houghton Mifflin, 1926.
Watson, Gary. Free Agency. Journal of Philosophy 72 (1975): 205220.
Watson, Gary. Responsibility and the Limits of Evil: Variations on a Strawsonian Theme. In Responsibility, Character, and the Emotions, edited by Ferdinand Schoeman, 256286. Cambridge: Cambridge University Press, 1987.
Williams, Bernard. A Critique of Utilitarianism. In Utilitarianism: For and
Against, edited by Bernard Williams and J. J. C. Smart, 77150. Cambridge:
Cambridge University Press, 1973.

252

BIBLIOGR A PH Y

Williams, Bernard. Moral Luck: Philosophical Papers 19731980. Cambridge:

Cambridge University Press, 1981.
Williams, Bernard. Shame and Necessity. Berkeley, CA: University of California
Press, 1993.
Wolf, Susan. Asymmetrical Freedom. Journal of Philosophy 77, no. 3 (1980):
151166.
Wolf, Susan. Above and Below the Line of Duty. Philosophical Topics 14, no. 2
(1986): 131148.
Young, Iris Marion. Justice and the Politics of Difference. Princeton, NJ: Princeton
University Press, 1990.
Zwiebach, Burton. Civility and Disobedience. New York: Cambridge University
Press, 1975.

253

IN DE X

Note: Locators followed by the letter n refer to notes.

abnormal moral contexts, 3637, 45, see
also excusing, in abnormal moral
contexts; injustice, conventionalized;
moral revolutionary
defined, 196
and feminist critique, 196198
and the obligatory, 206
Addelson, Kathryn Pyne, 190192
agent, see maturity, moral; identity, as
minimally well-formed agent
Aldrich, Virgil, 47, 50
Atwood, Margaret, 238, see also Life Before
Man
Auntie Muriel, 238240
autonomy, see integrity, and autonomy;
moral maturity, and autonomy;
shame, and autonomy
Baier, Annette, 1516
Baron, Marcia, 214
Bartky, Sandra, 5860
Becker, Marvin, 76n1
biographical sense, see sense-making,
biographical
Bleier, Ruth, 188
Blum, Lawrence, 119n24, 121
Blustein, Jeffrey, 123,
135, 142

Card, Claudia, 2930, 45

Categorical Imperative procedure, see
universalization
Chisholm, Roderick, 109
civility, 1920, see also civility, bounds
of; morally imperfect social world,
nature of civility in; subordination,
and civility
as communicative display, 8490
difference from treating with respect,
8590
and etiquette or manners, 7576
and liberal democracy, 8081, 9495
as polite virtue, 8183
scope of, 7778
and social norms, 76, 85, 89, 91
socially critical analysis of, 77, 9293
value of, 9192
civility, bounds of, see also function, of
civility
defined, 93
set from a socially critical viewpoint,
9394, 9798
set by social consensus, 9899
common decency, 2021, see also moral
gifts
as constructed category, 113114,
119120

255

INDEX
common decency (continued)
expected of minimally well-formed moral
agents, 116119, see also identity, as
minimally well-formed agent
as fulfilling minimal moral obligations,
106, 106n4, 106n5, 116117, 243
and moral gift giving conventions, 108,
119121
as motivationally nontaxing and
unambiguously good, 117120
not fully morally elective, 108111
relation to obligation, 111113
relation to supererogation, 107111
critically reflective viewpoint, 56, 76, 89,
see also civility, bounds of, set from a
socially critical viewpoint; getting it
right; morality, gap between socially
accepted and correct
Kantian and utilitarian, 77, 89, 9091
and shame, 70
Daly, Mary, 189190
decency, as grading term, 105
Deigh, John, 52n8
Dennett, Daniel, 237
doing the right thing, see getting it right
Driver, Julia, 109, 109n14
Duberman, Martin, 128129, 133
emotional work
and benevolence, 220
and feeling rules, 214
as giving others their emotional due,
214
as management of others emotions,
215, 217220
ethics of care, 198, 213
excuses, 2728, 91n24, 117118, 190,
235236
and forgiveness, 226227, 230
excusing, in abnormal moral context, 192,
199201
sanctioning force of, 201205
failure see also moral performance
of legibility, 30, 3639, 43
of justification, 3639, 43

