On Emotions

This page intentionally left blank

On Emotions
Philosophical Essays

EDIT ED BY JOH N DEIGH

1
 3
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide.

Oxford New York

Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto

With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Th ailand Turkey Ukraine Vietnam

Oxford is a registered trademark of Oxford University Press

in the UK and certain other countries.

Published in the United States of America by

Oxford University Press
198 Madison Avenue, New York, NY 10016

© Oxford University Press 2013

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

On emotions : philosophical essays / edited by John Deigh.
p. cm.
Includes index.
ISBN 978–0–19–974019–2 (alk. paper)
1. Emotions (Philosophy) I. Deigh, John.
B815.O5 2013
128′.37—dc23
2012012985

ISBN 978–0–19–974019–2

1 3 5 7 9 8 6 4 2
Printed in the United States of America
on acid-free paper
In memory of Robert C. Solomon
1942–2007
This page intentionally left blank
 CONTENTS

Contributors ix
Acknowledgments xi

Introduction 1
JOHN DEIGH

1. Justice as an Emotion Disposition 14

ROBERT C . ROBERTS

2. Equality and Love at the End of The Marriage of Figaro: Forging

Democratic Emotions 29
M A RT H A C . N USSBAU M

3. Spectator Emotions 59
PA U L W O O D R U F F

4. Comic Amusement, Emotion, and Cognition 76

NOËL C A R ROLL

5. Intellectual and Other Nonstandard Emotions 99

MICH A EL STOCK ER

6. Authenticity and the Examined Life 124

JEROME NEU

7. Self-Knowledge and the Affirmation of Love 149

L AU R ENC E T HOM A S

8. Love and Death 159

K AT HL EEN M A R I E H IG GI NS

vii
viii Contents

9. Guilt in War 179

NA NC Y SHER M A N

10. Emotion and Personal Identity 198

C H A R L E S N USSBAU M

11. The Emergence of Emotion as an Object of Scientific Study 215

JOHN DEIGH

Index 233
 CON T R I BU TOR S

Noël Carroll is distinguished professor of philosophy at the Graduate Center,

The City University of New York.
John Deigh is professor of law and philosophy at the University of Texas
at Austin.
Kathleen Marie Higgins is professor of philosophy at the University of Texas
at Austin.
Jerome Neu is professor of humanities at the University of California, Santa
Cruz.
Charles Nussbaum is associate professor of philosophy at the University of
Texas at Arlington.
Martha C. Nussbaum is Ernest Freund Distinguished Service Professor of Law
and Ethics at the University of Chicago.
Robert C. Roberts is distinguished professor of ethics at Baylor University.
Nancy Sherman is university professor of philosophy at Georgetown
University.
Michael Stocker is Guttag Professor of Ethics and Political Philosophy Emeritus
at Syracuse University.
Laurence Thomas is professor of philosophy at Syracuse University.
Paul Woodruff is Darrell K. Royal Regents Professor in Ethics and American
Society at the University of Texas at Austin.

ix
This page intentionally left blank
 ACK NOW L E DG M E N TS

Many of the papers in this volume were given at the 5th annual Royal Conference
on Ethics at the University of Texas at Austin in 2008. The conference was sup-
ported by the Darrell K. Royal Professorship in Ethics and American Society.
I thank Paul Woodruff for this support. Two of the papers, Robert Roberts’s
“Justice as an Emotion Disposition” and Michael Stocker’s “Intellectual and
Other Nonstandard Emotions,” are reprinted. Roberts’s paper originally
appeared in Emotion Review 2 (2010): 36–43. I am grateful to Sage Publications
for permission to reprint it. Michael Stocker’s paper originally appeared in The
Oxford Handbook of Philosophy of Emotion, Peter Goldie, ed. (Oxford: Oxford
University Press, 2010), pp. 401–423.

xi
This page intentionally left blank
On Emotions
This page intentionally left blank
 Introduction

Robert Solomon’s Theory of

Emotions in Retrospect
John Deigh

The philosophical study of emotions among British and American thinkers

had become, by the midpoint of the twentieth century, an intellectual back-
water. On the European continent, by contrast, it thrived. Phenomenology was
a major movement on the continent, and its principal lines of investigation
included the application of its methods to the emotions. The history of twenti-
eth century Anglo-American philosophy is rather different. In Britain, analytic
philosophy had become the dominant school by the 1930s, and in the years
after World War II it spread to America through both the influence of its British
contingent and that of the distinguished émigré philosophers from Central
Europe who had, before coming to America, achieved distinction as exponents
of logical empiricism. Unlike phenomenology, the agenda of analytic philoso-
phy did not include the study of emotions. The topic was given scant attention
even by those in the analytic school who concentrated on questions in the phi-
losophy of mind.
The reasons had largely to do with how analytic philosophers, during this
period, saw their work in relation to that of the empirical sciences. Philosophy,
as they understood the discipline, did not study the same things as science.
Rather, on their view, science studied the workings of nature, and philosophy
complemented these studies by working out the logic and semantics of language
either with the aim of constructing an ideal language for encoding scientific
knowledge or of working out a vernacular understanding of the principles of
reasoning and the basic concepts that various branches of knowledge applied.
The relevant science for the study of emotions was experimental psychology,
and it treated emotions first and primarily as felt states of physiological arousal

1
2 on emotions

and later, with the rise of behaviorism, as overt behavioral responses to the
same stimuli as produce those states. In either case, philosophy, understood as
an analytic enterprise that complements science, had little to add.
The writings of major analytic philosophers during this period tellingly
illustrate this view. Bertrand Russell in his book The Analysis of Mind devoted
six of three hundred pages to emotions.1 After first conceding that emotions
do have characteristics that make special investigation of them by psycholo-
gists warranted, he remarked that his concern is with their analysis. Yet
within a sentence of this remark he wrote, “Although what specially concerns
us is the analysis of emotions, we shall find that the more important topic
is the physiological causation of emotions. This is a subject on which much
valuable and exceedingly interesting work has been done, whereas the bare
analysis of emotions has proved somewhat barren.”2 Th irty years later Gilbert
Ryle, in his book The Concept of Mind, had somewhat more to say on the topic.
He devoted a thirty-page chapter to it.3 Most of the chapter, however, was
concerned with establishing the negative thesis that emotions are not mental
states that one can conceive apart from the behavior that manifests them. The
chapter offered few positive ideas about emotions. As a source of such ideas,
philosophical analysis, as Ryle practiced it, was no less barren than it was as
Russell applied it.
Things began to stir in these stagnant waters toward the end of the 1950s
as the influence of Wittgenstein’s remarks on philosophical psychology in his
Philosophical Investigations and related unpublished writings grew. His remark
at PI §476, in particular, supplied a theme that others elaborated. “We should
distinguish,” he wrote, “between the object of fear and the cause of fear. Thus a
face which inspires fear or delight (the object of fear or delight), is not on that
account its cause but—one might say—its target.”4 G. E. M. Anscombe spot-
lighted the remark at a crucial juncture in her book Intention, though because
her interest in doing so was in distinguishing something she called a mental
cause from causes that conformed to the notion of causation in the natural sci-
ences, she did not expound its theme.5 Others, however, did.
Among the first were Bernard Williams and Philippa Foot. Williams, in
“Pleasure and Belief,” examined the nature of being pleased at something—his
chief example was pleasure at inheriting a fortune—and argued for the irre-
ducibility of this state to any casual relation between pleasure and either its

1
Bertrand Russell, The Analysis of Mind (London: George Allen & Unwin, 1921), pp. 279–284.
2
Ibid., p. 279.
3
Gilbert Ryle, The Concept of Mind (London: Hutchinson & Company, Ltd., 1949), pp. 83–115.
4
Ludwig Wittgenstein, Philosophical Investigations, G. E. M. Anscombe and Rush Rees, eds.
(Oxford: Basil Blackwell, 1953), §476.
5
G. E. M. Anscombe, Intention (Oxford: Basil Blackwell, 1957), §10.
 Introduction 3

object or the belief that its object obtains.6 The state, he proposed, has, instead,
the structure of attending to something, and it may therefore, he conjectured,
in virtue of such a structure, which philosophers sometimes use the notion of
intentionality to capture, be open to fertile philosophical analysis. Foot, too, in
her landmark paper “Moral Beliefs,” focused on the relation between an emo-
tion and its object and also maintained that it eluded accounts of emotion like
Ryle’s.7 Her point, however, was different from Williams’s. She observed that,
with respect to a specific type of emotion—her chief example was pride—the
relation limited what could qualify as an object of that type. Thus, she argued,
the object of pride must be something one sees both as one’s own and as an
asset, achievement, or advantage. Hence, pride cannot, pace Ryle, consist in
behavior. Rather whatever behavior manifests pride must be explicable as the
product of a belief that what one is proud of has the properties that qualifies
it as an object of pride. No matter the behavior, in other words, one could not
be proud of, say, the sky unless one had beliefs, doubtlessly delusional, about
owning the sky or saving it from falling.
In the decade that followed more philosophers in the analytic tradition
began to examine the relation of emotions to their objects. A small literature
developed, which confirmed Williams’s conjecture that emotions, in virtue of
their intentionality, are objects of fertile philosophical analysis. Notable con-
tributions to this literature were David Pears’s “Causes and Objects of Some
Feelings and Psychological Reactions,”8 Anthony Kenny’s Action, Emotion,
and Will,9 and George Pitcher’s “Emotion.”10 Pears discussed the issues that
Williams had raised. Kenny and Pitcher pursued themes similar to Foot’s. Still,
no Anglo-American philosopher working on the topic in this period put for-
ward a general theory of the emotions. The standard practice was to focus one’s
research on narrowly defined problems and to limit their treatment to essays
suitable for publication as an article in a philosophical journal. This practice
prevailed until the 1970s.
The breakthrough in the study of emotions came in 1976 with the publica-
tion of Robert Solomon’s book The Passions.11 Solomon had rehearsed the book’s
core themes a few years earlier in his article “Emotions and Choice.”12 The article

6
B. A. O. Williams, “Pleasure and Belief,” Proceedings of the Aristotelian Society, Supplementary
Volume, 33 (1959): 57–72.
7
Philippa Foot, “Moral Beliefs,” Proceedings of the Aristotelian Society 59 (1959): 83–104.
8
D. F. Pears, “Causes and Objects of Some Feelings and Psychological Reactions,” Ratio 4
(1962): 91–111.
9
Anthony Kenny, Action, Emotion, and Will (London: Routledge and Kegan Paul, 1963).
10
George Pitcher, “Emotion,” Mind 74 (1965): 326–346.
11
Robert C. Solomon, The Passions: The Myth and Nature of Human Emotion (Garden City,
NJ: Anchor Press/Doubleday, 1976).
12
Robert C. Solomon, “Emotions and Choice,” Review of Metaphysics 17 (1973): 20–41.
4 on emotions

was unusually bold, indeed audacious. Unsurprisingly, it attracted immediate

criticism. In the book, Solomon preserved the article’s brash sensibility, while
qualifying somewhat his statement of its themes. In addition, he presented a
more complete theory of emotion to support them. The book’s exposition com-
bined a polemical attack on the prevailing theories of emotion in experimental
psychology, those reflected in Russell’s and Ryle’s books, in particular, with a
far-reaching argument for an alternative theory. A prime target of this attack
was William James’s famous definition of emotion as the feeling of physiologi-
cal changes provoked by the perception of an exciting event. Solomon regarded
this definition as emblematic of the theories he attacked. His alternative the-
ory included, as a major tenet, the distinction between the cause and the object
of an emotion that had given rise to the philosophical literature I canvassed
above. But the theory went significantly beyond this literature in its account of
the nature of emotions. Solomon situated it within the philosophical method
and thought of phenomenology. He did so, he said, to avoid ontological ques-
tions about the nature of an emotion’s object.
The method of phenomenology was well suited to this purpose. Its chief
precept is that one study phenomenal experience without any presupposi-
tion about the reality of its objects. Consequently, adopting this method
frees one from having to deal with the sort of metaphysical disputes over
the nature of mental or intentional objects with which the literature on emo-
tions in Anglo-American philosophy had become preoccupied. Questions
about whether mental objects that have no physical counterparts are real in
some other way, for instance, and whether such objects are ultimately elim-
inable can be sidestepped. Accordingly, Solomon put forward what he called
a subjective theory of emotions. Emotions, being intentional states of mind,
disclose a personal world or world of the subject’s own making. The extent of
the match between the personal world an emotion discloses and the objective
world, the world that defi nes the subject’s objective circumstances, is of no
concern. The theory, rather, is concerned only with the world as it appears to
the subject in virtue of the emotion he or she is experiencing. And to mark
the radical difference between how emotions are conceived on this theory
and the conceptions of them that one fi nds in the prevailing theories in psy-
chology, Solomon identified emotions as judgments. What distinguishes the
two conceptions is the attribution to emotions of thought as an essential
component when they are conceived as experiences that disclose something
to their subjects. When they are conceived as mere feelings that physiologi-
cal processes produce or that behavior manifests, by contrast, they lack this
component. To say, then, that an emotion discloses a personal world to its
subject is, Solomon maintained, to say that the subject judges the world or
particular things in it to be characterized in a certain way in virtue of his
experiencing the emotion.
 Introduction 5

In taking emotions to be judgments, Solomon rejuvenated a conception of

emotions whose roots were in ancient thought and whose chief exponents in
the ancient world had been the Stoics. Like the ancient Stoics, Solomon identi-
fied emotions with evaluative judgments about oneself and one’s place in the
world. The evaluative judgments with which the Stoics identified emotions
were judgments of objects, people, and events as these were understood to
affect one’s good, what the ancient Greeks called eudaimonia. Thus, sadness is
a judgment that one has lost something of value; fear is a judgment that one is
being threatened with harm; shame is a judgment that one has disgraced one-
self before others whom one holds in high esteem. Solomon, in his account of
these judgments, replaced eudaimonia with dignity, self-esteem, or a sense of
personal worth. Emotions, Solomon argued, are judgments of objects, people,
and events in which one has invested oneself. Accordingly, they are judgments
of things that reflect well or badly on one or one’s life, and their reflection on
one or one’s life for good or ill is given in their being judged to be a loss, a
threat, a disgraceful act or condition, and so forth. Summing up, Solomon, to
distinguish the evaluative judgments with which he identified emotions from
the many disinterested and dispassionate ones that we make in daily life,
characterized the former as “self-involved and relatively intense” evaluative
judgments.13
While Solomon’s theory, in identifying emotions with evaluative judgments,
revived this major doctrine of ancient Stoicism, it differed from the Stoics’ the-
ory in one important respect. The Stoics took the evaluative judgments with
which they identified emotions to be objective. Indeed, they thought these
judgments were mostly false. For they treated the question of how to achieve
eudaimonia as one the truth of which a wise man could arrive at and held
that ordinary people, who lacked wisdom, misunderstood what things men
needed to have and what conditions they needed to be in to achieve eudaimo-
nia. Ordinary people’s turbulent emotional lives, the Stoics argued, testified
to their lack of wisdom and consequent propensity to misjudge what things
are good to have and what conditions are good to be in. Solomon, by contrast,
took the evaluative judgments with which he identified emotions to be subjec-
tive. That is, he understood them as concerned with things that affect one’s
self-esteem, and he took a person’s self-esteem to be based on the person’s
self-image, an image that results from his coming to regard some things about
himself as more important than others. Because there is no truth about what
things about oneself are more important and what things are less, because the
degree of their importance to one is a matter of the amount of importance
one places on them, the evaluative judgments one makes of those things nei-
ther represent nor misrepresent facts in the world. They are neither true nor

13
Passions, p. 187.
6 on emotions

false. Rather they consist in one’s projecting value onto things, and thus, as
Solomon said, they constitute one’s world. Emotions, Solomon argued, because
they entail such acts of projection, are performances, things that we do, and
not conditions, things that happen to us. Accordingly, being things that we do
rather than things that merely happen to us, emotions fall within the class of
things for which we are responsible.
In putting forward this last thesis Solomon located his subjective theory
within the thought of twentieth-century existentialism, especially that of
Jean-Paul Sartre. The dominant theme in Sartrean existentialism is the ines-
capable responsibility of every rational human being for how he or she lives.
As Solomon developed this theme, no one can escape responsibility for his or
her emotions by pleading helplessness with respect to them. For none of us
is helpless when it comes to determining what to value in life and what ide-
als to aspire to in living our life, and therefore how we understand objects,
people, and events as they affect our dignity and self-esteem is not beyond our
power to determine. Such an understanding is at work in the evaluative judg-
ments that our emotions consist in, and consequently those judgments are the
result of our having, either consciously or tacitly, determined to see the world
as imbued with the values from which they proceed. The common view of emo-
tions as alien forces that act on us and push us one way or pull us another, often
against our will and despite our efforts to combat them, is, Solomon declared,
a long-standing myth. A legacy of Plato’s philosophy, it had been elevated to
a finding of science through the prevailing theories of emotion in psychology
that Solomon was attacking. His opposition to these theories was no more
acute than in his thesis that we are responsible for our emotions.
Solomon’s book was the first in a wave of monographs and collections by
Anglo-American philosophers that presented theories of emotion explain-
ing the phenomenon as essentially cognitive. Within a year after it appeared,
Jerome Neu published Emotion, Thought, and Therapy, in which, drawing on
Spinoza’s notion of emotions as confused ideas, he explained emotions as
thought centered.14 Three years later Amélie Rorty brought out an anthology,
Explaining Emotions, of twenty-two essays by both younger and established phi-
losophers, psychologists, and social scientists that exhibited the wealth of new
ideas about the cognitive character of emotions that had emerged in the field.15
There followed in the next decade books by William Lyons, Patricia Greenspan,
Ronald De Sousa, and Robert Gordon, among others, each of which developed
a theory of emotions based on the idea that to understand emotions required
explaining the place of judgments, thoughts, beliefs, or cognitions of some

14
Jerome Neu, Emotions, Thought, and Therapy (Berkeley: University of California Press,
1977).
15
Amélie Rorty, ed., Explaining Emotions (Berkeley: University of California Press, 1980).
 Introduction 7

other type in their constitution.16 By the 1990s cognitive theories of this sort
represented the dominant program within Anglo-American philosophy in the
study of emotions. The general view of the nature of emotion that Solomon
had drawn from his polemic against the theories of emotion that prevailed in
experimental psychology at the midpoint of the twentieth century was by the
end of the century widely accepted.
Criticism of the specific theory Solomon had expounded, his subjective
theory, had nonetheless been sharp. It tended to center on two of the theory’s
main theses: first that the states of arousal that experimental psychologists
identified with emotions are inessential to emotion; and second that emotions
are among the class of things for which their subjects are responsible. Solomon
responded to these criticisms in papers that appeared during the three decades
following the publication of The Passions.17 His responses consisted neither
in an uncompromising defense of his theory nor in a complete makeover of
it. Rather he acknowledged the lack of nuance in his original statement and
sought to accommodate what he saw as valid in these criticisms by moderating
its theses.
The earliest accommodation resulted from the criticisms of social scientists,
particularly anthropologists, whose studies of the emotional life of people liv-
ing in vastly different cultures from our own implied a kind of provincialism
in Solomon’s identification of emotions with judgments concerned with one’s
dignity or self-esteem. In response, Solomon broadened his understanding of
the self-involvement that, in his view, characterized the evaluative judgments
in which emotions consisted. His original account, he conceded, had suffered
from his uncritical acceptance of egoism as a theory of human motivation.
Rejecting this theory, he characterized an emotion as an evaluative judgment
in which what is judged is personally important to one and observed that a
thing’s personal importance does not necessarily translate into how something
affects one’s self-esteem. One’s investment in it, in other words, is not always
conditioned on one’s sense of worth.
Having thus weakened his commitment to a Sartrean account of emotions,
Solomon entertained further qualifying the thesis that emotions are among
the things for which their subjects are responsible. Because of the thesis, the