moral, 30, 35, 4344

nature of, 2728
Feinberg, Joel, 110
feminist critique, 3132, 35n13, 36, 37, 38,
42n17, 71, 129130, 186, 187188,
196198, 198n16, 200
forgiveness, see also excuses, and
forgiveness
aspirational, 223, 226227, see also
stories, of aspirational forgiveness
and condoning wrongdoing, 230234
deserved, 224225
elective, 225, 227, 228, 245, see also
moral gifts
and emotional work, 218, 219
minimalist, 223226, 227, 233, see also
stories, of minimalist forgiveness
paradox of, 225226
and repentance, 227230
Frankfurt, Harry, 124126
function
of civility to regulate dispute, 20, 79, 95,
97, 100
of morality, 12
of reproach, 156, 207
of shame, 60n17, 7072
getting it right, 1, 2, 14, 35n14, 40, 41, see
also critically reflective viewpoint
commitments involved in, 31
and ideal of life beyond reproach, 35
and integrity, 21, 140141, 140n34,
140n35, 149153
limited by available concepts and
methods, 7, 42n17
limited by available socially critical
knowledge, 179180
Gilligan, Carol, 198
gratitude, misplaced, 32, 37,
110111, 206
Gutmann, Amy, 81, 94, 96
Herman, Barbara, 158, 166
and routine moral performance, 12
Heyd, David, 108n10, 111n18
Hill Jr., Thomas E., 126
Hochschild, Arlie Russell, 213

256

INDEX
identity
and commitments, 134135
duplicitous, see Lugones, Maria
as minimally well-formed agent, 116
moral, 207208
as psychological identification, 132134
within social practice of morality, 67,
71, 72
individualism, 194, 215
ignorance, 193
in abnormal moral context, 193, 196,
199200
in normal moral context, 194196
social determinants of, 205207
injustice, conventionalized, 4, 9, 37, 159,
see also abnormal moral contexts;
slavery; subordination
detecting, 22
and socialization, 173174, 177, 178,
179, 185
integrity, see also subordination, and
integrity; getting it right, and
integrity
and ambivalence, 128130
and autonomy, 144, 145, 147
and civility, 98, 100
clean hands picture of, 136140
defined, 141
and forgiveness, 246
and hypocrisy, 150151
identity picture of, 130136
and inconsistency, 127128
integrated-self picture of, 124130
as personal virtue, 144145, 146147
and self-indulgence, 146147
and selling out, 140143
as social virtue, 146, 148153
and wholeheartedness, 125128, 130
intuitions, 7, 45, 45n20, 69n33
justification, 4041
failure of, 3639
theory of, and integrity, 138139
Kant, Immanuel, see also resistance,
to unjust maxims; slavery;
universalization

and compliance with social norms,

159162, see also injustice,
conventionalized, and socialization
and cooperative schemes involving
roles, 167169
and promising, 172175
on the right to revolution, 162
universal law and the kingdom of ends,
181183
Kantianism, 90, 112113
and integrity, 137140
Kekes, John, 50, 106
Kingwell, Mark, 81
Kolnai, Aurel, 225, 228, 229, 232
Korsgaard, Christine M., 166n18, 172n23,
177n32
Life Before Man, 238240
Lugones, Maria, 127128
maturity, moral, 51
and autonomy, 63
and shame, 55, 59, 62
McFall, Lynne, 123, 135
Mill, John Stuart, 148
Miss Manners, 76, 7778, 8283, 9394,
94n25, 96
moral aims, 1718, 35, 4246, 99100,
101102, see also getting it right;
morality, social practice of
moral context, see abnormal moral
context; normal moral context
moral gifts, 106, 115116, 212, 227
conventions of as source of normativity,
114116
moral luck, 2930, 4243
moral mediation, see stories, and
moral mediation; moral theory,
mediator-centered
moral performance, see also failure
idiosyncracy of, 30, 45, 51, 52, 69, see
also moral revolutionary
intelligibility of, 10, 13, 14, 15, 1819
moral revolutionary, 17, 27, 46, 51, 56
moral theory
aim of, 1
dyadic, agent-centered, 215216, 220

257

INDEX
moral theory (continued)
mediator-centered, 215, see also stories,
and moral mediation
role of the hypothetical in, 24, 1213,
39, 41, 53, 172, see also resistance,
to unjust maxims
morally imperfect social world, 4, 2122,
31, 46, see also subordination
and integrity, 142143, 143n39
nature of civility in, 8788, 9091
morality
conceived as genuine rather than
supposed, 16, 78, 1417, 54
expressive-collaborative model of, 11
gap between socially accepted and
correct, 5, 3637, 89, 92
genuine vs. social, 38
requiring activity in, 3, 10
as scheme of social cooperation,
4042, see also morality, social
practice of
theoretico-juridical model of, 1012
and unacknowledged loan on social
practice of morality, 1516
morality, social practice of, 5, 65, 68, see
also identity, within social practice
of morality; understandings, shared
moral; morality, as scheme of social
cooperation
as actual rather than hypothetical, 11,
40, 41, 71
and civility, 99100
and condoning wrongdoing, 231
co-participants in, 6465, 66
and moral theorizing, 810
as only moral game in town, 10, 13,
19, 71
nature of, 68
nature of moral norms in, 56
representative viewpoint in, 6567,
69, 72
Mulholland, Leslie, 164, 165
Murphy, Jeffrie G., 228, 229230, 232
Nagel, Thomas, 29, 85n18
nonideal, the, see morally imperfect social
world

normal moral context

defined, 194
transparency of, 194
norms, see also understandings, shared
moral; civility, and social norms
nature of, see morality, social practice of,
nature of moral norms in
North, Joanna, 223224, 233234
Nussbaum, Martha, 29
OHear, Anthony, 40
ONeill, Onora, 173
oppression, see subordination
Orwin, Clifford, 80
Piper, Adrian, 6062
practical weight, see weight, practical
progress, moral, 1617, 100, 176n31,
204
Rawls, John, 33, 41, 8081
reproach, 10, 142, 187, 206209, 220, see
also function, of reproach
of oneself, 35, 141
resentment, 117n22, 222225, 227228,
235, see also morality, requiring
activity in
resistance, 37
mutual, 184
principles of, 3234
to punishment, 183184
to unjust maxims, 175177
responsibility
assignment, 28n1
feminist views of, 186192
justification versus point of, 208209,
209n22, see also ignorance
taking of, 30
Richards, Norvin, 228, 229
Sabini, John and Maury Silver, 202
Scrooge, Ebenezer, 103104, 121122
self-legislation, 56, 194n12, 196, 207, see
also autonomy
sense-making
moral, 240241, 243244
biographical, 241, 244, 245

258

INDEX
shame, 28, 4344, see also critically
reflective viewpoint, and shame;
function, of shame; maturity, moral,
and shame; subordination, and
shame
in abnormal moral context, 52
and autonomy, 4958, 63
and hypothetical social world, 53
and integrity, 151
shame of the discriminating social
actor analysis of, 5457, 62, 63,
69
shame of the moral pioneer analysis
of, 5054, 61, 63, 69
as social emotion, 5253, 5557, 7172
slavery
definition of, 164165
Kantian arguments against, 163170,
170171n22
socially critical viewpoint, see critically
reflective viewpoint
stories
of aspirational forgiveness, 240,
242245
of minimalist forgiveness, 235238,
244245
and moral mediation, 218219
Strawson, P.F.
on reactive attitudes, 234235
on the objective attitude, 237238
suberogatory, 109
subordination, 32, see also injustice,
conventionalized, and socialization;
slavery
and civility, 100101
and integrity, 129130, 130n18
and promising, 174175
and reactive attitudes, 187188
and scientific practices, 189, 190192
and shame, 4849, 5862, 70

and social practices, 32, 186187, 188,

190
systematic, 198, 198n16
supererogation, see moral gifts; common
decency, relation to supererogation;
forgiveness, elective
Taylor, Gabriel, 123, 125, 134
Thompson, Dennis, 81, 94, 96
truth, indeterminacy of, 56, 9596, 149150
understandings, shared moral, 5, 10, 36,
40, 6465, 68, 101, 232, see also
moral performance, intelligibility of
universalization, 167169, 170n22,
177n32, see also Kant, Immanuel, and
cooperative schemes involving roles;
Kant, Immanuel, and promising
and socially critical knowledge,
178180
utilitarianism, 90, 112
and integrity, 137140
virtue, 31, 34, 144, 148n42, see also civility;
integrity, as personal virtue; integrity,
as social virtue; emotional work, and
benevolence
Walker, Margaret Urban, 6, 11, 1314, 68
Washington, George, 82
weight
epistemic, 6364, 67
practical, 64, 64n25, 6567, 72
social, 5657
Williams, Bernard, 50, 5457, 90, 123,
131132, 133, 136137, 140n35,
141142
Wolf, Susan, 193, 195
wrongdoing, conventionalized, see
injustice, conventionalized

259