16
See William Lyons, Emotion (Cambridge: Cambridge University Press, 1980); Ronald de
Sousa, The Rationality of Emotions (Cambridge, MA: MIT Press, 1987); Robert M. Gordon, The
Structure of Emotions: Investigations in Cognitive Philosophy (Cambridge, UK: Cambridge University
Press, 1987): and Patricia Greenspan, Emotions and Reasons: An Inquiry into Emotional Justification
(New York: Routledge, 1988). See also Gabriele Taylor, Pride, Guilt, and Shame: Emotions of
Self-Assessment (Oxford: Oxford University Press, 1985); and Helen Faye Nissenbaum, Emotions
and Focus (Stanford, CA: CSLI, 1985).
17
Solomon brought many of these together in a single volume, Not Passion’s Slave (Oxford:
Oxford University Press, 2003).
8 on emotions

theory implicitly applies only to human emotions. Beasts, after all, are not
responsible for their conduct. A fortiori they are not responsible for their emo-
tions. Nor did Solomon ever suggest that he agreed with the Stoics in denying
that beasts were capable of emotions. And once he recognized this restric-
tion on the emotions his theory covered, he allowed that the thesis needed
to be qualified to exclude its applying to the emotions of beasts, infants, and
other humans who lack responsibility for their actions. He then qualified it
further to exclude its applying to emotions, like being frightened by a car back-
firing, that result from the immediate triggering of some inborn disposition.
He thus came to allow that the range of phenomena that count as emotion is
significantly broader than what he originally had in mind when formulating
his theory. Indeed, he came to see the range as too broad and heterogeneous
to admit of a single framework that is best suited to the study of emotions.
Accordingly, he suggested that different frameworks fit different theoretical
interests. Consequently, he no longer flatly rejected as misconceived the theo-
ries of scientists and scholars who focused on emotions characterized by the
sudden irruptions of feeling and involuntary movement, for their theoretical
interests differed from his own. The latter was that of a moral philosopher, and
as such the emotions with which he was concerned were those that related to
questions of how one should live. It is with respect to his theory’s coverage of
them, he argued, that it should be assessed and against which the thesis about
our being responsible for our emotions should be tested.
One theory, in particular, that is better suited to a different framework from
that of moral philosophy convinced Solomon that he needed to be more recep-
tive of other theoretical pursuits in the study of emotions. This is Paul Ekman’s
Darwin-inspired theory. Like Darwin, Ekman studies the involuntary expres-
sion of emotions, specifically their involuntary facial expression. On the basis
of these studies, Ekman has concluded that certain broad types of emotion
are universal in humankind. Specifically, his studies have yielded evidence that
certain involuntary facial configurations are consistently identified across cul-
tures with the same distinct emotions, fear, sadness, anger, disgust, joy, inter
alia, and these are the universal types Ekman has distinguished. In addition,
Ekman associates each of these distinct facial configurations with a neurophys-
iological mechanism that produces it, and in view of these mechanisms’ being
universal in humankind he has taken them to represent the inherited emo-
tional capacities of human beings. They are, to use a phrase he borrowed from
Sylvan Tomkins, affect programs. On Ekman’s theory, then, each affect pro-
gram corresponds to a basic human emotion, and, given that these programs
collectively represent the inherited emotional capacities of human beings, every
human emotion is either a subtype of some basic emotion or a synthesis of two
or more basic emotions. Accordingly, Ekman has offered a conception of emo-
tions as affect programs. While Solomon rejected this conception of emotions,
 Introduction 9

he recognized that he could not ignore the evidence Ekman adduced to sup-
port it. He recognized, that is, that a theory of emotions on which emotions
are evaluative judgments had to account for the common recognition across
cultures of these facial configurations as expressions of emotion. He could not
treat it as mere coincidence.
To account for Ekman’s evidence, Solomon first noted that, on his theory, if
an emotion is universal, then the type of evaluative judgment it consists in is
universal. In other words, if it is universal, then people from vastly different
cultures make the same judgments. Second, he noted that, on his theory, for an
emotion to count as basic it must be such that the values it presupposes have an
important place in people’s lives. That is, on a theory of emotions on which they
are evaluative judgments, identifying an emotion as basic implies that the val-
ues it presupposes are themselves in some sense basic, and the relevant sense
is that they have an important place in people’s lives. It follows then, Solomon
observed, that an emotion’s being universal neither implies nor is implied by
its being basic in this sense. Hence, he could accept Ekman’s evidence without
accepting either Ekman’s typology or his identification of emotions with affect
programs as long as there were cultures in which the values that one or more
of the emotions Ekman identified as basic presupposed had no important place
in the lives of the people who belonged to those cultures. For in such cultures
the evaluative judgments in which these emotions consist would be periph-
eral or aberrant, and therefore to treat them as comparable to emotions that
presuppose values that have an important place in the lives of the people who
belong to those cultures would be, within the framework of the kind of theo-
retical pursuit Solomon was engaged in, to misconceive the emotional lives of
these people. That there were such cultures would allow Solomon to accommo-
date Ekman’s evidence by treating the affect programs Ekman identified with
emotions as the neurophysiological substrata of emotions whose character and
importance in people’s lives were determined by the values of the cultures to
which they belong.
Solomon’s chief example of such a culture was that of the Utku Inuit people
as described by the anthropologist Jean Briggs in her work Never in Anger.18
Briggs’s thesis was that the Utku were raised not to experience anger. While
they recognized anger in small children and foreigners, it had no place in their
lives. This is not to say that the Utku were unfamiliar with irritation or annoy-
ance or that they didn’t harbor hostile attitudes toward things they disliked.
Nor is it to say that they merely never showed anger owing to their having
learned to suppress it. Rather it is to say that anger and its variants, resent-
ment, indignation, outrage, fury, wrath, and so forth were simply not part of
the Utku’s standard emotional repertoire. Being absent from their emotional

18
Jean Briggs, Never in Anger (Cambridge, MA: Harvard University Press, 1975).
10 on emotions

repertoire meant, in Solomon’s view, that they did not take the actions of oth-
ers as insults, slights, or other offenses against them and likewise did not judge
others as having trespassed against them. They did not make evaluate judg-
ments about others or their actions of the type in which anger and its variants
consist. Nonetheless, Solomon observed, they had the same innate neurophysi-
ological mechanisms as other human beings, and therefore the affect program
that Ekman identified with anger was part of their biological inheritance. It
operated in very young children, whose display of anger was like that of the
very young in other cultures, but its manifestation in Utku adults was, on
Briggs’s description of their emotional lives, aberrant. Because the emotions
people experience are guides to how they judge themselves and the people and
things of the world they inhabit, it would misrepresent the Utku’s experiences
to include anger among their basic emotions.
Solomon’s mature theory of emotions thus remained at its core a subjec-
tive theory on which the phenomena were identified with evaluative judg-
ments. Phenomenology and conceptual analysis continued to be his preferred
methods of theorizing. At the same time, owing to the qualifications of the
theory’s more extreme elements that Solomon made in response to criticisms
of its initial statement, the theory in its final form no longer diametrically
opposed or directly challenged the objective theories of emotion that were the
products of experimental psychology. The polemical attacks on these objec-
tive theories that had accompanied the initial statement, his attack, in par-
ticular, on James’s understanding of emotion, evolved into more appreciative
critical treatments of these theories. He no longer saw his subjective theory
as a competitor to them. He no longer advanced it in the interest of defeat-
ing them. Rather he expounded his theory more in the spirit of John Stuart
Mill’s idea that truth lies in the synthesis of different perspectives, that no
one view on a subject captures the entire truth, and that his subjective theory
represented an important side of our understanding of emotions that scien-
tific theories tend to eclipse. Even James’s understanding of emotion found a
place in Solomon’s synthesis: it captured, Solomon said, the affective aspect of
those emotions that arise abruptly from the triggering of inborn dispositions
and whose physiological substrata are the affect programs that Ekman had
identified.
The change in Solomon’s view from staunch opposition to the theories
of emotion that dominated experimental psychology to open-minded con-
junction of these theories with his own raises new questions about his sub-
jective theory. Do these theories and his cover the same phenomena from
different perspectives, as Solomon sometimes suggests? Or do the emotions
that, for instance, Ekman identifies with affect programs comprehend emo-
tional states that are distinct from the ones Solomon identifies with evalu-
ative judgments? Does the anger the small Utku child displays consist in an
 Introduction 11

evaluative judgment that the child later learns not to make because he comes
to see such judgments as inappropriate? Or is the child’s anger and indeed
the anger of any small child regardless of the culture to which it belongs not
constituted by an evaluative judgment? These questions place the theory in
something of a bind.
Specifically, if the former is the case, if the child’s anger consists in an evalu-
ative judgment, then it is unclear why Solomon’s theory is in need of comple-
mentation by Ekman’s. It is unclear, that is, what is missing from Solomon’s
theory that Ekman’s supplies. If, on the other hand, the latter is the case, if the
child’s anger is not constituted by an evaluative judgment, then it is pertinent
to ask what the relation of this child’s anger is to adult anger when the latter
does consist in an evaluative judgment. A gap, that is, would appear to exist
between the account of anger due to Ekman’s theory and the account due to
Solomon’s since nothing in either theory explains why the small child’s emo-
tion and the emotion that consists in a judgment that another, the object of
one’s emotion, has committed an inexcusable offense against one are both
instances of the same emotion.19 On neither theory are the identity conditions
of anger sufficient to apply the theory to what the other theory understands as
anger. To achieve the full synthesis of these theories that Solomon envisioned
requires, then, either explaining how the one theory complements the other or
fi lling this gap. It is, one might say, the unfinished business of the program for
the philosophical study of emotions that Solomon, through the evolution of his
thought, came to pursue.
Solomon, in his last book on the subject, True to Our Feelings, offered a poign-
ant summary of this program. Remarking on the course of the book’s progress,
he wrote:

. . . I have tried to focus our attention on the emotions from an eth-

ical point of view, that is, how our emotions fit into and function in
our lives, their meaning. Accordingly, I have not paid as much atten-
tion to the new neurological and biological discoveries about emo-
tions as many of my colleagues would like, but I hope that I have also
made clear that I do not think that these perspectives are necessarily
at odds or in competition. They complement one another, face differ-
ent kinds of questions, but they also inform one another. So, too, the
social sciences appear to have a very different agenda than ethics and
the humanities, but that opposition is illusory, the product of an over-
emphasis on the science of these disciplines and a sometimes studied
neglect of their ethical origins.

19
I have discussed this gap and offered a way to close it in my “Emotions and Values.” See John
Deigh, Emotions, Values, and the Law (Oxford: Oxford University Press, 2008), pp. 72–102.
12 on emotions

After all, the question of how our emotions fit into and function
in our lives needs to be understood in a number of different but obvi-
ously related ways. Looking for the meaning of emotions is one way.
Studying their neurological substratum is another. Speculating on
their evolutionary histories is yet another. And then there is all of
that rich literature in the social sciences. Finally, there is the study
of the emotions in literature, the arts, and in music. A philosophical
perspective, with its emphasis on ethics and self-knowledge, is in no
way opposed to these but can benefit from them in all sorts of ways.
Because ethics, in one sense, is a thoroughly natural discipline and
not at all dependent on the commands of a supernatural God or some
mysterious notion of conscience. It has to do with our natural endow-
ments and propensities, our lives in society with other people, our per-
sonalities and character, our personal itineraries and choices in life.
Everything we can find out about human life, whether from science or
through literature and the arts, is important to us.20

*********

The essays collected here uphold this vision of what philosophy can contribute
to the study of emotions. Several of them, those of Robert Roberts, Laurence
Thomas, Kathleen Higgins, and Jerry Neu, speak directly to the questions of
ethics and self-knowledge that Solomon regarded as the soul of the philosophi-
cal study of emotions. Others, Martha Nussbaum’s, Paul Woodruff ’s, and Noël
Carroll’s, are studies of emotions in literature, arts, and music, specifically,
opera, theater, and comedy. Following Carroll’s study of comic amusement,
Michael Stocker’s essay argues for the importance of intellectual emotions
generally to the study of emotions. Stocker’s point, in keeping with themes
close to Solomon’s work, is to counter the tendency of theorists of emotion to
focus narrowly on the most primitive and explosive emotions in human life,
such as sudden fear, anger, and sorrow, and to treat them as paradigms of the
subject. Nancy Sherman and Kathleen Higgins, in their essays, draw on stud-
ies in psychiatry and clinical psychology to explore certain profound, human
emotions brought on by death, the guilt of soldiers for deaths of comrades or
innocents they accidentally caused, and the grief of spouses over the loss of
their mates. Charles Nussbaum’s essay, drawing on a different branch of sci-
ence, exemplifies the use of influential work in neuroscience on emotions to
enlarge our understanding of the perplexing philosophical problem of personal
identity. The collection’s final essay is my own. It is a study of the emergence

20
Robert C. Solomon, True to Our Feelings: What Our Emotions Are Really Telling Us (New York:
Oxford University Press, 2007), pp. 203–204.
 Introduction 13

in theoretical writing on emotion of the notion that late nineteenth and early
twentieth century experimental psychology embraced and that was the object
of Solomon’s polemical attack in The Passions.
The authors of these essays were among Solomon’s many friends, colleagues,
and admirers. Some of us were closer to him than others, but we all felt his
influence. Each of us learned from his writings and benefited from his wisdom.
These essays are a tribute to his pioneering and subsequently sustained philo-
sophical work on emotions that helped us come to a greater, richer understand-
ing of their place in human life.
 1

Justice as an Emotion Disposition*

Robert C. Roberts

Introduction
Like much of his work, Robert Solomon’s writings on justice are brash, polemi-
cal, fresh, passionate, insightful, inspiring, and provocative. I present here an
account of justice as a personal excellence—what we have in mind when we say
someone has a strong sense of justice, or a passion for justice, or is a just person.
My account has much in common with the one Solomon presents in A Passion
for Justice and some of his papers on justice. I intend this paper as a salute to
him and his life’s work.
From his earliest writings to some of his latest, Bob Solomon championed
the idea that emotions are a kind of evaluative judgment (Solomon, 1973, 1976,
2003). In fact, he was probably the philosopher most responsible for the grow-
ing currency, in the 1970s and 1980s, of “cognitive theories” of emotions. His
motive for thinking of the emotions in this way, rather than as, say, neurologi-
cal events, adaptive (or maladaptive) states of action-readiness, or perceptions
of body-states, was his desire to incorporate discourse about the emotions
into ethics, broadly speaking. He wanted to stress how our emotions struc-
ture our “world,” especially our social world, how they give us ourselves in rela-
tion to our fellow human beings and other significant aspects of our world,
how they shape our consciousness of things and so predispose our judgments
and actions. Of course he never denied that emotions have a neurology or that
they often involve bodily sensations and readiness to act (or at least behave)
in characteristic ways, and in some of his later writings he even stressed the
bodily character of the judgments that he identified as the passions (Solomon,
2004). But these alternative angles on emotions were less to his purpose as
a philosopher than the powerful ethical idea that emotions are constitutive
evaluating (and thus motivating) judgments. I think he was right about this,

*
Reprinted from Emotion Review

14
 Justice as an Emotion Disposition 15

and the present paper is a defense of the idea that something like a judgment
account of emotions opens special and important channels of insight into the
virtues and is thus an indispensable aid for the moral psychologist.
Within a large area of agreement about the ethical nature and importance of
emotions, Solomon and I sparred over the years on whether “judgment” is the
best choice of concept in this connection. Judgment suggests assent to what
our emotion is “telling” us, and it seems clear that we do not always agree with
what our emotions tell us about the world. If fear, for example, is my judgment
that I am facing a significant threat, then I must be disposed to say “yes” if
someone asks me, while in that state, whether I am facing a significant threat.
But if I have a phobia of riding in airplanes, I may admit (even while having a
mild panic attack in an airliner over the Atlantic) that I am a lot safer 30,000
feet above the earth than my fear is telling me I am. In my view, this is the main
objection to the claim that emotions are judgments. They are broadly “cogni-
tive,” and no doubt they predispose judgments (just as sense perceptions do),
but they neither are nor presuppose the corresponding judgment.
However, even if my fear is not a judgment that I am facing a significant
threat, it is a kind of perception or impression to the same effect. I “see” the situa-
tion as involving a significant threat, even if I don’t think it really involves one.
This is how it strikes me. It’s a bit like an optical illusion that I know to be such:
my eyes tell me something I don’t believe. In the emotion case, my emotion
tells me something I don’t believe. Of course I usually believe my emotions, as
I usually believe my eyes; but the fact that emotion and judgment come apart
in cases like that of the knowing phobic suggests that it is better to think of an
emotion as a kind of impression of my situation in certain terms than to think
of it as a full-fledged judgment.
So I have suggested that emotions are a kind of perceptual construal in vari-
ous kinds of “terms” such as threat, offense, good prospects, bad prospects,
blessings, honor, and dishonor. These are the kinds of things that matter to
people; they are the terms in which people typically care about situations. The
fact that some of our perceptions are based on our concerns or cares about
things would explain why such perceptions are evaluative. Thus, I have sug-
gested that emotions might be well thought of as “concern-based construals”
(Roberts, 1983, 1984, 1988, 2003, 2007b).
In personal correspondence during the 1980s, Solomon wrote to me that
what I call a construal is just what he means by judgment, and when he vetted
my review of Not Passion’s Slave for Mind (Roberts, 2004) he did not respond to
a repetition of my critique of the judgment theory. So it is a little puzzling why
he continued to think judgment the best cognitive category for understanding
emotions. Strictly speaking, he was not a judgment theorist. In this regard,
he differs from Martha Nussbaum, who explicitly raises the issue of whether
emotions involve assent to their constitutive impressions, and answers, with
16 on emotions

the Stoics, in the affirmative (see Nussbaum 1994, especially chapters 9 and
10; Roberts 1999 for critique; Nussbaum 1999 for response to my critique;
however, at a recent conference on Solomon’s work, John Deigh referred to
Nussbaum as a leading “perception theorist” of the emotions, and she did not
correct him). In a moment I will try to show how the notion of a concern-based
construal might support and refine Solomon’s account of the virtue of justice.

General Justice and Particular Justice

Let us begin with a distinction that Aristotle was the first, as far as I know, to
make clearly. Plato’s Republic is about justice, and especially about how justice
has value. Is justice desirable “for its own sake,” or does its value derive from its
consequences? Is it worthwhile for a person to be just if he doesn’t get anything
from it (say, a good reputation, honor, money, or avoidance of punishment)? One
of the definitions offered near the beginning of Republic is that justice (dikaio-
sunê) is telling the truth and paying what one owes (Plato, 2003, 331c, p. 19),
so it seems that the discussion will be about justice as people getting what is due
them. Something in this neighborhood is our everyday concept: justice is nei-
ther acquitting the guilty nor condemning the innocent; it is paying a fair wage,
respecting people’s rights, keeping one’s promises, paying one’s debts, and the
like. But somewhere early in Republic the concept of dikaiosunê changes into that
of the proper functioning and interrelation of parts of something—in particular,
a city or a psyche—so dikaiosunê becomes a state of general well-being. And of
course if that is what justice is, it is obviously desirable to the one who exempli-
fies it; any rational state or person who understands that justice is its own consti-
tutional well-being cannot help but desire it—for its own sake.
So we find at least two concepts of justice in the Republic; however, Plato
does not distinguish them clearly, and consequently his argument is murky or
invalid. Aristotle helps the discussion by distinguishing general justice (dikaio-
sunê), which is people acting generally virtuously toward one another and thus
producing a healthy social order, and particular justice (dikaiosunê), which is a
matter of fair distributions of goods, fair exchanges, and the fair meting out of
punishments for criminals and honors for the worthy (Aristotle, 1934, V.1–2,
pp. 253–67). What Aristotle calls general justice and what Plato mostly dis-
cusses in Republic is broader than what modern speakers of English usually call
justice; it is instead a very general concept of moral rightness, what we might
call, speaking a little quaintly, “righteousness.” This is the standard translation
of dikaiosunê in the New Testament, for example, where the word does not usu-
ally mean justice in the narrow sense.
Like Plato, in A Passion for Justice Solomon sometimes blurs the distinction
between general and particular justice. He says, “Compassion, this ability to feel
 Justice as an Emotion Disposition 17

for those less fortunate than oneself, is the cornerstone passion of our sense
of justice” (Solomon, 1995a, p. 225–26). But compassion, as I will argue here,
has a different conceptual structure from the emotions expressive of the virtue
of justice. Compassion is certainly included in general justice (righteousness),
yet it is clear from elsewhere in the book that Solomon intends to be writing
about justice in our ordinary (stricter) sense of the word. This is evident, too, in
a later comment on the book: “I would admit that we should resist the tempta-
tion to talk about the virtue of justice in terms of feelings, even ‘compassion’
and ‘benevolence’” (Solomon, 2001, p. 175); and in an affi xed footnote he says,
“This is something of a retreat from my argument in A Passion for Justice, which
is mainly concerned with the role of feelings in justice rather than with justice
as a virtue” (ibid., p. 185). Oddly, he seems here to distinguish justice as a dispo-
sition to various emotions from justice as a virtue. His basic view (which I think
is right) is that justice as a virtue is (among other things) a disposition to various
emotions. But I think he has good reason in the 2001 publication to deny that
compassion is one of the characteristic emotions to which the passion for justice
is a disposition. Toward the end of this paper I will try to explain why.
In the sense of general justice or righteousness, justice encompasses the sev-
eral virtues of compassion, generosity, forgiveness, mercy, and perhaps (depend-
ing on one’s outlook) gratitude and humility. All of these traits are morally
“right” in a broad sense; they are righteous (dikaios), to use the language of the
New Testament. But in a narrower sense of justice, these virtues differ notably
from justice and may even involve a mitigation or “violation” of strict justice. For
example, generosity is giving someone more of good than he is due; forgiveness
and mercy involve giving a person less than his due of evil—in each case, more
or less than is strictly “just.” Full human well-being would not be achieved by
having a society of people who were just only in the strict and narrow sense, nor
would an individual flourish if he did not have the virtues that go beyond strict
justice. The question of this paper is about the relation of emotions to the virtue
of justice in the strict or narrow sense.

Objective Justice and Justice as a Virtue

Solomon objects to the idea of theories of justice, seemingly on two kinds
of grounds: (1) theories of justice are always distorting oversimplifications
(Solomon 1995a, pp. 153–197); and (2) theories of justice neglect or deny or
forestall the possibility of justice as a personal virtue (Solomon 1995a, pp. 17,
33). I am sympathetic to the first complaint but will have little to say about
it here, except that it perhaps amounts to a plea for more attention to what
Aristotle calls “equity” (epieikeia), the personal ability to make refined judg-
ments in matters of justice to which formulated rules do not provide a
18 on emotions

defi nitive conclusion. As to the second complaint, it is true that recent philo-
sophical discussions have been strongly weighted in favor of “objective” jus-
tice—that is, the justice of states of affairs, policies, laws, and institutions—in
contrast with justice as a trait of personal character.

I want to shift the focus from impersonal institutions and govern-

ment policies to individual people and their personal passions. Justice
is not, as the tradition presumes rather than argues, an ideal to which
our society (or any other) more or less badly conforms. Justice is first
of all a function of personal character, a matter of ordinary, everyday
feeling rather than grand theory. (Solomon, 1995a, p. 3)

I don’t think it will do to throw out all conceptions of objective justice or to

think that philosophical efforts to conceptualize it necessarily lead to denying
justice as a personal virtue. Aristotle thought both conceptions necessary for
ethics, and at the beginning of his Digest Justinian quotes a third-century jurist,
Ulpian, who defines justice (iustitia) as a steady and enduring will to give to each
person what is due him (suum ius cuique tribuere) (discussed in Wolterstorff,
2008, pp. 22–24). What is due to each person is objective justice, and the steady
and enduring will to give each his due is the virtue of justice. Solomon accuses
Bernard Williams of thinking that “the particular notion of justice as a virtue is
a kind of category mistake, especially in Aristotle . . . ” (Solomon, 1995a, p. 301).
But the article Solomon cites (Williams, 1980) is about the personal virtue of
justice. Williams is in the virtue tradition of Aristotle and Ulpian when he says:

The notion of a fair distribution is prior to that of a fair or just person.

Such a person is one who is disposed to promote just distributions,
look for them, stand by them, and so on, because that is what they are.
He may also be good at inventing just distributions, by thinking of a
good method or proposing an acceptable distribution in a particular
case: this will be a characteristic of Aristotle’s epieik ês, the person who
is good at particular discriminations of fairness. But even there, it is
important that, although it took him, or someone like him, to think
of it, the distribution can then be recognized as fair independently of
that person’s character. (Williams, 1980, pp. 196–197)

Disagreeing with Solomon on the priority issue, Williams is pointing out that
a fair distribution is possible independently of any fair person (a distribution
might be fair by serendipity), but the very idea of a (distributively) fair person
is that of one who aims habitually at fair distributions, that is, one whose will is
steadily interested in bringing about objectively just situations when he can.
 Justice as an Emotion Disposition 19

Emotions, Passions, and Justice

Solomon’s most usual way of describing the relation between justice and par-
ticular emotion types is:

Justice . . . consists first of all of a constellation of feelings . . . . There

are several quite different and often confl icting feelings involved in
our sense of justice—a sense of pity and compassion for those in dire
need, a sense of righteous indignation when one’s hard-earned prop-
erty is taken away, just to name two. . . . The emotions are . . . the very
substance of our sense of justice. (Solomon, 1995a, p. 30)

Solomon writes of “the emotions that make up [the] sense of justice” (ibid.,
p. 198), yet he also clearly wants to allow that the virtue of justice is a character
trait. For example, he says, “The sense of justice (and injustice) is not an ‘epi-
sode.’ It is often, for better or for worse, a way of life” (Solomon, 1994, p. 294).
The virtue of justice is not an episode, either of action or emotion. Actions and
emotions occur at given times and often take a bit of time to be accomplished
or play out, but the virtue of justice does not occur and cease, then occur again
and cease. It is a disposition, a steady, ongoing trait of the person. Nor is it quite
right to think of the virtue merely as a class or grouping of actions or emotions;
the virtue is, rather, the psychological rationale for grouping a set of rather
diverse actions and emotions together as expressions of the virtue of justice.
It is “a systematic totality of emotions” (ibid., p. 296, italics added), the psycho-
logical explanation of the actions and emotions that express the virtue.
I propose that a natural way to mark the difference between the episodes of
action and emotion to which the virtue of justice gives rise, and the virtue itself,
is with the words passion and emotion. The title of Solomon’s book about the vir-
tue of justice is exactly right: A Passion for Justice. I take it that this title suggests
(or should suggest) that justice as a virtue is a passion (a steady dispositional
concern, desire, will, interest, or caring) about justice (that is, objective justice,
ultimately the state of affairs of any and all people getting what is due them).
In Greek and Latin, the mental states that we call emotions were called pas-
sions ( pathê, passiones) and so also in an earlier English: for example, Book II of
David Hume’s Treatise of Human Nature is titled “Of the Passions,” and it’s about
emotions such as pride and shame. But in today’s English, we rarely call an emo-
tion a passion, and when we do we usually mean intense anger: “the fans flew into
a passion over the referee’s call.” We also call the strong emotions consequent on
a romantic attachment (such as joy, despair, hope, disappointment) passionate,
but we are less inclined to call the individual emotions passions. We do, however,
speak of people having a passion for fashion, a passion for fitness, a passion for
20 on emotions

French cooking, a passion for movies, a passion for chocolate—or more seriously,
a passion for justice. A passion in this sense is not an emotion but a concern: a
caring about something, a desire for or attachment to something or some class
of things, an interest in something. But isn’t a concern just a kind of emotion?
Let me offer two arguments: the first to the conclusion that passions, as a kind
of concern, are not emotions; and the second to the conclusion that passions, as
something distinct from emotions, exist.
Argument that Passions Are Not Emotions. Assume that standard emotion types
have special names such as anger, joy, disappointment, hope, relief, shame, anx-
iety, pride. Each of these emotions has a distinguishing conceptual structure,
which writers about emotions analyze (for many examples, see Solomon, 1976,
chapter 8; Roberts, 2003, chapter 3). When psychologists argue about whether
there are basic emotions and if so which ones they are, the candidates are fear,
anger, joy, and the like. So emotions come in types. Now note that emotions of
any or all of these types can be based in any one of the passions mentioned in
the previous paragraph. A person with a passion for justice, for example, may as
a consequence experience fear, joy, hope, gratitude, anger. If he sees justice as
in danger of being undone or violated, his emotion may be anxiety or fear. If he
sees an injustice as being dramatically rectified, his emotion may be joy. If he
sees good prospects that an injustice will be rectified, he will feel hopeful, and if
efforts fail he will be disappointed. He will feel grateful to those who help in the
cause of justice and angry at those who intentionally obstruct it. The passion for
justice, like other passions, yields emotions of the various types. It is a possible
basis for emotions of many types, and, when it is the basis for some emotion, the
emotion expresses the passion. But the passion for justice is not an emotion.
Argument that Passions Exist. “But,” someone may retort, “why suppose that
passions, as you use the term, exist in addition to emotions? Let us admit that
the passion for justice is not identical with any particular type of emotion. Why
not suppose that ‘the passion for justice’ is just a collective name for any emo-
tion that is about justice?” Passions, in that case, would not be some additional
psychological reality that gives rise to emotions but a fiction used to organize
our thought about emotions.
That passions exist is suggested by the fact that emotions of a diversity of
types can be systematically explained by reference to them. We can infer from
one emotion type to another, contingent on characteristic variations in
the circumstances. Consider the following example. A subject gets angry on
the occasion of an injustice, and our “hypothesis” is that he is angry about the
injustice, that is, that the explanation of his anger is that he cares about justice.
We “verify” the hypothesis by varying the circumstances. If the hypothesized
object of concern is justice, then we will expect a subject who is tracking the
circumstances accurately to vary emotionally with the following variations in
 Justice as an Emotion Disposition 21

the circumstances of the object: if the prospects of justice have recently taken
a turn for the worse, anxiety or fear; if the prospects of justice have recently
taken a turn for the better, hope; if the injustice has a salient culpable agen-
tial origin, anger, or indignation; if the injustice has been saliently rectified by
the free action of an intervening agent, gratitude, or admiration; if the subject
himself has been responsible for an injustice, guilt; and so forth. If we observe
such changes in the circumstances and suppose that the subject is aware of the
change yet the subject does not experience an emotion of the predicted type,
this will count as evidence against the hypothesis of his passion for justice.
For example, consider again the person who becomes angry on the occasion
of an injustice. The injustice involves a harm to himself or someone he is espe-
cially attached to (say, a loss of money). Now we might wonder, what does the
subject really care about here? Does he care more about justice or about losing
the money? So we rearrange the situation so that the victim gets his money
back, but in such a way that the injustice is transferred to someone else, and
he knows it. If, knowing that the injustice has been transferred to another, the
person’s anger turns to relief or joy when the money is returned, we have rea-
son to think that his concern (passion) is less for justice than for money.
All of these variations depend on the “tracking” in question, which is to say
on the subject’s actually construing the situation in terms of the mentioned cir-
cumstances. The fact that we have this system of anticipations or predictions
bears out the hypothesis that the emotions are systematically ordered by the
concern. If we suppose that the so-called passion is nothing but the disposition
to this range of emotions, it remains mysterious how this range of type-diverse
emotion episodes comes to be just the disposition that I am calling the passion.
Psychologists who think of the emotions of the different types as belonging
to separate “affect programs” or neurological configurations or body-sensation
configurations tend to sideline this commonsense view of the systematic inter-
connections among the types.
In the foregoing arguments I have written of passion and concern without
distinguishing them, but not all concerns are passions. Some concerns are
temporary, passing, whimsical, or shallow, and these are not passions in the
sense I am exploiting. I am using passion in the way I think Solomon intends
it in his title A Passion for Justice. Here the concern for justice is deeply fi xed
in the personality. It is the kind of concern for justice that we see in William
Wilberforce or Nelson Mandela. It is a concern that would pass the variation
tests that sketched above a moment ago not just in a short time span or a local
environment but across the years.
The distinction between emotions and passions can be of great service to the
ethicist who wants to understand the emotional structure of certain virtues,
because it enables one to identify the basic nature of a virtue like justice and
22 on emotions

at the same time to sketch the relation between the virtue and the emotions
that it generates, the emotions that “exemplify” it or indicate its presence and
display the psychological core of the virtue, and to distinguish such emotions
from others that bear some other relation to the virtue. A virtue like justice can
thus be identified with a passion for (objective) justice. At the same time, this
psychological core of the virtue explains the actions characteristic of the just
person: voting, protesting, writing letters, speaking to officials, speaking and
writing to the public, risking harm to oneself, bringing legal action, contribut-
ing wealth and time. Such actions are often motivated by the anger, hope, fear,
gratitude, admiration, and sadness expressive of justice as a personal trait.

Four Kinds of Emotional “Object”

The foregoing discussion suggests a fourfold distinction among kinds of emo-
tional objects that can be helpful in a moral psychology of the virtues. Such a
moral psychology aspires to understand the relations between emotion episodes
and the virtues that dispose us to them, the moral value of emotion episodes,
and the relations among different virtues. For example, the schema might help
us understand the relation between compassion as a virtue and justice as a vir-
tue, as well as between each of these and the emotions generated on the basis
of them. It might help us to understand when an emotion of a given type is an
expression of the virtue in question and when it is not. Here is the schema:

1) The object of concern. All normal human beings are concerned about some
things—our retirement account, our job, our hobbies, the people we care about,
and what others think of us, for example. Some people are concerned about
justice, and some are concerned about the amount and kind of suffering in the
world. As I pointed out already, each of such concerns can generate a whole range
of emotions of different types, depending on how we perceive the situation of
the things we care about. The object of a concern is crucial to emotions, but the
object is not the object of an emotion. For example, someone who is angered in
such a way that his anger expresses his concern for justice is not angry about
justice (justice being each person getting what is due him); he is angry about
this case of injustice, and more particularly he is angry at the perpetrator.
2) The formal “object” of an emotion. We have noted that emotions come in
types, and that we have names for some of these types: anger, fear, jealousy,
etc. The emotion types fit situations of different types: anger posits some kind
of culpable offense, fear some kind of threat, jealousy some kind of actual or
threatened rival for another’s love or devotion, etc. The situation type that fits
an emotion of a given type can be called the formal object of the emotion. It is
the kind of thing that anger, fear, jealousy, or whatever is about. Each distinct
 Justice as an Emotion Disposition 23

emotion type has its own formal object. My way of specifying the formal object
of an emotion (type) is to formulate what I call the emotion’s defi ning proposi-
tion (see Roberts 2003, chapter 2, section 3).
3) The intentional object of an emotion. When a subject perceives something
that he or she is concerned about as involved in a situation of one of these types,
the subject is in the state of an emotion of the type that corresponds to the situ-
ation of that type. Thus, the intentional object of an emotion is the particular
situation as the subject of the emotion construes it. To communicate this to some-
one, it is very typical to tell a story in which figure particular things and people
that the subject cares about and the standard-type relations among these, as
they are represented in the mind of the subject. The explanatory story will have
elements belonging to the formal object of the explained emotion. This passion-
ate or concern-based perception, which Solomon (2004) calls an engaged judg-
ment, is the emotion itself on the kind of view that Solomon and I promote.
4) The actual object of an emotion. This is the object “in the world” that the
emotion’s intentional object needs to match if the emotion is to get the situa-
tion right. The intentional object may or may not match or fit the actual object
(Solomon 1995a, p. 53). For example, someone with a passion for justice might
be indignant on hearing the story of a racial injustice, but if the situation devi-
ated significantly from the narrative representation the indignation would not
fit the actual object. Here we see that the virtue of justice is more than a passion
for what the subject takes to be justice, since to be a genuinely just person one
must have good judgment about matters of justice; Solomon frequently stresses
that such judgment is not “dispassionate” (Solomon 1995a, pp. 165–166). If
one’s conception of objective justice is very poor, or one is very inept at recogniz-
ing justice and injustice in the particular situations of life, then one’s “passion”
is not really for justice and thus is not a passion for justice.

The Importance for Ethics of an Approach like Solomon’s

My thesis is that a “cognitive” approach to emotions like Solomon’s affords
resources for framing a moral psychology of the virtues by allowing us to see
how the “thought” that informs ethics is integrated into a person’s character
and experience. It clarifies how emotions with moral content can be expres-
sions of character traits and how virtues differ from one another. It allows us
to decide which emotions belong to a virtue and which do not and to distin-
guish the diverse ways that emotions can contribute to the virtues. It honors
the moral importance of discriminating carefully among the various possible
reasons a person might have for feeling angry, resentful, hopeful, joyful, or
disappointed. I want now to illustrate some of these points by considering how
the virtue of justice is related to compassion.
24 on emotions

I noted that in some of his earliest work on justice, Solomon (1995a, p. 226))
called compassion “the cornerstone passion of our sense of justice” but later
backed off from this claim. His basic orientation to the emotions and charac-
ter allows us to honor both of these intuitions. Objective justice is the state of
affairs in which people get what is due them, and the virtue of justice is a pas-
sion for justice, that is, a dispositional desire that people get what is due them.
The emotions that arise out of the passion for justice—anger toward those who
do injustices, guilt about one’s own injustices, frustration at obstacles thrown
up to the accomplishing of justice, joy at seeing justice done, and admiration of
those who act justly are all, in their different ways (formal objects), about people
getting what is due them. This is the theme that marks the emotions expressive
of the passion for justice, and such cognitive content is essential to these emo-
tions’ being expressions of justice.
Compassion, by contrast, is not about what is due somebody but about his
suffering. Its object of concern is different. Compassion as an emotion is a con-
strual of a sufferer as one who is suffering and that suffering as something lamen-
table, something that calls for remedy. This is the conceptual content of the
emotion of compassion, of its formal object. We can distinguish the emotion
of compassion from the passion (virtue) of compassion. Compassion as a vir-
tue is the sustained and deeply engrained dispositional desire that people (and
other beings that can suffer) not suffer. This is the object of compassion as a
passion. Notably, it is expressed in the emotion of compassion, but it can also
be expressed in many other emotions that are based in the desire that people
not suffer: joy when one’s efforts to relieve suffering succeed, frustration when
one’s efforts repeatedly fail, anger at those who culpably cause suffering, hope
in the good prospects of means of relieving suffering.
Injustices sometimes—often—cause suffering, and in such cases a person
who is both compassionate and just may well feel emotions expressing both
virtues—say, indignation at the perpetrator of the injustice and compassion
for the suffering victim. But situations calling for compassion are not always
unjust (we can feel compassion for someone whose suffering is due to his own
clumsiness, rather than any injustice), and unjust situations do not always
involve suffering (for example, being maligned in a way that has no conse-
quences for one’s wellbeing). And even when the injustice brings on some suf-
fering, the suffering may not be significant enough to justify compassion. One
is on a fishing trip with one’s buddies, poised to enjoy an evening of philoso-
phy and theology around the campfire and “a bunch of rugged outdoor types
who have nothing but scorn for city slickers like us camp next door and play
their rock-and-roll so loudly and spiel off their raunchy jokes so raucously that
our attempt at philosophico-theological discourse is completely frustrated”
(Wolterstorff, 2008, p. 247; the maligning example is also Wolterstorff ’s). The
city slickers have a right to be angry about the injustice, but it would be silly to
 Justice as an Emotion Disposition 25

feel compassion for them. So compassion is not one of the emotions logically
proper to the virtue of justice; it belongs to a different conceptual territory.
Yet the virtue of compassion coordinates, in the whole ethical personal-
ity, with the virtue of justice via a passion more basic than either of them—
benevolence, or wishing others well. One can see this point by trying to
imagine a person who has a passion for people getting what is due them,
but without caring about their well-being. Or to put the matter another way,
try to imagine a person who cares deeply that every person be given his due
but sees no value in many of the persons in question. He thinks many of the
persons whose rights he wants to be respected are worthless. Th is mind-set
seems to me incoherent. So valuing persons, wishing them well, or valuing
their lives as something of very high intrinsic worth lies behind or under
both compassion and justice as full-fledged virtues and connects them.
In most of us, I think, the virtue of justice, to the extent we have it, is some-
what abstract. It is a desire that people be given their due (at least some people),
yet without the full realization of who these people are, of their status as per-
sons just like ourselves. The virtue of justice seems better able to exist in this
state of abstraction than the virtue of compassion, and one might well think
that compassion is therefore a cornerstone, a corrective humanizing influ-
ence, on justice. In compassion we feel another’s humanity, and if this feeling
is strongly associated with justice, then we need compassion if we are to have
justice in the best possible way. As Solomon (1995a) remarks, “You cannot have
a sense of justice unless you care about the wellbeing of particular other people”
(p. 96, italics original).

Causal Connections between Compassion and Justice

In A Passion for Justice, Solomon shows considerable sympathy for the
eighteenth-century Scottish “sentimentalists,” philosophers like David Hume
and Adam Smith who attempted to ground ethical principles in very basic and
universal human emotional responses. But Solomon’s own use of the idea of
such emotions seems to be more modest and more plausible than that of his
philosophical forebears.
Several types of human emotions appear to have automatic or reflex-like
antecedent versions or counterparts, the explanation of which strains the kind
of account of emotions that Solomon and I commend as of primary use in moral
psychology. For example, if we suppose that fear, in adults, is a perception of
one’s situation as containing a significant threat to something one cares about
(thus a concern-based construal or engaged judgment), what do we make of
the very young child’s reflex-like fear response to an unprotected precipice?
Or if we think of compassion as the painful perception of a person whose
26 on emotions

well-being we care about as suffering some compromise of his well-being, what

do we make of the sympathetic crying of a newborn upon hearing another
baby cry? (For a discussion of empathic crying in babies, see Hoff man, 2008,
pp. 444–445.) In both cases there may seem to be neither an independent con-
cern nor the kind of propositional “take” on the situation that are the hallmarks
of the view we favor. I have laid out in some detail (Roberts, 2009a, 2009b)
how I think a view like Solomon’s and mine best understands the emotions of
beasts and babies. The gist of that account is that, from the most primitive to
the most sophisticated, emotions are perceptions that are more than sensory
takes on things—construals—that are evaluatively laden so that the unpro-
tected precipice looks scary and the cry of a human infant sounds distressed,
even to the least cognitively sophisticated human beings if they are normal. On
this biologically given base, proper human development builds the emotional
capacities of well-formed adults, among which are those we call the virtues.
The base is not enough to provide detailed norms for the virtues (norms that
are culturally and philosophically variable and endlessly contentious in their
details), but it provides a rough outline. The emotions of beasts and babies are
thus strongly analogous to the more sophisticated, language-based emotions
of human adulthood.
I think Wolterstorff (2008) misses the analogical relationship between sen-
timents and their more mature counterparts when he says, “I doubt that com-
passion is based on a concern; I think the Scots moral theorists were right in
appealing instead to a ‘sentiment’ to explain it” (p. 218, note 6). Compassion
in adults is seldom just a sentiment; it is a construal based on a concern for the
well-being of the sufferer. This seems evident from at least two considerations.
When efforts to relieve the sufferer succeed, the distress of compassion turns to
relief or joy, and the plausible explanation of the joy is that it is the satisfaction
of the concern for the sufferer’s well-being. (No doubt, in less mature people,
an important part of the explanation is that the agent has been relieved of the
distress of his own compassion—or “compassion,” as the case may be.) Also, in
most people, compassion is selective: we feel more compassion for our friends
and children when they suffer than we do for people we care less about.
Several types of emotions come in versions that antedate cultural learn-
ing. In addition to compassion and fear, we have sentiments that are obviously
antecedents of justice. When an older brother gets a privilege that his junior is
denied, the younger may resent it, feeling it’s not fair. He is appealing implic-
itly to a parity principle, yet the reaction is purely emotional. The feeling can
be an occasion for teaching a more complex conception of justice: difference in
age justifies difference in treatment. Similarly, the desire to give back harm for
harm (vengeance, a product of anger; Solomon 1995b) has sentiment status.
People don’t need to be taught it, and it is as obviously related to justice as
resentment is. Both sentiments are antecedents of a conception of what is due
 Justice as an Emotion Disposition 27

someone. Solomon’s (1999) point about sentiments like these is that they lie in
the psychohistory of the virtue, both in the individual and the human race. The
virtue would never have arisen among us (or would not have the character that
it does have), were we not the kind of beings for whom such responses are “nat-
ural.” To the emotional dispositions characteristic of the virtue of justice the
saying of Aristotle applies: “nature gives us the capacity to receive them, and
this capacity is brought to maturity by habit” (Aristotle, trans. 1934, 1103a25,
p. 71), to which we might add that our nature lays down their broad grammar,
but our moral traditions and considered beliefs about the world specify their
particularities.
Though compassion, in not being about what is due someone, has a different
“logic” from resentment and vengeance, the sentiment version of it lies like-
wise in the psychohistory of justice. It is a primitive expression or antecedent
of benevolence, and we have seen that benevolence lies in the background of
justice. The most mature desire that people be given their due depends on the
desire that they do well.
Compassion has other causal connections with justice. The view of a wrong-
doer’s suffering is often a trigger for forgiveness, and forgiveness, though it
goes beyond strict justice, often serves justice in the longer run by putting a
stop to cycles of vengeance. And being treated with kindness disposes people
to give others their due.

Conclusion
I hope to have made a little clearer the ethical-philosophical power of Robert
Solomon’s basic orientation toward human emotions by exploring, and perhaps
refining just a bit, some of his ideas about the virtue of justice.

References
Aristotle (1934). Nicomachean Ethics. Trans. H. Rackham. Cambridge, MA : Harvard University
Press.
Hoff man, M. (2008). Empathy and Prosocial Behavior. In M. Lewis, J. Haviland-Jones, and
L. Feldman Barrett (Eds.), Handbook of Emotions (3rd ed., pp. 440–455). New York:
Guilford Press.
Nussbaum, M. (1994). The Therapy of Desire. Princeton, NJ: Princeton University Press.
Nussbaum, M. (1999). Reply to Papers in Symposium on Nussbaum. Therapy of Desire in
Philosophy and Phenomenological Research 59, 811–819.
Plato (2003). Republic. Trans. P. Shorey. Cambridge, MA : Harvard University Press.
Roberts, R. (1983). Spirituality and Human Emotion. Grand Rapids, MI: Wm. B. Eerdmans
Publishing Company.
28 on emotions

Roberts, R. (1984). Solomon on the Control of Emotions. Philosophy and Phenomenological

Research 44, 395–403.
Roberts, R. (1988). What an Emotion Is: A Sketch. Philosophical Review 97, 183–209.
Roberts, R. (1999). Emotions as Judgments. Philosophy and Phenomenological Research 59,
793–798.
Roberts, R. (2003). Emotions: An Essay in Aid of Moral Psychology. Cambridge, UK : Cambridge
University Press.
Roberts, R. (2004). Review: Not Passion’s Slave. Mind 113, 588–590.
Roberts, R. (2007a). Compassion as an Emotion and as a Virtue. In I. Dalferth and A.
Hunziker (Eds.), Mitleid (Religion in Philosophy and Theology), vol. 28. Tübingen: Mohr
Siebeck, pp. 119–137.
Roberts, R. (2007b). Spiritual Emotions: A Psychology of Christian Virtues. Grand Rapids, MI:
Wm. B. Eerdmans Publishing Company.
Roberts, R. (2009a). Emotions and the Canons of Evaluation. In P. Goldie (Ed.), The Oxford
Handbook of the Philosophy of Emotion, pp. 561–583.
Roberts, R. (2009b). “The Sophistication of Non-Human Emotions” in Robert W. Lurz, ed.,
Philosophy of Animal Minds. Cambridge, UK : Cambridge University Press, pp. 218–237.
Solomon, R. C. (1973). Emotions and Choice. Review of Metaphysics 27, 20–41.
Solomon, R. C. (1976). The Passions. Garden City: Anchor Doubleday.
Solomon, R. C. (1994). Sympathy and Vengeance: The Role of the Emotions in Justice. In
S. van Goozen, N. de Poll, and J. Sergeant (Eds.), Emotions: Essays on Emotion Theory
(Festschrift for Nico Frijda). New York: Lawrence Erlbaum, pp. 291–311.
Solomon, R. C. (1995a). A Passion for Justice: Emotions and the Origin of the Social Contract.
Lanham, MD: Rowman and Littlefield.
Solomon, R. C. (1995b). Justice as Vengeance, Vengeance as Justice: A Partial Defense of
Polymarchus. In J. Sterba (Ed.), Morality and Social Justice. Lanham, MD: Rowman and
Littlefield, pp. 251–300.
Solomon, R. C. (1999). Vengeance: On Law as Expression of Emotion. In S. Bandes (Ed.), The
Passions of Law. New York: New York University Press, pp. 123–148.
Solomon, R. C. (2001). Justice as a Virtue. In J. Sterba (Ed.), Social and Political Philosophy:
Contemporary Perspectives. New York: Routledge, pp. 169–186.
Solomon, R. C. (2003). Not Passion’s Slave: Emotions and Choice (The Passionate Life). New York:
Oxford University Press.
Solomon, R. C. (2004). Emotions, Thoughts, and Feelings: Emotions as Engagements with
the World. In R. C. Solomon (Ed.), Thinking about Feeling: Contemporary Philosophers on
Emotions. New York: Oxford University Press, pp. 76–88.
Williams, B. (1980). Justice as a Virtue. In A. Rorty (Ed.), Essays on Aristotle’s Ethics. Berkeley:
University of California Press, pp. 189–199.
Wolterstorff, N. (2008). Justice: Rights and Wrongs. Princeton, NJ: Princeton University Press.
 2

Equality and Love at the End of

The Marriage of Figaro: Forging
Democratic Emotions
Martha C. Nussbaum

I have no idea what those two Italian ladies were singing about.
Truth is, I don’t want to know. Some things are better left
unsaid. I’d like to think they were singing about something
so beautiful, it can’t be expressed in words, and it makes your
heart ache because of it . . . , and for the briefest of moments,
every last man in Shawshank felt free.
Morgan Freeman (“Red”), The Shawshank Redemption,
of the “Canzonetta sull’aria,” from Act III,
The Marriage of Figaro

My great peace woman has only a single name: she is called

universal justice, humaneness, active reason . . . Her function, in
accordance with her name and her nature, is to inculcate dis-
positions of peace.
Johann Gottfried Herder, Letters for the
Advancement of Humanity (1793–1797)

“Happy in that Way”

The ancien régime sings in a loud and authoritarian voice, saying, “No, no, no,
no, no, no.” So, just before the end of Mozart’s opera, the Count, as yet secure in
his status, rejects the urgings of the other characters to mercy and sympathy, as
they kneel, one by one, before him. To Almaviva, revenge for insulted honor is
all-important (“the only thing that consoles my heart and makes me rejoice”).1

1
“Già la speranza sola/Delle vendette mie/Quest’anima consola,/E giubilar mì fa,” the end of
his third-act Aria. Th roughout I rely on the edition of the libretto in Wolfgang Amadeus Mozart:
Three Mozart Libretti (New York: Dover Publications, 1993).

29
30 on emotions

To display kindness to the imploring, as they humbly kneel, is a noble prerogative,

not a general human virtue. He can give it, or he can withhold it. If he chooses the
latter course, putting slighted honor ahead of generous condescension, nobody
can say he is wrong. That’s how the ancien régime operates, animated by a moral-
ity of status, shame, and kingly prerogative.
But suddenly the Countess, removing her disguise as Susanna, reveals her-
self—revealing, at the same time, the stratagem that has trapped her husband
in both error and hypocrisy. (Having boasted of ending the droit du seigneur,
he has all the while been scheming to enjoy it.) Everyone present exclaims in
hushed voices that they don’t know what is going to happen next: “Oh heav-
ens, what do I see! Madness! A hallucination! I don’t know what to believe!”
The strings, coursing up and down, with rapid modulations, express immense
agitation and uncertainty. In retrospect, it is the uncertainty of transition
between two political regimes.
And now the Count, kneeling before the Countess, sings—in a voice newly
softened by confusion—a phrase of a type, lyrical and legato, hushed, almost
gentle—that we have never heard from this man before: “Excuse me, Countess,
excuse me, excuse me.” There is a long pause.2
The Countess then sings softly out of the silence: “I am nicer, and I say yes”
( più docile io sono, e dico di sì).3 The musical phrase arcs gently upward and then
bends down as if, almost, to touch the kneeling husband. And now, in hushed
and solemn tones, the entire assembled company repeats the Countess’s phrase,
this time to the words, “Ah, all of us will be happy in that way” (tutti contenti
saremo così). The choral version of the phrase is reminiscent of the solemn sim-
plicity of a chorale (which, in this Catholic musical universe, denotes a sudden
absence of hierarchy).4 A hesitant orchestral interlude follows.

2
The length of the pause is interpreted variously by different conductors, but both Solti
and Karajan hold it for 4 seconds, which feels very long. In the score, the pause is designated
by a quarter-note rest with a fermata. (See Wolfgang Amadeus Mozart, The Marriage of Figaro
(Le Nozze di Figaro) in Full Score [New York: Dover, 1979], p. 422.)
3
Docile is difficult to translate: one could also say “gentler,” or “kinder.” I’ve chosen “nicer” in
order to convey the fact that this is a very everyday word, not an exalted moral or philosophical
one. It also connotes (to some extent in Da Ponte’s time, even more so today) tractability and
yielding, perhaps even submission. However, one cannot read even the libretto as saying that the
Countess simply acquiesces in her subordinate role: she says not, “I am docile,” but rather, “I am
more docile,” suggesting that being docile is a virtue that all should possess and that the Count
possesses to a deficient degree. I would think of it, then, as denoting yielding to life’s complexities
and imperfections, being pliant rather than rigid.
4
See also Joseph Kerman’s “half hymn-like” in Opera as Drama (Berkeley: University of
California Press, 1956), 87. For related observations about Mahler’s use of Bach in the Second
Symphony, see my Upheavals of Thought: The Intelligence of Emotions (New York: Cambridge
University Press, 2001), chapter 15. Michael P. Steinberg, in both Listening to Reason: Culture,
Subjectivity, and Nineteenth-Century Music (Princeton, NJ: Princeton University Press, 2004) and
Judaism: Musical and Unmusical (Chicago: University of Chicago Press, 2007), has drawn attention
 Equalit y and Love at the End of T he Mar riage of Figaro 31

The group now bursts out, a sudden eruption of dizzy elation:5 “This day of
torment, of craziness, of foolishness—only love can make it end in happiness
and joy.” Love, it seems, is the key not only to the personal happiness of the
central characters but also to the happiness of “all,” of the whole community, as
they sing: “Let us all rush off to celebrate” (corriam tutti a festeggiar).
The usual story about Mozart’s Le Nozze di Figaro (1786ƒ) is that it is a
cop-out. Taking the radical Beaumarchais drama of 1778, whose essential
point and emphasis is political, a denunciation of the ancien régime and the
hierarchies it imposes, Mozart and his librettist Lorenzo Da Ponte have fash-
ioned an innocuous drama of personal love, defanging the text by omitting, for
example, Figaro’s long fifth-act monologue denouncing feudal hierarchy and
substituting a more extensive treatment of women and their private desires.
The Beaumarchais play, which is usually understood to be a major harbinger
of the French Revolution, was refused production for many years, and even
in 1784, when it was allowed production in France, becoming wildly popular,
it remained controversial.6 Mozart and Da Ponte, by contrast, decided (so the
story goes) to escape controversy. The relatively progressive Joseph II had for-
bidden the Beaumarchais play to be performed in theaters within his realm.
Da Ponte, however, persuaded the Emperor that an acceptable opera could be
written on the basis of the play.7 In the process, however, says the received
story,8 he and Mozart, despite producing a wonderful love-drama, sold out the
radicalism of the original.
I shall argue, by contrast, that the opera is as political and as radical as the
play, and more deeply so: for it investigates the human sentiments that are
the necessary foundation for a public culture of liberty, equality, and frater-
nity. This construction of sentiment is accomplished more clearly in Mozart’s
music than in the libretto, so making my argument will require going into

to the many ways in which the period’s religious tensions are worked out in its musical culture, in
such a way that Protestant and Jew are frequently aligned together in a repudiation of a Catholic
culture of representation, “idolatry,” and hierarchy. Here, we needn’t think of J. S. Bach in par-
ticular, since his music was rediscovered only later; the allusion is to the general culture of the
Protestant chorale.
5
At this point the key changes from G major to D major, and the tempo is marked Allegro
assai.
6
For the Beaumarchais play, I use the edition by Malcolm Cook, Beaumarchais: Le Mariage de
Figaro (Bristol: Bristol Classical Press, 1992).
7
For an excellent treatment of the historical background of the opera, see Tim Carter,
W. A. Mozart, Le Nozze di Figaro (Cambridge Opera Handbooks), (New York: Cambridge University
Press, 1987), an excellent treatment of the opera covering a wide range of issues.
8
The received story has some foundation in Da Ponte’s Memoirs, which do at least tell us of
what he said to try to persuade Joseph II. Th at hardly shows that the libretto’s real intent was
apolitical, however, and even if Da Ponte’s intent had been utterly apolitical, that would hardly
show us that the music that animates the libretto is apolitical.
32 on emotions

considerable musical detail.9 I shall argue that Mozart agrees (in effect)10 with
Rousseau in understanding that a political culture requires a new shaping
of human attitudes in the realm of love but that he disagrees with Rousseau
about the specific shape of the new attitudes. Whereas Rousseau emphasizes
the need for civic homogeneity and solidarity, a patriotic love based on manly
honor and the willingness to die for the nation, Mozart envisages the new pub-
lic love as something gentler, more reciprocal, more feminine—“nicer,” to use
the Countess’s everyday word—connected more to Cherubino’s horror of war-
like exploits than to Rousseauesque ideas of valor. In the process, Mozart also
eschews Rousseauian homogeneity, emphasizing that the new fraternity must
protect spaces for the free play of mischief, craziness, humor, and individuality—
all of which are connected, in the opera, to the women’s world.
In arguing that the culture of the Enlightenment requires a new form of
subjectivity, one that the music of this period explores and in part creates, I am
agreeing with the central thesis of Michael P. Steinberg’s important Listening to
Reason and with the general direction of his interpretation of the opera. I shall
argue, however, that Steinberg’s more or less total neglect of the politics of gen-
der in Figaro means that he cannot offer us a deep or precise enough account of
what needs to change if the revolution is to be humanly possible. Symptomatic
of the difficulty in Steinberg’s reading (which also neglects the Countess and
her yes) is its total neglect of the male–female character Cherubino (a teenage
boy sung by an adult woman), but Cherubino, I shall argue, is in many ways the
opera’s pivotal character, a male who can be both delightful and loving, capable
of empathy and reciprocity, only because he has been educated by women and
by music and thus has grown to love jokes instead of put-downs singing rather
than marching, mischief rather than revenge.
Pursuing such connections between music and emotion, and between both
of these and political thought, seems an appropriate way to honor the career of
Robert Solomon, whose bold defense of a then unpopular view concerning the
cognitive content of emotions opened that field to all those of us who see emo-
tions as rich and multilayered rather than as simple pushes and pulls. Indeed,
Solomon’s work was a major part of what made an inquiry like this one—which
links emotions to political thought not only by seeing them as motivations for
action but also, and especially, by showing how they supply political thought
with part of its content—seem potentially fruitful rather than absurd. The
work I have already done on emotions in music, developed in chapters 5 and 15
of Upheavals of Thought (which provides a basis for my arguments here) could

9
See Kerman, 90–91, who emphasizes Mozart’s transformation of his material, especially at
the end of the opera.
10
I see no reason to suppose that Mozart read Rousseau, but these ideas about civic sentiment
were all around in the 1780s.
 Equalit y and Love at the End of T he Mar riage of Figaro 33

not have been begun without the stimulus of Bob’s thought and the courage
of his example. I hope this essay does at least a little justice to the wonderful
career so prematurely and tragically interrupted. I also hope to honor the life of
teamwork between Bob and Kathy, which so often focused on the intersection
of music and love.11
The project will also have the advantage of allowing me to engage in detail
with two recent works that I greatly admire: Charles Nussbaum’s The Musical
Representation,12 which has done so much to enhance our understanding of
musical listening as a mental phenomenon; and Michael Steinberg’s Listening
to Reason, which impressively pioneered the enterprise of locating politi-
cal thought in Enlightenment-era music. If I focus on areas of criticism and
disagreement, contending that Charles Nussbaum has imputed to music too
simple a set of human goals and purposes and that Steinberg has neglected the
role of gender in Mozart’s politics, my deep admiration for both books should
be presupposed.

The Ancien Régime and the Male Voice: Honor,

Shame, Disgust
According to the received view, Beaumarchais dramatizes the opposition
between an ancien régime, based on hierarchy and subordination (personified
by the Count) and a new democratic politics, based on equality and liberty (per-
sonified by Figaro). The key moment of the Beaumarchais play is thus Figaro’s
Act V monologue, in which he denounces the Count’s hereditary privilege.
Mozart, omitting this political speech, has depoliticized the opera, turning the
confl ict between the Count and Figaro into a merely personal competition over
a woman.
This view contains a tacit premise: that the contrast that should hold our
interest, as political thinkers, is the opposition between the Count and Figaro.
It is because Mozart does not locate the center of the political confl ict here that
his version is felt not to be political at all, but merely domestic. Let us, however,
keep an open mind. Let us not assume that the Figaro represents the new citi-
zenship (as the Count so clearly represents the old).
If we do keep an open mind, we are likely to notice very soon that Figaro
and the Count are very similar, both musically and thematically. What do they

11
I shall later discuss Kathleen M. Higgins, The Music of Our Lives (Philadelphia: Temple
University Press, 1991); for the collaboration, see especially Robert Solomon and Kathleen
Higgins, The Philosophy of (Erotic) Love (Lawrence: University of Kansas Press, 1991).
12
Charles O. Nussbaum, The Musical Representation: Meaning, Ontology, and Emotion
(Cambridge, MA: MIT Press, 2007).
34 on emotions

sing about when they are alone? Outraged honor, the desire for revenge, the
pleasure of domination. The energies that drive these two men are not alien but
deeply akin. (Indeed, the two roles are set in such a way that one and the same
singer might, in principle, sing either role, and their musical idioms are so alike
that it is easy to confuse them.)13 Figaro’s initial aria, Se vuol ballare, follows his
discovery that the Count has plans to sleep with Susanna. But if we simply look
at what Figaro says in the aria, we would not discover that any such creature as
Susanna ever existed. All his thoughts are about his rivalry with the Count, and
his insistent negatives (non sarà, non sarà) anticipate the Count’s peremptory
negations at the opera’s end (as well as those in the Count’s Act III aria). What
energizes Figaro? The thought of paying the Count back in kind, teaching him
to dance in Figaro’s dancing school.
Similarly, two acts later, the Count imagines Susanna, his own future prop-
erty, being possessed by Figaro (ei posseder dovrà), whom the Count sees as “a
base thing” (un vile oggetto), a mere object.14 This thought torments him—not
because he is fi lled with any love or even any particularly intense desire for
Susanna but because the idea of being bested by a mere “thing” is intolerable.
To this competitive loss he, like Figaro, has to say no: “Ah no, I am not willing
to allow you to enjoy this happiness in peace. Brazen one,15 you were not born
to give me torment, and perhaps even to laugh at my unhappiness.” Figaro, not
Susanna, is the person whom he addresses in the second person. Like Figaro,
his head is fi lled with the picture of another man, laughing at him, insulting his
honor, putting him to shame. In return for that tormenting picture, to which
he (like Figaro) says no, he proposes (like Figaro) to substitute the image of a
tamed enemy dancing to his tune, in this case the picture of Figaro forced to
marry Marcellina and separated forever from Susanna, whom the Count can
then enjoy: “Now only the hope of revenge consoles my heart and make me
rejoice.”16 Figaro’s Se vuol ballare is closely based upon the Beaumarchais text;
this aria of the Count’s, however, is a complete innovation of Da Ponte, since
Beaumarchais gives us only what forms the recitative before the aria, not the
aria’s extended development of sentiments of humiliation and reactive rage.
Musically as well as textually, the Count’s aria is a cousin of Figaro’s: full
of an ill-governed fury that bursts out as the voice reaches the words felice un

13
Bryn Terfel, for example, well known in performance for his Figaro, has also recorded the
Count.
14
The Countess understands that she too is a thing to him: later, when he addresses her as
“Rosina,” she replies, “I am no longer she, but the wretched oggetto of your abandonment.”
15
I translate audace in this awkward way because to supply “man” or “person” would constitute
an acknowledgment that Figaro is human, which is what the Count has just been denying.
16
Literally, the Count speaks of “revenges” in the plural—thinking, presumably, of the way in
which he will both force Figaro to marry Marcellina and then humiliate him further by sleeping,
himself, with Susanna.
 Equalit y and Love at the End of T he Mar riage of Figaro 35

servo mio and then again at ah non lasciarti in pace; anger, in the music, is com-
plemented by sneering irony (the downturning phrase accompanying un vile
oggetto). The libretto gives us some indication of the kinship between the two
men, but the expressive range of the music goes much further to emphasize
their rhythmic and accentual similarity, as both express attitudes that range
from snide contempt to furious rage.17 What emotions are absent? Love, won-
der, delight—even grief and longing.
According to the conventional political reading of Beaumarchais, Figaro
becomes, by Act V, the apostle of a new type of citizenship, free from hierar-
chy. Mozart’s Figaro makes no such progress. As Michael Steinberg aptly notes,
throughout the opera (or, at least, until late in Act IV), Figaro dances, musically,
to the Count’s tune: “[H]e hasn’t found a musical idiom of his own; his political
and emotional vocabulary suggests a similarly unfortunate mimetic duplica-
tion of the Count’s”18 —both in Non più andrai, at the end of Act I (where he
reenacts “the authority with which [the Count] has just dispatched Cherubino
to serve in one of his regiments, forming his phrases from the relevant military
march”)19 and even at the opening of Act IV, when, waiting to catch Susanna in
infidelity, he sings, once again, of slighted honor, asking all males to “open your
eyes” to the way women function as agents of humiliation. Again, it is men, not
women, far less a particular woman, whom he addresses in the second person.
Now maybe this means that Mozart has failed to understand the opposi-
tion between Figaro and the Count that Beaumarchais has depicted. But let’s
not pass judgment so quickly. Perhaps, instead, Mozart sees something that
Beaumarchais does not see: that the ancien régime has formed men in a cer-
tain way, making them utterly preoccupied with rank, status, and shame, and
that both high and low partake of this social shaping. What one does not wish
to lose, the other wishes to enjoy. For neither, given their obsession, does any
space open out in the world for reciprocity or, indeed, for love.
The suspicion that Mozart is deliberately subjecting the male morality of
status to critical scrutiny is confirmed when we see what Mozart puts in the
mouths of two males who have little to do with the plot. Perhaps, someone
might argue, the sentiments of Figaro and the Count are not to be read as seri-
ous political thought: after all, the plot requires them to compete in this way. We
have seen that even so, Da Ponte constructs a parallel that is not so plain in the
original text and that Mozart takes this similarity much further by giving the
two men a similar expressive musical range. Nonetheless, someone might still

17
For a general account of emotional expression in music, on which I rely here, see my Upheavals
of Thought chapter 5, whose main contentions are summarized in the Appendix; on the ability of
music to go beyond a text or to make more precise the emotional meaning of an indeterminate
text, see the reading of Mahler’s Kindertotenlieder in that chapter.
18
Steinberg, Listening, 43.
19
Ibid.
36 on emotions

argue that Mozart and Da Ponte are simply amplifying the suggestions of the
Beaumarchais plot. This, however, cannot be said of the treatment of Bartolo
and Basilio, whose role in the story line is minimal. Each sings an aria—Bartolo
in Act I, Basilio in Act IV (though typically cut in performance)—both of which
supply crucial commentary on the morality of maleness. Neither of these arias
is based on anything in Beaumarchais’s text.
Bartolo is an emotional first cousin of Figaro and the Count. Vocally distin-
guishable, since he is a basso, he nonetheless sings from the same expressive
palate: similar outbursts of rage, tempered by a type of sneering already known
to us from Figaro’s Se vuol ballare. Textually, his role appears to be to offer a gen-
eral theory of what Figaro and the Count both exemplify: “Revenge, oh revenge!
It is a pleasure reserved for the wise. To forget humiliations and outrages is base-
ness, is utter lowness.”20 So, life is more or less utterly occupied by competition
for status and the avoidance of shame between males, and the smart thing to
do is to play that game to the hilt. The recommended attitude not only causes
outrage and humiliation to eclipse love and longing (Bartolo, like Figaro and the
Count, has no thought at all for Rosina, whom he has lost to the Count, through
Figaro’s scheming) but also precludes any kind of mercy or reconciliation. It is
this attitude that leads to the Count’s six consecutive no’s at the opera’s end.
Bartolo also shows us something else pertinent to citizenship: for he is very
interested in reason and law. His attitude is that the law is an instrument of male
revenge, and someone who knows the law will be ahead of someone who does
not, because he can find the little weaknesses and loopholes that will allow him
to do down his enemy. At this point the aria becomes rapid, joyful, with a kind of
sneering playfulness, a patter song of legal one-upmanship: “If I have to search
through the whole legal code, if I have to read all the statutes, with an equivo-
cation, with a synonym, I will find some obstacle there. All Seville will know
Bartolo! The rascal Figaro will be defeated!”21 (Here the music, once again, goes
well beyond the text, expressing the sly joy of legal cleverness dragooned into the
service of humiliation.) The aria ends foursquare and martial as it began (though
with a little sneer accompanying the words il birbo Figaro). Bartolo announces
that “he” will be known to all: “All Seville will know Bartolo”—showing us that
he takes his practical identity to be utterly encompassed in his revenge project.
His joy at the thought is unqualified—despite the fact that the revenge in ques-
tion would never return Rosina to him. She is simply not on his mind at all.
At the opening of Act IV, another minor character has his say, and it
both inverts and ultimately reinforces the morality of Bartolo. Basilio, a

20
La vendetta, oh la vendetta è un piacer serbato ai saggi. L’obliar l’onte, gli oltraggi, è bassezza,
è ognor vilt à.
21
Se tutto il codice dovessi volgere, se tutto l’indice dovessi leggere, Con unequivoco, con un sinonimo,
qualche garbuglio si troverà . Tutta Siviglia conosce Bartolo: il birbo Figaro vinto sarà.
 Equalit y and Love at the End of T he Mar riage of Figaro 37

music-master, is a less powerful character than Bartolo, and it was he who, in

the predecessor play The Barber of Seville, discoursed enthusiastically on the
crushing humiliation that gossip and slander can offer someone who wants
to defeat an enemy. Da Ponte portrays him throughout as both spiteful and
weak, lacking the resources to compete on an equal footing with the nobles,
lacking the cleverness to compete on an equal footing with Figaro. His Act IV
aria offers advice to men who are in this weakened position.22 He begins by tell-
ing the audience that it is always risky to enter a competition with the grandi:
they almost always win. So, what should one do? A story from his youth offers
guidance. He used to be impulsive and didn’t listen to reason; then, however,
Lady Prudence made her appearance before him and handed him the skin of
an ass. He had no idea what it was for, but when, shortly after that, a thunder-
storm began, he covered himself in that ass’s skin. When the storm abated he
looked up—and found a terrible beast before him, almost touching him with
his mouth. He could never have defended himself from a terrible death, but the
disgusting smell of the ass’s skin scared the beast away. “Thus fate taught me
that shame, danger, disgrace, and death can be escaped under an ass’s skin.”23
This aria offers advice diametrically opposed to the advice in Bartolo’s, which
told us to use reason and law to hound the person who has caused one’s humilia-
tion. It’s obvious, however, that the difference is slight. Both men see the world
in the same way, as a zero-sum game for honor and status. The only differ-
ence is that Basilio is aware that some are bound to be losers, and he wants to
give advice about damage control to those losers. If you are perceived as smelly
and low anyway, use that spoiled identity to protect yourself from yet further
outrage. Sung in a sneering reedy tenor, the aria, like Bartolo’s, complements
Figaro and the Count from the other direction. It shares with them a view of
what the world is really about. If women figure in the aria at all, it is only in the
way Basilio confesses to a kind of “fire” and “craziness” in his youth—a false
direction soon put right by the counsels of Lady Prudence. The ancien régime
does not like low-class people who allow their fire free rein.

Females: Fraternity, Equality, Liberty

The females of the opera inhabit a musical and textual world that is from
the beginning depicted as utterly unlike that of the men. First of all, it con-
tains friendship. Susanna and the Countess might have seen one another as
rivals: after all, the Count is trying to seduce Susanna. However, the thought

22
Because this aria is so commonly cut it is not in the Dover edition of the libretto, so I use the
text from the libretto accompanying the 1983 Solti recording of the opera.
23
Cosi conoscere me fe’la sorte ch’onte, pericoli, vergogna e morte col cuoio d’asino fuggir si può.
38 on emotions

does not occur to them: they understand that they share a common set of
purposes and that the desired outcome for both is that both men, Figaro
and the Count, become loving and faithful husbands focused on affection
and pleasure rather than revenge and jealousy. (The Count is as dominated
by jealousy as is Figaro, despite his apparent lack of love for his wife.) Like
the two men, the two women share a musical idiom—so much so that they
can be mistaken for one another even by the men who ostensibly love them
(until, interestingly, Figaro does at least recognize Susanna by her voice,
“the voice that I love”).
Unlike the men, however, the women use their similarity not for mutual
combat but for cooperation and, in particular, for the complex masquerade
that ends up revealing the Count’s hypocrisy. When we focus on their team-
work, we notice, as well, that there is absolutely nothing like teamwork and
reciprocity among the men. The women’s partnership, moreover, despite their
class difference, appears to be quite unhierarchical, as they benefit one another
with genuine mutual friendliness. (Susanna, for example, is surprised that it
is she—presumably not all that well educated—who is to write the letter to
the Count suggesting the rendezvous: “I should write? But madame. . . . ” The
Countess will have nothing of her deference: “Write, I say, and I will take the
consequences.”) One way we see their reciprocity is in the nature of their jokes:
for here there are no sneering put-downs, no snide spitefulness, only mutual
solidarity and the equal love of a good scheme.
Once again, all this is in the libretto, but the music takes the suggestion
of reciprocity and equality much further. As Countess dictates the letter
and Susanna writes it down, the women take inspiration from one another’s
musical phrases, exchanging ideas with a sinuous capacity for response and
a heightened awareness of the other’s pitch, rhythm, and even timbre. They
begin by exchanging phrases, as in a conversation. As the duet continues, how-
ever, their reciprocity becomes more intimate and more complex, as they wind
around one another, ultimately achieving close-knit harmony. Their musical
partnership expresses a kind of friendly attunement that is, we might say, not
only an image of mutual respect but also a reciprocal affection that is deeper
than respect. Neither runs roughshod over the utterance of the other, yet each
contributes something distinctive of her own, which in turn is recognized by
the other and is carried forward.24
This duet has acquired fame in American popular culture because of its
use in the fi lm version of Stephen King’s The Shawshank Redemption, when

24
For discussion of a similar moment in the fi nal movement of Mahler’s Second Symphony,
the contralto and soprano voices wrapping around one another, see Upheavals chapter 15. I now
believe that Mahler, the lifelong opera conductor, may have derived the inspiration for this musi-
cal depiction of reciprocity from Figaro. It is also, as he makes clear, an image of freedom.
 Equalit y and Love at the End of T he Mar riage of Figaro 39

Tim Robbins, the convict who has become the prison’s librarian, figures out
a way to play it for all the prisoners over the PA system and, locking the door,
stops the prison hierarchy from interfering until the duet is done. The men of
Shawshank certainly are not fans of classical music, but they hear something
in this music and stop in their tracks, transfi xed by a promise of happiness. As
Morgan Freeman expresses it, looking back:

I have no idea what those two Italian ladies were singing about. Truth
is, I don’t want to know. Some things are better left unsaid. I’d like
to think they were singing about something so beautiful, it can’t be
expressed in words, and it makes your heart ache because of it . . . , and
for the briefest of moments, every last man in Shawshank felt free.

What do the prisoners hear in the duet? Freedom, they say. But why and how?
First, they cannot help hearing an absence of hierarchy in the evenly matched
voices and a partnership based on responsiveness rather than dictatorial power.
This, in the context of Shawshank, is already freedom. But, as the voices soar
out over the squalor of the prison yard, I think there is more to be heard in it:
the idea of a kind of internal freedom, a freedom of the spirit that consists pre-
cisely in not caring about hierarchy, seeking neither to avoid being controlled
by others nor to control them. Suppose we imagine Tim Robbins playing the
Count’s Vedrò mentr’io sospiro or Bartolo’s La vendetta. Well, those two power-
ful men express, in their own way, an idea of freedom: freedom as power to
dominate, escaping the shame of being dominated. But we know that the men
of Shawshank would not have been transfi xed by that image of freedom: after
all, it is what they live every day. The promise of the duet is not simply a prom-
ise of freedom as reversal, freedom as getting your turn to humiliate the one
who has humiliated you. It is a freedom that takes us beyond that anxious and
always unsettled picture of what liberty, for men, might consist in. It is free-
dom as being happy to have an equal beside you, freedom as not caring who is
above or beneath. And that’s a freedom that does take the mind way away from
Shawshank and from the American society of which that institution is the apt
mirror.
In other words, this music has invented democratic reciprocity. Whatever
the faults of the fi lm—which is in many ways sentimental—this moment
contains a correct insight into Mozart’s politics and into the politics of equal-
ity more generally. You don’t get the right kind of liberty, the idea is, with-
out also having this type of fraternity and this type of equality. To shoot for
liberty without fraternity, as Beaumarchais’s Figaro does, is simply to turn
the hierarchy upside down, not to replace it with something fundamentally
different. If there is to be a new regime, if there is ever to be something like
a politics of equal respect in this world, the suggestion is, it must start by
40 on emotions

singing like those two women, and this means becoming a fundamentally
different type of man.25
To put it a different way, the male world of Figaro is its own prison, as each
man goes through life dominated by rank anxiety. What those prisoners heard
in the duet was the promise of a world without that tension, a world in which
one would then really be free to engage in the pursuit of happiness. The new
regime, as it never has been realized in any nation in the world.26

Creating a Man: “Mischievous Looks,” a “Good

Outside Myself ”
The headings of the last two sections referred to “females” but to “the male
voice,” and it is the male voice, not maleness itself, that the opera associates

25
Where, if ever in opera, do men sing like that (in close-knit interweaving harmonies, each
taking cues from the other)? The duet in Bizet’s The Pearl Fishers comes to mind, but it is not
nearly as complex: the men simply sing together in close harmony. Similar is the wonderful lib-
erty duet (Dio, che nell’alma infondere) sung by Carlos and Roderigo in Verdi’s Don Carlo —close
harmony and, we might say, solidarity, but without responsiveness to the separate moves of the
other. So it would seem that men, in opera, can on occasion attain solidarity and unanimity but
perhaps not responsiveness or attunement. One might also study the Otello–Iago duet in Verdi’s
Othello, where they both swear vengeance together. Here there is an appearance of attunement,
but it is only superficial, since at a deeper level the two are profoundly at odds; insofar as there is
an attunement, it is an unhealthy attunement. I welcome other examples and counterexamples.
As for men and women singing together with the responsiveness of the Countess and Susanna,
the supreme example that comes to mind is a strange one: the fi nal love duet, Pur ti miro pur ti
godo, pur ti stringo, pur t’annodo (“I wonder at you, I enjoy you, I embrace you, I entwine you”)
sung by Nero and Poppea, alone on the stage, at the end of Monteverdi’s L’incoronazione di Poppea.
There, the expressive purpose of the sinuous and intertwining lines (sung, usually though not
always, by two female voices) is to express and represent the reciprocity of good lovemaking—an
idea not irrelevant to this moment in Figaro, as we notice that the women’s capacities for respon-
siveness are sadly unmatched by any similar capacity on the part of the men they love. However,
in Monteverdi the political intention is clearly to show that people are not all of a piece: the bru-
tal tyrant Nero, who has just killed Seneca and many other good people, is capable of the most
respectful and egalitarian sort of sexual passion. Figaro is more skeptical: brutality in the public
domain is very unlikely to be accompanied by anything else in the private and vice versa.
26
Th is new world surely involves transformation on the part of real-life women as well—for
although the world of males has its distinctive pathologies, it would be absurd to claim that the
world of real-life women is a stranger to jealousy and rivalry. (We should not forget Susanna’s
sniping at Marcellina, and vice versa, in that Act I duet—although that rivalry is harmoniously
resolved soon enough.) In this sense we ought to view Mozart’s men and women as symbolic
placeholders for types of human beings that one might be, or become. And Marcellina’s Act IV
aria Il capro e la capretta, like Bartolo’s and Basilio’s often cut in performance, tells us that the new
world will also require change in the position of real-life women: men and women, she says, are
at war with one another in a way unknown in the rest of nature—because “we poor women” are
treated cruelly and subjected to all sorts of suspicions.
 Equalit y and Love at the End of T he Mar riage of Figaro 41

with the endless and exhausting fight against the “lowness” of shame. There
is, however, a male in the opera who does not sing in a male voice: the teenage
boy Cherubino, performed by a female mezzo-soprano. This already seems sig-
nificant, and Cherubino’s education, it shortly emerges, is the focal point of the
opera’s depiction of the new egalitarian citizenship.
Cherubino is usually treated superficially, as a running joke throughout the
opera, and this is, more or less, the way Beaumarchais treats him. His preoc-
cupation with women and sex is indeed the source of much of the plot, as he
turns up repeatedly in places where he should not be, to the consternation of
the possessive males around him. In many productions he is treated as a person
with no sentiments but only very intense bodily desires. Let us, however, pay
closer attention to what he says and what he does.
Cherubino is clearly, in crucial ways, masculine. He is tall (Susanna has to
ask him to kneel down so that she can put on his bonnet), good-looking (Figaro
and the Count are both jealous of him), and sexually active (with his teenage
girlfriend Barbarina)—indeed, very likely, the only male who is actually hav-
ing sex with anyone during the time span of the opera.27 On the other hand,
the fact that he is sung by a female voice forces us to pay attention to the ways
in which that voice, and the sentiments it expresses, differ from all the male
voices in the opera. So, what does Cherubino talk about?
He talks about love. He is the only male in the opera who has the slight-
est interest in that emotion. Certainly, the breathlessness of Non so più
expresses the promiscuous quality, as well as the confusion, of adolescent
infatuation: “Every woman makes me blush, every woman makes my heart
leap.” Still, even when he is reporting his state of sexual obsession, he is
talking about love: “I talk about love when I’m awake, I talk about love in my
dreams, I talk about it to the water, to the shadow, to the mountains, to the
flowers, to the grass, to the fountains, to the echo, to the air, to the winds.” 28
He shows, here, a romantic and poetic conception of what he is after that is
quite unlike the ideas of all the other males in the opera, who all see sex as
a means of asserting domination over a key piece of property in the male
world. The musical idiom, breathless yet tender, is utterly unlike the tense
accents of the adult males. Indeed, it is the musical idiom, far more than the

27
The Countess is eloquent about her husband’s neglect and indifference. So much is made of
the idea of Susanna’s virginity at the time of marriage that it seems plausible to think that she and
Figaro have not yet occupied the bed that he is so anxiously measuring at the opera’s opening.
28
Da Ponte has altered Beaumarchais here in an interesting way: in Beaumarchais, the pas-
sage goes, “Finally, the need to say ‘I love you’ to someone has become so urgent for me that I say
it when I’m all alone, when I’m running in the park, I say it to your mistress, to you, to the trees,
to the clouds, to the wind that carries the clouds and my lost words away together.” Th is comi-
cally confused utterance—he can hardly tell the difference between one woman and another, or
between a woman and a tree—is subtly altered by Da Ponte into something much more delicate,
a mood that the musical idea brings out more vividly still.
42 on emotions

Beaumarchais-inspired text, which makes us see that Cherubino’s sensibil-

ity is poetic and romantic rather than simply energetic.
When we reach the Countess’s chamber, Cherubino’s difference from other
males becomes even more evident. Deeply infatuated with the Countess, he has
decided to make her a present. What sort of present? What naturally occurs to
him is to write a poem, set it to music, and sing it himself. Thus, accompany-
ing himself on the guitar, Cherubino becomes the only leading character in
the opera who sings a solo, that is, whose solo singing represents singing.29
Growing up in a world of sentiment and musicality, he naturally gives his pas-
sion a musical shape.30
The content of that passion (in the beautiful area Voi che sapete) is remark-
able for its utter difference from the arias of all the other males. First of all,
Cherubino simply talks about his feeling of love and about its beautiful female
object. He has nothing to say about other men, and he seems utterly impervi-
ous to all questions of honor, shame, and competition. Second, he is eager to
learn something, and he is eager to learn it from women: “You who know what
sort of thing love is, women, tell me whether that is what I have in my heart.”
All the other men are eager to teach rather than to learn; what they are eager to
teach is a lesson in competitive one-upmanship, and they are eager to teach it
to other males. (Figaro imagines himself as the dance-master running a school
that will teach the Count to dance to his tune; Bartolo is eager to show “all
Seville” that he can defeat Figaro; the Count is eager to show Figaro that his
“cause” is not, as Figaro believes, “won,” but, rather, lost.) Third, Cherubino,
unlike all the other males, is utterly vulnerable, and he makes no attempt to
conceal his vulnerability, which is emotional more than bodily: “I feel my soul
in flames, and then it turns to ice in a moment.” He describes an intense long-
ing that leaves him no peace. Finally, and most remarkably, he locates what
he is pursuing in a place outside of his own ego: “I seek a good that is outside
myself” (ricerco un bene fuori di me). Hearing these words, we realize that no
other male in the opera does seek a good outside himself: all are preoccupied
with winning a competitive victory or shielding the ego from shame.
The music of the aria would tell us all this without the words, and indeed it
communicates, well beyond the words, the young man’s delicacy, vulnerability,

29
I say “leading character,” because the various choruses saluting the Count for his wisdom
and virtue —Giovani liete, Ricevete, o padroncina, and Amanti costanti are presumably to be imag-
ined as real-life singing inside the plot: Figaro at one point says, “The music-makers are already
here.” I say “solo” because of the duet between Susanna and the Countess, the canzonetta sull’aria,
already discussed.
30
In Beaumarchais he simply takes a traditional folk melody and writes his own words to it.
The words themselves express love for the Countess, though they are far less interesting than the
Da Ponte text; the music, however, is utterly banal, the tune of “Malbrough s’en va-t-en guerre,” a
bouncy somewhat aggressive war song.
 Equalit y and Love at the End of T he Mar riage of Figaro 43

and sheer kindness. Indeed it is hardly by accident that people who have utterly
no idea what Cherubino is saying should have found in this aria (as in the duet
between Susanna and the Countess) an image of emotional integrity.31 Here, if
anywhere, Mozart’s music moves well beyond Da Ponte’s text.
How did Cherubino get to be this way, a way that promises real reciprocity in
passion? Answer: he was brought up by women and kept a stranger to the men’s
world. Indeed, we’ve already seen that the prospect of military service utterly
confuses and appalls him. In the scene, at the end of Act I, in which Figaro tells
him what to expect when he goes off to the army (Non più andrai), Figaro’s joke
to Cherubino is that he has lived in the women’s world of sentiment, music,
tenderness, and delicacy—and now, suddenly, he will have to enter a world of
drunken men (they swear by Bacchus) with inflexible necks (collo dritto), tough
faces (muso franco), long mustaches ( gran mustacchi), and “lots of honor” (molto
onor). Now, in Act II, we see more fully how much the young man will have
to unlearn in order to enter this male world: in particular, lovely, sensuous
music. “What a beautiful voice,” says the Countess when Cherubino finishes his
aria—drawing attention, again, to the fact that this is singing singing. Figaro
has already told Cherubino, however, that the world of male honor knows noth-
ing of beautiful music: its only music is “the concerto of trumpets, of shells and
cannons, whose shots, on all pitches, make your ears whistle.”32 The aria itself,
with its boringly foursquare military rhythm, now, in retrospect, contrasts
sadly with the grace and elegance of Cherubino’s composition.
By singing so beautifully, Cherubino shows himself to be a candidate for
fraternity, equality, and the female type of liberty. But before he can be finally
confirmed as lovable-with-good reason, one thing more must happen to him:
he must put on women’s clothes. The plot requires the disguise, but Mozart con-
nects this moment to the deeper sentiments of the heart.
It has often been sensed that Susanna’s tender aria, “Come, kneel down,”
Venite, inginocchiattevi, is a pivotal moment in the opera, that something pro-
found is going on when Susanna, first perfecting Cherubino’s female disguise,
then takes a look at him and sings, “If women fall in love with him, they cer-
tainly have their good reasons” (se l’amano le femmine, han certo il lor perchè).
The music is perhaps the most sensuous and tender in the opera, as Susanna,
asking him to turn around, adjusts his collar and his hands, shows him how to
walk like a woman—and then notices how the guise complements the young
man’s mischievous eyes and graceful bearing: che furba guardatura, che vezzo,

31
Indeed, it is formally very different. The text is a repetitious strophic song, and each strophe
is like the preceding, rather like a simple folk ditty. It is Mozart who supplies the contrasting
middle section, with its more complex expressions of longing, fear, and delight.
32
“. . . al concerto di tromboni / Di bombarde / di cannoni / Che le palle in tutti i tuoni All’orecchino
fan fischiar.”
44 on emotions

che figura! What Mozart slyly suggests, by making this aria so riveting and at
the same time so playful, is that here, in an intimate moment of tenderness,
the seeds of overthrow for the ancien régime are decisively sown.
To begin with, the aria concerns kneeling. There is lots of kneeling in this
opera, and in every other place (until the very final moments) kneeling is a sym-
bol of feudal hierarchy: exalted status on one side; obedience on the other. In
the women’s democratic world, however, kneeling is just kneeling. You kneel in
front of your dressmaker so that she can fi x your bonnet and collar. Kneeling
has no symbolism; it is just a useful action. Hierarchy is simply out the window,
irrelevant, a nonissue. The music itself expresses this thought: instead of the
thumping accents of the quest for honor, we hear little trill-like bursts from the
violins, playful jumpings, like muffled outbursts of laughter, that not only betray
no hierarchy but also positively subvert the whole idea.33 Bit by bit, the woman’s
costume is assembled, the woman’s walk learned—until, at the end, Susanna
surveys, with wonder and amazement, the result she has produced. “Admire
(mirate, wonder at) the little devil, admire how beautiful he is. What mischie-
vous looks (che furba guardatura), what charm, what allure. If women love him,
they certainly have their good reasons.” Cherubino is alluring, it seems, pre-
cisely because, while manly and drawn to women, he is not drawn to controlling
them or using them as pawns in games with other men: instead of domination,
charm, and grace; instead of plots to conceal shame or avenge insult, “mischie-
vous looks,” as he joins the women in their love of jokes and gossip.34
All of this is in the libretto—after a fashion. We can, however, imagine
musical settings of the text that would have signaled irony, skepticism, or bit-
terness (certainly an emotion that we could imagine Susanna feeling at this
time). Instead, the music expresses both tender sensuousness and, as I have
said, with the playful movements of the strings, laughter, suggesting that
these two attitudes go well together and that both are key parts of the woman’s
world. We are now led to recall a feature of the Overture whose significance we
might have missed before: the same type of muffled laughter from the violins
is present there, suggesting that subversive play is a major theme of the opera
as a whole.

33
A possible allusion in the text is to the Christmas carol venite adoremus, O come let us adore
him (which would often be a prelude to kneeling). Here, Susanna says, “Come, kneel down”—but
it is not adoration of the transcendent that she seeks; it is fun and play.
34
Here Da Ponte has made major alterations to Beaumarchais. The stage direction says that
Cherubino kneels, but Susanna does not ask, so the inversion of feudal kneeling is not empha-
sized. Far more important, when Cherubino becomes a woman Susanna says that she, as a woman,
is jealous of him. Th is not only puts rivalry and jealousy into the women’s world, whereas Mozart
and Da Ponte represent that world as a world of reciprocity but also fails to state that a man
is more attractive as a man for behaving in ways that we have heretofore associated with that
world.
 Equalit y and Love at the End of T he Mar riage of Figaro 45

This reading of the aria is shortly confirmed by the duet Aprite, presto, aprite,
as Susanna and Cherubino plot together about getting him safely out of his
compromising hiding place. The two sing, extremely rapidly, in hushed con-
spiratorial voices that show a rare degree of attunement—foreshadowing the
more developed duet between Susanna and the Countess in Act III. Cherubino
shows that he has now, in effect, become a woman: a coconspirator, a voice of
fraternity and equality, and therefore, as if we didn’t know it already, a person
internally free from the bonds of status.
As we look at Cherubino, we realize afresh how unradical Figaro’s apparent
radicalism is. It’s not only that he takes over from the ancien régime its propri-
etary attitude to women; it’s something more global than that. Figaro simply
sees the world the way the Count sees it: in terms of the quest for honor and the
avoidance of shame. He doesn’t understand reciprocity, and he really doesn’t
understand humor. (His idea of a joke is a mean-spirited put-down.35) If the
new world has citizens like that, its commitment to equality and fraternity will
be bound to be problematic. New hierarchies will be thrown up to replace the
old, like ramparts defending the male ego. Could there, however, be citizens
who simply like to laugh and to sing?
In her fascinating reflections on eighteenth-century pornography, Lynn
Hunt has argued that the pornographic idea of the interchangeability of bod-
ies is closely linked to the revolutionary call for democratic equality.36 Legal
theorist Lior Barschack argues that the new subjectivity created by Mozart’s
operas is just this hedonistic idea of sexual freedom.37 No doubt such ideas
were prominent in the eighteenth century, as people (meaning men) tried to
make sense of the new world they inhabited.

35
In a fascinating survey of the use of diminutives in the libretto (sent to me as a personal
communication), scholar Marco Segala fi nds defi nite patterns in the use of different types of
diminutives. The diminutives ending in— etto/a typically connote tenderness and playfulness,
almost never sarcasm or irony; diminutives ending—ino/a often have an ironic or sarcastic mean-
ing. When diminutive (usually in— etto/a) connote affection or play, the speaker is always femi-
nine—with the exception of Cherubino, who has learned to speak the women’s language. The men
of the opera mainly use diminutives (usually in ino/a) to express sneering or sarcasm.
36
Lynn Hunt, ed., The Invention of Pornography: Obscenity and the Origins of Modernity, 1500–
1800 (Cambridge, UK: Zone Books, 1993), 44. Robert Darnton’s earlier study of eighteenth-century
pornography (and, in particular, of the anonymous novel Therèse Philosophe) comes to a subtly
different conclusion: the new idea is not one of intersubstitutability of bodies but rather the
idea of women’s control and autonomy: thus, the relationships that are prized are personal and
long-lasting but include contraception. See his The Forbidden Best-Sellers of Pre-Revolutionary
France, (New York: HarperCollins, 1997).
37
Lior Barshack, “The Sovereignty of Pleasure: Sexual and Political Freedom in the Operas of
Mozart and Da Ponte,” Law and Literature 20 (2008), 47–67. Later, Barshack seems to arrive at a
more nuanced view: “as Mozart saw, the libertine account of humaneness is as one-sided as the
sentimental” (63). I’m not sure I have fully understood his argument at that point, or how it is
related to his earlier contention.
46 on emotions

If I am right, however, Mozart sees the world rather differently and more
radically. The objectification of bodies as interchangeable physical units is
itself, the opera suggests, just one aspect of the ancien régime, which invented
and depends upon the idea that some classes of people, including, prominently,
women, are just oggetti and thus can be used at will in one’s quest for personal
gratification. Seeing bodies as interchangeable is, indeed, a clever route to what
the ancien régime wanted all along: male control and invulnerability. What
would be truly opposed to the ancien régime would be not the democratization
of bodies as interchangeable machines but love. Which, as Cherubino under-
stands, means seeking a good outside oneself, a scary idea. It is, nonetheless,
an idea that Figaro must learn before he can be the kind of citizen Mozart (not
Beaumarchais) demands—and learn it he does, as, in the recitative before his
still-defensive aria in Act IV, he acknowledges both longing and pain. Saying “O
Susanna, what suffering you cost me,” he, like Cherubino, seeks a good outside
himself.
What’s suggested here, then, is that democratic reciprocity needs love.
Why? Why wouldn’t respect be enough? Well, first of all, respect is unstable
unless love can be reinvented in a way that does not make people obsessed
all the time with hierarchy and status. That private obsession, unchallenged,
threatens to disrupt the public culture of equality. But, more deeply, the public
culture needs to be nourished and sustained by something that lies deep in the
human heart and taps its most powerful sentiments, including both passion
and humor. Without these, the public culture remains wafer-thin and passion-
less without the ability to motivate people to make any sacrifice of their per-
sonal self-interest for the sake of the common good.38

Cherubino, Rousseau, Herder: Spaces for Craziness,

“Dispositions of Peace”
Now that we have a general idea of what Mozart is attempting, we may make
our impressions more precise by comparing the opera’s insights about citizen-
ship in the new era to those of two of Mozart’s philosophical contemporaries,
Jean-Jacques Rousseau and Johann Gottfried Herder. Both share with Mozart
the view that a new political culture needs to be sustained by new sentiments,
and both also share with him the view that these sentiments not only must be
the calm sentiments of respect and civic friendship but also must include and

38
See Nussbaum, “Kann es einen ‘gereinigten Patriotismus’ geben? Ein Playdoyer f ür globale
Gerechtigkeit,” in Kosmopolitanismus, ed. M. Lutz-Bachmann et al. (Göttingen: Velbr ück, 2010),
242–76; and Nussbaum, “Teaching Patriotism: Love and Critical Freedom,” University of Chicago
Law Review 79 (2012), 213–50.
 Equalit y and Love at the End of T he Mar riage of Figaro 47

be sustained by, something more like love, directed at the nation and its moral
goals. Here, however, the resemblance ends.
In the important final section of On the Social Contract titled “On Civil
Religion,” Rousseau makes it clear that intense love-like bonds of patriotic sen-
timent are needed to bring citizens together, rendering egalitarian institutions
stable over time.39 Early in human history, he observes, people “had no kings
but the gods” and needed to believe that their leaders were indeed gods. Both
paganism and feudalism were sustained by some such fiction. “A lengthy altera-
tion of feelings and ideas is necessary before men can be resolved to accept a
fellow man as a master, in the hope that things will turn out well for having
done so” (220). These new sentiments must have the intensity of the religious
sentiments they replace, or they will not succeed in holding the new political
order together.
Christianity looks at first blush as if it might be that “civil religion,” since it
does teach the brotherhood of all human beings (224). On further inspection,
however, Christianity has a number of fatal flaws from the point of view of
the political order. First, it teaches people to hope for a salvation that is other-
worldly and spiritual rather than political; thus, “it leaves laws with only the
force the laws derive from themselves, without adding any other force to them”
(224). Second, Christianity turns people’s thoughts inward, as each is urged to
examine his own heart; this teaching produces indifference to political events.
Third and finally, Christianity teaches nonviolence and even martyrdom, thus
teaching people to be slaves. “Its spirit is too favorable to tyranny for tyranny
not to take advantage of it at all times” (225). Christian emperors, Rousseau
argues, ruined the Holy Roman Empire: “when the cross expelled the eagle, all
Roman valor disappeared” (225).
The civil religion we need must inculcate “sentiments of sociability, without
which it is impossible to be a good citizen or a faithful subject” (226). These sen-
timents are based on some quasi-religious dogmas, including “the sanctity of
the social contract and the laws” (226). But what are the sentiments themselves
like? It is clear that they involve an intense love of the nation and its laws. They
also involve a type of fraternity grounded in unanimity and homogeneity: the
person who dissents from the “civil religion” is to be banished “for being unso-
ciable, for being incapable of sincerely loving the laws and justice” (226). Civic
love, then, is incompatible with active critical thought about the political order
and with a sense of the separateness of the individual from the group. The
test for sincerity is unanimity. Furthermore, one thing that citizens must be
unanimous about is the willingness to die for the nation—presumably without

39
I am citing the work in the translation by Donald A. Cress in Jean-Jacques Rousseau: The Basic
Political Writings (Indianapolis: Hackett, 1987), original publication 1762; page numbers are from
that edition.
48 on emotions

thinking critically about the plan for war and whenever the sovereign body
of citizens so decrees. The sentiment of civil love has, then, a strong commit-
ment to the suspension of both individuality and reasoning. Indeed, we might
say, more generally, that the person-to-person dimension is missing, since the
approved sentiments of communal bonding do not lead to or rest upon any sen-
timents directed at individuals, even sentiments of concern and respect.40
Notice, then, that despite Rousseau’s intense hatred of the feudal order, he
has not been able to think his way very far beyond it, in thinking about the
emotions. Civic love, like feudal love, is obedient, hierarchical. (Even if “the
general will” is sovereign, not an individual, it nonetheless bears to the way-
ward individual a strongly hierarchical relationship.) There is no room for the
sort of reciprocity exemplified by Mozart’s women, a reciprocity based on plot-
ting, joking, a sense of the free space within which people can live and be them-
selves. Moreover, despite Rousseau’s attempt, in Book IV of Emile, to substitute
the egalitarian sentiment of pitié for sentiments based upon feudal inequality,
that experiment remains deeply unrealized in his idea of the civil religion, since
the civil religion counters the allegedly excessive meekness of Christianity by
relying, it would seem, on the very ideas of manly courage, assertiveness, and
honor that sustained the ancien régime. There is a shift, in that the object
of civic shame, civic anger, and civic assertiveness is now the nation, seen as
embodiment of the general will. The sentiments themselves, however, feel very
much the same, as the nation seeks to establish itself in the world’s hierarchy
of nations. It’s still true, in this new world as in the old, that “to forget shame
and insult is baseness, complete lowness of status.”
Mozart, by contrast, proposes a radical alteration of the very content of civic
love. No longer must love revolve in any way around ideas of hierarchy and
status. Instead, it must be aspirational: like Cherubino, it must “seek a good
outside myself.”
To sustain truly egalitarian institutions, moreover, this aspirational love
must remain vulnerable to criticism, ultimately to the fact that each individual
has a quirky mind that is not exactly the same as any other. (Indeed, although I
speak of “civic love,” it is crucial to the new conception that there is a family of
types of love that play a role, interwoven, sharing some features but differing
in other respects, as do the people these loves connect.) Rather than Rousseau’s
homogeneity, the Mozartian regime seeks real-life heterogeneity, gives it space
to unfold, and takes delight in its oddness. The fondness of the women’s world
for plotting, joking, every subversion of tradition and obedience, is the sign
of something that ultimately becomes crucial to the Enlightenment, in its
Kantian and especially Millian forms: the idea of the mind of the individual

40
I owe this point to Daniel Brudney, who also points out that the Letter to d’Alembert on the
Theater contains a rather different picture of the preferred types of social interaction.
 Equalit y and Love at the End of T he Mar riage of Figaro 49

as containing an untouched free space, a funny unevenness that is both erotic

and precious. What the women’s world knows is that those “mischievous looks”
are precisely what make Cherubino worth loving (so that, if women love him,
they “certainly have their good reasons”). They also know that the aspirational
nature of his love is deeply interwoven with his capacity for subversion. Civic
love, then, also has a downward movement: it can aspire in a healthy way only
if it is also capable of poking fun at itself, noticing the everyday messiness and
heterogeneity of real people.41
How do we imagine this civic love expressing itself? We connect Rousseau’s
love with solemn public ceremonies, with anthems, with the drumbeat of the
call to arms. Mozart’s love, by contrast, is expressed—well, through many dif-
ferent types of artistic and musical performance but, crucially, through comedy,
including comedy that pokes fun at the call to arms (as in Non più andrai)—and
points out some of the unpleasant realities of what passes in some quarters for
warlike glory: “instead of the fandango, a march through the mud (il fango).”42
At this point, we notice that Mozart has an eighteenth-century ally: Johann
Gottfried Herder, whose Letters for the Advancement of Humanity (1793–1797)
develops a remarkably similar conception of a reformed patriotism that would
need to be inculcated if the world were ever to become a world of peace. Herder
begins by making the point that if patriotism is an attitude toward an entity
called “Fatherland,” it had better figure out what is valuable in the relationship
of a child to its father.43 If we ask this question seriously, he argues, we will see
that we want this love to contain aspiration to genuine merit but also a love of
peace, since we all remember with greatest longing and love the peaceful times
of our childhood. Moreover, what delighted us in those peaceful times was
“games of youth,” so the new patriotism must at the same time be something
playful. Above all, it would never involve blood lust and revenge: “Fatherlands
against fatherlands in a combat of blood [Blutkampf ] is the worst barbarism in
the human language.”
Later in the collection, Herder returns to this theme, making it clear that he
conceives of the animating spirit of the new patriotism as less paternal than
feminine and as requiring a profound gender transformation on the part of
males. Here he alludes to what he has managed to learn about Native American
Iroquois customs, which, he argues, involve casting one of the potentially

41
Th is idea of the two aspects of decent patriotism is a central theme in M. Nussbaum and
J. Israel, Loving the Nation: Toward a New Patriotism, book in progress (under contract to Yale
University Press).
42
Here I echo ideas already developed by Israel in an unpublished paper (untitled), presented
at a conference at the University of Helsinki, June 2009.
43
Letters, extracted and translated by Michael N. Forster in Herder: Philosophical Writings
(Cambridge, UK: Cambridge University Press, 2002), p. 378.
50 on emotions

warring tribes in the role of “the woman” and then requiring all the others to
listen to what “she” says:44

Hence if at some time the men around her are at blows with each other
and the war threatens to become severe, then the woman should have
the power to address them and say to them: “You men, what are you
doing that you belabor each other about with blows in this way? Just
remember that your wives and children are bound to die if you do not
stop. Do you, then, want to be responsible for your own annihilation
from the face of the earth?” And the men should then pay heed to the
woman and obey her. (401)

By dressing the (members of the) chosen nation in women’s skirts and women’s
jewelry, they express the thought that “from now on they should no longer
occupy themselves with weapons” (401). Herder now notes that the members
of the Iroquois nation address one another as “sister-children” and “fellow
female playmates.”
Now to Europe. Herder observes that at one time feudal hierarchy played,
after a fashion, the role of this “woman,” making people keep the peace. Now
that we have rejected feudalism, however, we have to put the women’s clothes
on all of us, in effect, and this means inculcating in all citizens “dispositions
of peace.” His “great peace woman” (whom he equates with “universal justice,
humaneness, active reason”) will seek to produce seven (emotional) “disposi-
tions” in the citizens of the future. First is a “horror of war”: citizens should
learn that any war not limited to self-defense is mad and ignoble, causing
endless practical pain and deep moral degeneration. Second, they will learn
“reduced respect for heroic glory.” They should “unite to blow away the false
sparkle that dances around a Marius, Sulla, Attila, Genghis Khan, Tamerlane”—
until citizens have no more awe for these mythic “heroes” than they do for
common thugs. Herder does not say how this “blowing away” should be accom-
plished, but comedy is clearly a useful technique. Third, the peace-woman will
teach a “horror of false statecraft.” It’s not enough to unmask warlike heroics:
we must also teach disobedience and disrespect to the sort of political author-
ity that likes to whip up war to advance its own power interests. To produce this
result we must teach active critical citizenship: “The universal voice-vote must
be victorious over the value of mere state rank and of its emblems, even over the
most seductive tricks of vanity, even over early-imbibed prejudices” (406). Of
course this critical spirit must be taught in conjunction with an admiration for
and aspiration to what is really fine.

44
Here Herder is apparently paraphrasing a writer named G. H. Loskiel (1740–1814), a priest
of the United Brethren, who published an extensive account of Iroquois customs in 1794.
 Equalit y and Love at the End of T he Mar riage of Figaro 51

Fourth, peace will teach patriotism, but a patriotic love that is “purified” of
“dross,” above all purified of the need to define the lovable qualities of one’s
nation in terms of competition with other nations and even war against them.
“Every nation must learn to feel that it becomes great, beautiful, noble, rich,
well ordered, active, and happy, not in the eyes of others, not in the mouth of
posterity, but only in itself, in its own self” (406). The fifth, closely related,
disposition is that of “feelings of justice towards other nations.” The sixth is
a disposition to fair principles for trade relations, involving a ban on monop-
oly of the seas and a determination to make sure that poorer nations are not
sacrificed to the greedy interests of the richer. Finally, citizens will learn to
delight in useful activity: “the maize stalk in the Indian woman’s hand is itself a
weapon against the sword.” All of these, concludes Herder, are the principles
“of the great peace goddess Reason from whose language no one can in the end
escape” (408).
Herder and Mozart are in harmony. Each sees the need to feminize the cul-
ture of male one-upmanship if civic love is to be productive of true happiness.
The Countess, we might say, is Herder’s great peace-woman Reason, whose
gentleness and refusal to focus on insulted pride show a way in which “all” can
be happy. And Cherubino, her pupil, has learned from her a horror of war, a
horror of false statecraft, and a love of mischievous subversion of the countless
ways in which men try to make the world a world of war. Herder emphasizes
the negative side more than does Mozart: the decent society must teach appro-
priate fear and even horror, not just appropriate love. And he rightly empha-
sizes the fact that critical Reason plays a crucial role in the new approach. In
that sense he is echoing Kant’s call for an approach to world peace that rests
on the Enlightenment value of a critical public culture. But in essence he and
Mozart are on the same page. The adult men are led, in the end, to put on those
long skirts and think about daily life in a reasonable way, rather than a way
informed solely by the insatiable greed of honor run amok. What we have in the
end of the opera is something that would appall Rousseau, a world of craziness,
foolishness, joking, idiosyncratic individuality—and, inseparably, a world of
peace.45 It’s the world that Kant and Mill would have depicted, if either of them
had had a sense of humor.

Transcending the Everyday? Nussbaum contra Nussbaum

What, then, happens at the opera’s close? Temporarily, at least, the male world
yields before the female world, asking for pardon. And then there is a pause. As

45
Is the new temper sufficient for peace? Clearly not, since external dangers may still threaten.
But it seems to be his view that it is necessary for peace.
52 on emotions

Steinberg nicely says, “From Mozart to Mahler, the rest, the musical pause, the
moment of silence is the indicator of a first-person musical voice taking stock
of itself. Music stops to think.”46 And what, in this silence, might the Countess
be thinking, before she says yes?
If she has any sense—and we know that she has a great deal—she will be
thinking, What on Earth does this promise of renewed love really mean? Has
this man, who has behaved badly for years, really become a new person just
because our joke succeeded and he is publicly embarrassed? And when, like the
sensible woman she is, she gives herself the answer, “Surely not,” then she must
think again, asking herself, “but then, shall I accept him as he is, with his arro-
gance, his status-consciousness, his anxiety-driven infidelities? Shall I agree to
live with just the hope or promise, and the occasional reality, of reciprocal love,
rather than its assured stability?”
When, then, after that pause, she answers, “I am nicer, and I say yes,” with
that downward-leaning phrase, she is saying yes to the imperfection in all their
lives, accepting the fact that love between these men and these women, if fre-
quently real, will always be uneven and far from blissful; that people will never
get the entirety of what they long for; that even if men are capable of learning
from women—and Steinberg has nicely shown how Figaro learns from Susanna
a newly tender musical idiom47—nonetheless we hardly have reason to expect
these achievements to be stable, given the pressures culture and upbringing
exert on human development. Indeed, it seems far more likely that Cherubino
will be corrupted by the male world around him than that the other men will
drop their quest for honor and status and learn to sing like Cherubino. Even in
the best of cultures, the aversion to shame and the narcissistic desire for con-
trol are profound human desires; they are unlikely to go away, yielding a world
in which all lovers get everything they want. (And wouldn’t the image of such
a world itself be a narcissistic fantasy that might inhibit the real perception of
another individual reality?)
So, when she says yes, she is agreeing to love, and even trust, in a world of
inconstancy and imperfection—an affirmation requiring more courage than
any of the battlefield exploits mentioned by Figaro in Non più andrai.
And what she agrees to, here, is also what the ensemble also agrees to. The
new public world is a world of happiness in that way. What that seems to mean is
that all present say yes to a world that seeks and aims at reciprocity, respect, and
attunement without being starry-eyed about perfection, a world in which people
commit themselves to liberty, fraternity, and equality while understanding that

46
Listening to Reason, 45. He is talking about a pause in Susanna’s aria Deh vieni, non tardar,
which he discusses so nicely that I refrain from adding anything further.
47
“In his emotional maturity, Figaro is awarded by Mozart with a musical sensuality that
departs from his earlier, metronomic ditties” (45–46).
 Equalit y and Love at the End of T he Mar riage of Figaro 53

these transcendent ideals are not to be attained by exiting from the real world
into a pristine world but rather by pursuing them in this one in episodes of love
and craziness. The new regime will fail if it demands perfection. It will succeed
only on the basis of a realistic conception of men and women and what they are
capable of. But sustaining the hope of fraternity without being starry-eyed (and
therefore, in due course, disillusioned and cynical) requires something like an
unjaundiced trust in the possibility of love (at least sometimes and for a while)
and, perhaps above all, a sense of humor about the world as it is.48
These ideas of trust, acceptance, and reconciliation are not in the text but
only in the music.49 As has long been felt by interpreters, Mozart’s sensibility
is, at the very least, more determinate than his librettists’s and possibly at odds
with it, if we locate in the libretto leanings toward a kind of detached cynicism
that Da Ponte at times exhibits.50 But Mozart’s music is not in some unattain-
able heaven; it is in the middle of our world and in the bodies of those who sing
it; it reshapes the world by reshaping breath itself.
In his magnificent new book, The Musical Representation: Meaning, Ontology,
and Emotion, Charles O. Nussbaum gives us the best picture we have in philoso-
phy to date of the experience of musical listening, the nature of the virtual space
it creates and the mental representations it evokes. At the end of the book, he
adds a chapter that is in many ways underdetermined by the book’s overall argu-
ment, in which he argues that the great interest we take in (Western, tonal)
music derives from our horror of the merely contingent, our desire for an experi-
ence of transcendence and unity that is akin to religious experience. This chapter
contains fascinating material on quite a few philosophers, including Kant, Hegel,
Schopenhauer, Nietzsche, and Sartre—all of whom supply arguments that har-
monize with Nussbaum’s contentions in a variety of ways. It also contains mate-
rial on mystical experience that convinces one that the religion–music parallel
Nussbaum investigates is real and illuminating for at least some music.51

48
Here I think Barshack is perceptive: “Affective intensity” (in Mozart) “does not result in a
retreat from the play of variations and ambiguities which make up everyday existence . . . In the
height of passion, Mozart often invokes the frivolous and the commonplace.”
49
See Kerman, 91; Carter, 120–121, who writes that the text is “perfunctory to an extreme”
and that the musical setting is “magnificent in its serenity and translucence.”
50
I am grateful to Tim Carter for valuable correspondence on this point; and see his Cambridge
Companion. The same tension is even more evident in Così Fan Tutte, where the libretto is cyni-
cal and detached throughout, but the music achieves moments of deeply moving tenderness and
reciprocity—all of which make the challenge of staging it well profoundly difficult.
51
Consider, however, the illuminating argument in Higgins (above n. 10) that music always
promotes an attitude of receptivity, thus cultivating capacities to approach other people and
the world in a “nondefensive, noncompetitive fashion” (156). I am not sure that noncompetitive
dispositions are always promoted, but her claim about the cultivation of receptivity is a power-
ful one, and to the extent that we accept it we would have further reasons to doubt Nussbaum’s
claims about transcendence of the earthly.
54 on emotions

Yet why should we be so inclined to suppose that music offers one particular
type of good to human life rather than many types of good? C. Nussbaum is
far too subtle to claim such a thing outright,52 but in his insistence on this one
function of music he at least suggests the primacy and centrality of this type of
good. Like, and continuous with, philosophy, however, music would appear to
assume different argumentative positions, seeing the world from different and
contrasting points of view, in such a way that Steinberg’s metaphor of “listening
to reason” appears more apt than C. Nussbaum’s idea of a single type of experi-
ence. (Indeed, religion itself contains many types of experience—including the
mystical impression of transcendence, but including, as well, the passion for
earthly justice and the acceptance of an imperfection in earthly striving.)53
I have tried to offer a reading of Mozart’s opera that makes it plain that it
offers a different sort of happiness, a happiness that is comic, uneven, uncer-
tain, wary of grandiose claims of transcendence. Indeed, the music itself laughs
up its sleeve at pretensions of that sort (as in those little muffled bursts of
laughter in the Overture and in Venite, inginocchiatevi).54 C. Nussbaum may well
respond that opera, that impure mixed art form, which relies on real bodies
and real sights, is not the musical medium that his argument (based on music’s
disembodied and invisible nature) addresses. Even the human voice itself
appears an anomaly within the Nussbaum conception of a musical art without
bounded spatiotemporal existence. All musical instruments refer in some way
to the human body, but the voice, alone among the instruments, is a part of the
body and always expresses bodily frailty as well as potentiality.55
Just as there is a love that seeks transcendence and a love that repudiates
that aspiration as immature and a precursor of disillusionment, so too, I think
there are both sorts of music. It is no accident that Beethoven is on the cover

52
Indeed, he states that his intention is “to argue that the assuagement of the horror of the
contingent came . . . to be one direct proper function (though by no means the only one) of the
musical representations belonging to the musical style under consideration in this book” (286).
He also refers to the analysis in the chapter as dealing with “certain central cases” of musical
emotion (295).
53
Indeed, I’d be inclined to say that Nussbaum’s characterization of religious experience is
more at home in Christianity than in Judaism, with its emphasis on the earthly nature of our
ethical duties.
54
And if I am right about the allusion to venite adoremus, the aria quite directly pokes fun at
the search for transcendence.
55
Here a comment made by Mollie Stone, assistant conductor of the Chicago Children’s Choir,
is illuminating. Describing the contribution the Choir makes to the political and social develop-
ment of children from a wide range of ethnic and racial backgrounds, she commented that the
children become close to each other because they actually share their breath with one another, a
kind of physical reciprocity that is much more intimate than anything that would be involved in
orchestral performance (Interview, June 5, 2008). Th is comment fits well with the fi ne analysis
of our physical engagement with music offered in Higgins’s The Music of Our Lives (above n. 11),
especially 150–156.
 Equalit y and Love at the End of T he Mar riage of Figaro 55

of C. Nussbaum’s book56 and is a major source of his musical examples. But

the yearning for the transcendent that is indeed embodied in Beethoven’s ver-
sion of the Enlightenment is first cousin to cynicism: realizing that the world
embodied in the Ninth Symphony and Fidelio doesn’t really exist, what can one
do but make a sour face at the real world?57
If, however, one follows Mozart’s version of Enlightenment politics, one will
still see that the world as it is needs a great deal of work, and one will not stop
aspiring to get that work done, making the world of the male voice somewhat
more like the world of the female voice, with its commitment to fraternity,
equality, and liberty. One will not stop seeking to educate young men to love
music rather than the concerto of shells and cannons. One will, however, also
at the same time embrace real people—even men!—as they are, and one won’t
stop loving them because they are (no doubt like oneself)58 a mess. That, sug-
gests the pause within the music, is a more hopeful direction, if not the only
possible direction, for a workable conception of democratic political love.

Appendix
Emotional Expression in Music: Upheavals of Thought
For those unfamiliar with the approach to the expression of emotion in music
that I defend in Upheavals of Thought (chapter 5), it seems worth summarizing
briefly its main contentions. In the earlier chapters of the book, I defend a con-
ception of emotions according to which they all involve intentional thought or
perception directed at an object and some type of evaluative appraisal of that
object made from the agent’s own personal viewpoint, which ascribes signifi-
cance to the object in terms of the agent’s scheme of goals and ends. Thus, we
do not grieve for every death in the world but only for deaths of people who
appear to us to be important in our lives; and so forth. These appraisals need
not involve language or even complexity: I argue that most animals make at
least some such appraisals of objects and have emotions in consequence. All
that is required is that they see the object (a bit of food, say) as good from the
point of view of their own pursuits and goals.

56
A cover chosen by the author.
57
As was done in a remarkable concert version of Fidelio directed by Daniel Barenboim in
Chicago several years ago, in close connection to his political activism in the Middle East, in part-
nership with the late Edward Said. The added text by Said expressed the message that Fidelio just
is unreal (just as unreal, Said suggests, as the idea of a decent Israel), and we should all be both
angry and pessimistic about the world as it is.
58
Because, after all, as I’ve said, the representation of female reciprocity in the opera does not
by any means imply that real-life women are free from narcissism and the urge to control others.
56 on emotions

I then investigate the role of noncognitive components (feelings, bodily

states) in emotions, and I argue that, although some such elements are likely to
be present in most of our emotional experience, these noncognitive elements do
not have the constancy and regular association with the emotion type in ques-
tion that would be required if we were to include them in the definition of an
emotion of a particular type. Even with an emotion as simple as fear, which is
indeed frequently associated with something like shivering or trembling, there
are numerous counterexamples—including the very common case of the fear
of death, which most of us have most of the time, in a way that has psychologi-
cal reality and motivational power, but (usually) without any conscious aware-
ness of trembling or shaking. With other more complex emotions, for example
grief compassion, it is not easy even to begin to identify the bodily feelings
that would belong to it. And often, even when we think we have identified such
elements (grief feels like a pain in the stomach, say), we find, on closer inspec-
tion, that one may continue to have grief over time while these bodily mani-
festations change, often greatly. (A grieving person may sometimes feel achy,
sometimes exhausted, sometimes endowed with extra energy, yet it would be
wrong to say that she is not still grieving.)
This general analysis is then complemented by an analysis (in chapter 3) of
the ways in which social norms embodied in cultural traditions produce subtly
different emotional repertories for actors in those traditions. Thus, anger is in
a way a cultural universal, but specific forms of anger are very strongly shaped
by social norms regarding what an insult is, what honor is, and so forth. This
chapter is of obvious relevance for the account of Figaro given here.
I then study (in chapter 4) the developmental character of emotions: the fact
that our earliest emotional experiences date from a time at which we cannot
use language and are unable, even, securely to identify objects and individuate
them one from another. And those archaic patterns often persist into adult life,
underneath the often sophisticated structure of adult love and grief.
I then turn to music, asking how we can best account for the fact—which
seems evident to most listeners—that music embodies and expresses emotions.
In other words, the fact that we say, with good reason, that this passage is joy-
ful, this one expressive of deep grief, and so forth. We ascribe these emotional
properties to music with great specificity, indeed with no less specificity than
we would find possible when talking about poetry. (For example, we can talk
about the specific type of love embodied in the Liebestod of Wagner’s Tristan,
and contrast it with the very different type of love embodied in Cherubino’s Voi
che sapete.) I point out that this question is not the same question as the ques-
tion about the actual emotional experiences of the listener, since the listener
might be distracted, or musically ignorant, etc. But, following the lead of Wayne
Booth’s illuminating analysis of literary response, I connect the expressive
properties of the music with the experience of the “implied listener,” meaning
 Equalit y and Love at the End of T he Mar riage of Figaro 57

a listener who follows, knowledgeably, attentively, and appropriately, the musi-

cal experience mapped out in the work.
I now point out that philosophers have had difficulty explaining how music
could embody anything like an emotion. On the one hand, some theorists
(e.g., Eduard Hanslick) have said, correctly, that emotions embody evaluative
thought—but they have not been able to see how music, being nonlinguistic,
could possibly embody such thought, so they have ended up denying that music
can embody emotions. On the other hand, other theorists (e.g., Schopenhauer,
Suzanne Langer) have started from the observation that music does indeed
embody emotions, but—agreeing with Hanslick that all thought is essentially
linguistic in nature—they have concluded that emotions can’t involve thought
after all but must be seen as stirrings or movements of the blood, without any
intentionality or cognitive content.
The mistake made by both is to suppose that all thought is essentially linguis-
tic in nature. By this point, I’ve already argued that both nonhuman animals
and small children have many nonlinguistic emotions, involving perceptions
of what is salient for one’s own well-being. What now needs to be added is the
idea that these nonlinguistic experiences need not be archaic or primitive; lan-
guage does not hold a monopoly on cognitive sophistication. Once we see that
a nonlinguistic form of representation can contain as rich an array of possibili-
ties as can language, we are ready to start thinking seriously about emotions
in music.
Next, following the guidance of Mahler’s letters, Proust’s remarks on
music, and Paul Hindemith’s excellent book, A Composer’s World, I argue that
music differs from language in one important way: it is not the shopworn
medium of our daily utilitarian communications, and thus it can often seem
to have a superior power to illuminate the depths of the personality. In virtue
of not being the language of habit, music possesses many properties in com-
mon with dreams.
I argue that the ways in which music expresses emotions are shaped by cul-
ture and by the history of a particular type of musical art form, as well as, more
specifically, by the particular composer’s own expressive development. I am
unable to hear music from a tradition unknown to me and to identify securely
its emotional content. The “implied listener” then must be well educated in
the specific musical tradition, including the composer’s own way of expressing
musical ideas. Thus, it is not possible to give an accurate account of a passage in
a Mahler symphony, say, without being aware of how he typically uses the oboe,
or the harp, and so forth.
Finally, I discuss the vexed question of the relationship between music and
text, arguing that although a verbal text may give a definiteness of reference
to music that it would otherwise lack; nonetheless the text often underdeter-
mines the emotional trajectory of the music. Studying two songs from Mahler’s
58 on emotions

Kindertotenlieder, I argue that the fact that these songs are about the deaths
of two children is supplied by the text, yet the precise nature of the grief they
express (as to whether it is consoled by religion, or ultimately hopeless) is
underdetermined by Rückert’s text and is supplied by the music. This discus-
sion is of obvious relevance to my account of Mozart and Da Ponte here.

Acknowledgment
For comments on earlier drafts, I am grateful to Douglas Baird, Daniel Brudney,
Emily Buss, Tim Carter, Jeff rey Israel, Douglas Maclean, Charles Nussbaum,
Sara Protasi, Adam Samaha, Geoff rey Sayre-McCord, Marco Segala, Michael
Steinberg, Cass Sunstein, Miira Tuominen, Susan Wolf, and Diane Wood.
 3

Spectator Emotions
Paul Woodruff

My subject comes from a review Solomon published about Bijoy H. Boruah’s

book, Fiction and Emotion (1988).1 Solomon ends his review as follows:

I would add that the problem itself, focusing on emotions as a more

or less generic concept, ignores the very dramatic difference between
such “spectator” emotions as sympathy and horror, and more engaged
emotions such as fear, jealousy, and embarrassment, a perhaps fatal
result of such heavy reliance on philosophical argument alone at the
expense of a rich sampling of well-explored examples.2

Groundwork
Spectators, as such, are detached, but emotions are not. Emotions (1) typically
motivate their subjects to action and (2) typically take their subjects as pri-
mary objects (one feels fear for oneself, one feels pity as a pain in oneself, etc.).
Spectator emotions appear to do neither of these things. If you cast yourself
as a spectator, you sit back and watch, insulating yourself from any possibility
of engaging actively in the events you are watching. You may, however, react
passionately to what you see, but your passions have objects fairly remote from
you—the events from which you have insulated yourself as spectator, and the
people who are agents for those events.
This paper considers how emotions function for people who are in specta-
tor mode. I will begin by clarifying what I mean by “spectator” and “emotion”
and then proceed to discuss four questions. First, I ask whether spectator
emotions constitute a species more or less distinct from that of the emotions

1
Boruah (1988).
2
Solomon (1990).

59
60 on emotions

Solomon calls “more engaged”: Could you experience as a spectator any emo-
tion that you could experience directly, and vice versa? Or not? I call this the
two-species problem, and I will argue for considering spectator emotions not
as a distinct species. Spectator emotions are regular emotions experienced in
a certain way.
Second, I ask whether spectator emotions can really be genuine emotions,
since they merely incline spectators to action without actually leading them
to act. This is the action problem. I will argue that emotions need not lead to
action, so long as they lead to an inclination to act.
Third, I ask whether spectator emotions can be genuine emotions in view
of their detachment from the subject’s own concerns. How can emotions be
directed merely at the people and actions watched, that is, at objects entirely
outside the subject’s direct concerns? I call this the engagement problem.
A fourth problem, the knowledge problem, has been much discussed.3 It
appears that full bore emotions require knowledge, but (a) fiction entails false
propositions and (b) knowledge can’t be of false propositions. So it appears we
cannot have emotions in response to fiction. Examples that raise the knowledge
problem are generally of spectator emotions, and we will see that our account
of spectator emotions will provide an answer for this problem as well.

Spectator Mode
Watching a TV monitor from a densely crowded medical waiting room, on
September 11, 2001, I felt horror, as two tall buildings erupted in plumes of
smoke, and I imagined the terror of those inside who must have been trying to
escape, and then I felt a deeper horror as the buildings collapsed with crowds
still inside them. There was nothing I could do, except to react emotionally. I
was not involved; I was a spectator of a distant event.
I soon became aware of responses that were more directly engaged than my
own. I was told that the x-ray technician, on whom we all were waiting, would
shut down his work until he heard that his sister, who had been inside one of
the buildings, was safe. He was more personally involved than a mere spectator
could be; through his love for his sister, he had something at stake that I had
not. Meanwhile, I saw an old man drop to the floor in a fetal position and heard
him call on the president to avenge this crime against the country, a crime he
felt was against us all. He was more engaged than I, as I had not considered this
event as about me in any way. The old man had felt the event as one that was

3
The best discussions, in my view, are to be found in Walton (1990, 241–255) and Carroll
(1990, 62). See also Woodruff (2008, 160–162). My thinking has evolved since writing that sec-
tion. I am persuaded by Boruah’s main argument in Fiction and Emotion that what he calls evalua-
tive beliefs are essential to emotions while existential beliefs are not.
 Spectator Emotions 61

directly about him as a citizen of a nation attacked. Similarly, one may watch
a football game as a spectator, but observe fans taking every success or failure
personally, so closely are they engaged with the team they support.
In deference to Solomon, I have begun to explore the details of my experience
in the waiting room, to show by example that there are modes of engagement.
My initial response was in purely spectator mode. The people in the burning
towers were not spectators; their fear was entirely their own. By contrast to
both extremes, the technician and the old man on the floor were more directly
engaged than I without moving entirely out of spectator mode. They were in a
mode that combines spectatorship with direct personal involvement.
My distinction here is between modes in which emotion can be experienced;
it is not a distinction between levels or degrees of engagement. Looking back, I
do not believe that I was any less engaged than the others in the waiting room.
My emotions were as strong and deep as theirs; I was totally consumed by hor-
ror, dread, and fear. But my emotions were not about me; they were about other
people. That made me a spectator pure and simple.
What came in degrees in this case was the directness of the engagement.
Others were not entirely in spectator mode; they were more directly engaged;
by this I mean that they were emotionally engaged to some extent on their own
behalf or on behalf of someone close to them. They were nevertheless in specta-
tor mode because, like me, they were constrained to be spectators. When your
engagement is on behalf of other people not connected with you in any way,
and you recognize that you are unable to do anything but watch and feel, then
you are engaged in purely spectator mode. But, as the example shows, others
may be spectators and yet more directly engaged.
Sitting in a roadhouse window, casually looking out at the highway, I may
watch a fourteen wheeler jackknife across the road, while three cars slew into
it and a fourth loses control, crosses the median, and collides with a fifth com-
ing the other way. Horrifying. I feel deeply, but I am not directly engaged. I will
drive more carefully after lunch, but there is nothing else I can do about what I
have seen. Again, in this case I am in pure spectator mode.
Watching a play in the theater, a game in a stadium, a pickup match of ulti-
mate Frisbee in the college quad—these are all opportunities for me to experi-
ence emotions in spectator mode. The producers for the theater and the team
owners for the stadium know that I am to be a mere spectator. They do not want
me interrupting the play or the game by bursting out on stage or on the playing
field during the event. But they hope to engage me as intensely as possible even
so, and they do everything they can to ensure that my emotional engagement
is not attenuated by my being in spectator mode.
Sometimes, however, what starts as a spectator experience ends as some-
thing else. Engagement obtrudes and pushes me out of spectator role. Watching
a production of Hamlet I am startled to see that the actor playing the lead role is
62 on emotions

my former girlfriend, whom I still love but who has discarded me for a woman
lover. I cannot be a pure spectator any longer. This evening is no longer about
Hamlet for me; it is about my spoiled relationship and my anger at the lover who
betrayed me.
Identification also can impose itself on a spectator. Suppose I am watching
a play staged by my nephew with a company of traveling players, The Murder of
Gonzago. Suddenly I find that this play is about me—about my action in killing
my brother. I rise from my throne, call for lights to take me to my chapel, and
try to pray. From the moment I saw that the play is about me, I was not in spec-
tator mode, and soon I ceased watching altogether. This is the Claudius example
from Shakespeare’s Hamlet, an extreme case of identification. Claudius finds
himself watching himself, seeing himself as criminal for the first time, and
for the first time wishing to pray for absolution. Meanwhile the performance
of the play, as a piece of theater, is ruined. No one is watching it now. Audience
members look to their safety. The actors fold their costumes away and depart
hastily. There is no theater without watchers in spectator mode.
Weaker forms of identification than Claudius’ occur in theater and else-
where. Authors may write fiction to engage readers through identification;
moviemakers usually try to entice their audiences into identifying with an
attractive and successful lead character. But any level of identification attenu-
ates the spectator mode and renders the watchers less attentive to what they
are actually watching. The art of theater, as I have argued elsewhere,4 generally
aims to put its watchers in spectator mode, so that they pay attention to the
actions that the art presents for their watching. The art of theater discourages
direct engagement by the audience, at least most of the time, and is totally
opposed to extreme identification of the sort experienced by Claudius.
Some forms of theater (say, romantic melodrama) thrive on identification,
but identification limits the degree to which audience members can carefully
or thoughtfully attend to what they are seeing. If I identify with a character in
a play or movie, I will resist facts that do not fit my imagination—either by not
noticing them or by breaking off my identification. Suppose I am identifying
with the woman in a romantic comedy, and she is madly in love (as I would like
to be) with the handsome male lead. If this male lead then shows signs of being
gay, I may refuse to notice these (like the woman with whom I am identifying)
or else I may break off altogether and leave the theater. I cannot watch this
action through identification, though I or another watcher could do so in a less
direct way.5
Let this suffice for an account of the spectator mode. The traffic accident
example is the clearest case of pure spectator mode. Others shade off. But an

4
Woodruff (2008, 142–143).
5
Woodruff (2008, 176–179).
 Spectator Emotions 63

account of spectator emotions should work for the pure cases, in which the
subjects are as personally disengaged as possible.

Emotions
A genuine emotion, on the theory I propose, has four features:6

Subjectivity

An emotion is a feeling that someone consciously has, so for every occurrence

of an emotion there is a subject who feels it.

Motivation to Action

Not every conscious feeling is an emotion. An emotion moves its subject

toward at least one specific action—that is to say, it makes the person who
feels the emotion also feel like doing something. Fear, for example, makes
you feel like running away; shame, like vanishing from the spot; disgust, like
being sick; pity, like giving comfort; regret, like reliving and repairing your
life; grief, like weeping; horror, like shutting your eyes or calling for help;
guilt, like seeking absolution. (Despair does not make you feel like doing any-
thing at all; it is a most unsatisfactory feeling and not, by my defi nition, an
emotion at all. It is a bottomless pit yawning in front of you, a vacuum left
when you cannot fi nd the emotion that would lead you toward the action that
would make you feel better.)7

Intentionality

Actions are chosen by their agents. If in my fear I run from the bear that
frightens me, I am taking action (albeit stupidly; never run from a bear).
Actions are chosen, and they are chosen in relation to the things of this
world. Because actions must be connected to objects, emotions must be con-
nected to objects as well. I cannot choose to flee, and not choose to flee from
something in particular. Therefore the action requirement entails that emo-
tions be consciously directed at objects. In this case the object is the bear;
she is the thing I fear and feel like running away from. Here is the bear that
has put fear into me, and she is the particular bear I feel like escaping. If the
bear threatens my child, then my fear has two objects—the bear, which I

6
Th is account of emotions is similar to the one I gave in Woodruff (2008, 154–157).
7
Deigh (1996) has helped me to an understanding of the difference between despair and
guilt.
64 on emotions

fear, and the child, for whom I fear and whom I feel like rescuing. Th is sec-
ond object seems to vanish in the common case in which I am afraid only
for myself, but we need to keep the possibility of a second object in mind
for any emotion, although the need is most apparent for emotions involved
in loving or caring. The fi rst object is what the emotion is at; the second is
what (or whom) the emotion is about. Of course, the second object does not
really vanish in the common case; we simply do not bother to specify it: I am
afraid of the bear (fi rst object) and on behalf of me (second object). So emo-
tions pick out objects, and their ability to do so is called “intentionality” by
philosophers.

Judgment

I follow an ancient tradition adopted by Solomon in holding that emotions

are ways in which we understand things we encounter and pass judgment on
them.8 My fear of the bear expresses the judgment that the bear is dangerous;
my envy of your brilliant success expresses my judgment that your success is
indeed brilliant. And so on. If I am unable to fear the bear, then I am unable to
understand the situation. This is true when the bear threatens me, but it is also
true when the bear threatens someone else.
Judgments are to be evaluated with respect to information available to
the judge. A good judgment may turn out to be false, and a bad judgment may
turn out to be true. The bear may turn out not to be dangerous, although all
the evidence available to me indicated that it was. If it turns out to be a pet
bear looking for human love, rather than lunch, I have still shown good judg-
ment in feeling like taking defensive measures. Even if it turns out that you
are pretending to be a bear, making truly bear-like sounds in the underbrush,
my fear might express good judgment, although it would represent a failure of
knowledge. (I will discuss mimesis and fiction in the section on the problem of
knowledge.)
The judgment I express through my envy of your success may be both false
and good at the same time. A judgment is good when it accords with the evi-
dence available to the person making the judgment. In fact, the success I envy
may be poisoning your life, though I cannot know it. But the evidence available
to me supports my judgment that your success is wonderful for you, so I may
still be showing good judgment in feeling envy over your success.
Embarrassment presents similar opportunities for good judgments
that are false, that is, when evidence supports the false judgment that the
embarrassment expresses. But false judgments may of course be bad as well.

8
Solomon (1977, 185–191). An elegant more recent statement of the new Stoic view of emo-
tion as judgment is given in Nussbaum (2001, especially at 19 and 43–44).
 Spectator Emotions 65

Suppose I am embarrassed over the quality of my latest book as it nears

publication, because I imagine, without evidence, that it is full of foolish
errors. In fact, no one will be able to spot an error, and the book is not at all
embarrassing. I am being foolish. In that case my embarrassment represents
a judgment that is false and shows bad judgment as well, since I have no
evidence to support it. Still, the embarrassment is real, and it is a genuine
emotion. I am embarrassed about mistakes that do not exist, but I am truly
embarrassed.
Many emotions express combinations of judgments.9 Some emotional
responses are so complex or incoherent that no one could pin down precisely
what judgment they express. That is why I do not offer a coherence requirement
for emotions. Emotions do not need to be coherent.10

Summary

A standard occurrence of an emotion, then, is an occurrent feeling for which a

complete account should aim to identify four things:

1. Who has it? Subjectivity

2. Toward what action? Action
3a. What or whom is it at? Intentionality (a)
3b. About what or whom? Intentionality (b)
4. What sort of judgment does it express? Judgment

The Two-Species Problem

The question is whether spectator emotions comprise a distinct species or
whether any emotion can be experienced in either of the two modes. Solomon
proposed two lists, suggesting a two-species theory. I argued against this in a
brief letter to him soon after his review appeared in print. I thought then (as I
do now) that the art of theater aims to arouse virtually any emotion in people
who are in spectator mode and to arouse emotions as intensely in theater as
those emotions can be felt in the world of our experience outside theater. That
is often the aim of theater. I know that the art of theater does not always suc-
ceed in this way, but I know that it can. It has with me.

9
See Nussbaum (2001, 321) for the case that compassion has a three-part cognitive
structure.
10
See Nussbaum (2001, 35) on the potential for incoherence in judgment through emotions.
66 on emotions

Solomon listed two emotions as spectator emotions: sympathy and hor-

ror.11 Additional candidates are wonder and awe.12 On the other side of the
ledger Solomon listed fear, jealousy, and embarrassment. Taking my cue from
his insistence on examples, I will examine these lists by describing some cases
in detail.

The Spectator List

Horror, to begin with, I have illustrated as a spectator emotion in the case of
the traffic accident. But horror can be experienced more directly. Looking at
my left hand mangled by my mismanagement of a table saw, I was horrified
at what I saw, horrified on my own account. Looking back on an experiment I
conducted with fire as a small child, I am horrified at myself and at what I did.
The word “looking” in both examples, however, shows that I am a spectator of
myself. That is true, but it does not entail that I am in pure spectator mode.
When I am the object of my own horror, I am directly engaged as the object of
my own emotion to the highest degree of directness.
Horror seems in some cases to be disabling of action, and this would seem to
put horror solidly on the spectator list. But on this point, horror is analogous to
fear, which can devolve into panic—a feeling that disables both judgment and
action and is therefore not an emotion. On the other hand, fear may stay at the
level of emotion, in which case it can motivate one to take appropriate action.
In a state of horror at my mangled hand, I could have been paralyzed. But I
was not. I judged that I needed help and called for help, so I was not merely a
spectator of myself. I did something as a result of the horror. Ideally, we would
have two words (as we do for panic and fear), one for the paralyzing horror that
makes one a mere spectator, and the other for the sort of horror that could
lead to action. I conclude that horror has spectator features, but these do not
prevent it from leading to direct engagement of the sort we find in emotions
on the other list.
Sympathy is first on Solomon’s list of spectator emotions. In traditional
usage the word “sympathy” means fellow feeling in one form or another; in
recent usage it more often means compassion. In traditional usage as fellow
feeling, “sympathy” is hard to distinguish from what “empathy” has come to
mean. Fellow feeling entails a spectator relationship, and so does compassion.

11
Solomon did not list laughter as a spectator emotion, but I expect some readers would like
to see it on the list. I have argued elsewhere that laughter is not an emotion at all. Emotions
engage us; laughter, blessedly, allows us to disengage. On laughter as a defense against emotion,
see Woodruff (1997) with Woodruff (2008, 188–190).
12
Wonder and awe are offered by Nussbaum (2001, 53–54) as challenges to her thesis that
emotions are eudaimonistic, that is, connected to the personal goals and projects of the person
experiencing the emotions.
 Spectator Emotions 67

To experience either of these, you have to be paying attention to someone else.

It remains to see whether they are spectator emotions in our sense.
Compassion first: I understand compassion in the classical way as a painful
feeling of sorrow on behalf of someone who is suffering something bad, for
either of two related reasons:

(a) The suffering is undeserved and could have happened to me, and I therefore
understand what that feels like (as in the case of a civilian prisoner of war).
(b) The suffering was earned by an action that was wrong but I understand why
the sufferer took that action; indeed, I feel that in those circumstances I
might well have done the same.13

Compassion is always for someone else; it may be experienced only by a per-

son who is paying attention to someone else and is therefore functioning as
a spectator. As in the case of horror, however, this spectator status does not
preclude my having compassion for myself if I am able to see myself as an other.
And more clearly than in the case of horror, compassion involves judgments
that may lead to action. The more compassionate judge gives a lighter penalty
to the guilty prisoner, because she understands how the prisoner came to do
what he did in spite of his generally good character. The compassionate traveler
pulls the child from the well and hands it to the distraught mother. After a
death, the compassionate friend assists the family in burying their dead, at
the risk of his own life.14 I conclude that compassion, like horror, has spectator
features that do not preclude the sort of direct engagement that puts an end to
spectator status. Compassion can put you in the center of the action.
Fellow feeling second: in my book on theater I have written a chapter dis-
tinguishing nine different forms that what we now call empathy might take.15
I will blur those distinctions here and make a brief comment on the genus.
Sympathy as fellow feeling is not a true emotion by my definition, because sym-
pathy with another person can lead to a wide variety of actions and express a
wide variety of judgments. If you are angry, and I am in sympathy, I feel like
hitting the person who hurt you, and I judge that this person has done you an

13
An example of the fi rst: Heracles’ wife in the Women of Trachis feels compassion for a woman
who has been, like her, won at spearpoint (lines 243, 303–313; Meineck and Woodruff 2007, 74,
76–77). An example of the second: Odysseus feels compassion for Ajax, even though Ajax has just
tried to murder him, because he (Odysseus) recognizes that a rage similar to Ajax’s could drive
him too around the bend into homicidal mania (opening scene of Sophocles’ Ajax; see note 14).
For my discussion of compassion as the understanding that seasons justice, see Woodruff (2011,
98–109). For a more thorough and tightly defi ned concept of compassion, see Nussbaum, 297 ff.
14
So Odysseus secures the proper burial of Ajax, even after Ajax has tried to kill him. Sophocles,
Ajax, 1332–1345 (Meineck and Woodruff 2007, 58).
15
Woodruff (2008, 165–185).
68 on emotions

injustice. If you are grieving, and I am in sympathy, I share your sense of loss;
I judge that you have in fact suffered a great loss, and I feel like trying to make
you feel better.
Such examples show that far from being a specific emotion, sympathy sets
you up for a variety of emotions related to the person with whom you are in sym-
pathy. The same is true of empathy and, for that matter, love, which Solomon
also treats as a distinct emotion. On my theory, sympathy (as fellow feeling),
empathy, and love have the structure of Aristotelian virtues. Such virtues are
dispositions to a range of emotions specified for each disposition.
Courage, for example, shows when you experience confidence in the face of
something you should face (such as physical danger in a good cause), or you
experience fear in the face of something you should avoid (such as betraying
a beloved friend). Courage makes you feel like doing different things on dif-
ferent occasions and expresses different judgments. What unites the various
emotions and judgments that derive from courage is their connection to the
balance between fear and confidence. What unites the various emotions and
judgments I have because I love you is their connection to you.
Even if sympathy as fellow feeling does not belong on a list of emotions,
Solomon would be right to observe that it belongs to spectators. The emotions
that come to me through fellow feeling with you come to me only because I
am paying attention to you, as a spectator. You, not I, are the primary object
of whatever emotions I have through sympathy for you. And that makes me a
spectator when I am in sympathy with you. But the emotions that I may experi-
ence through sympathy include many from the regular list—grief, fear, anger,
joy, and so on. And these are not exclusively spectator emotions.

Nonspectator Emotions
This result leads to a new way to ask the question: Could sympathy or empa-
thy or love lead me to have any emotion in spectator mode? Or are some emo-
tions so personal that I cannot feel them except on my own behalf? Turning
to Solomon’s second list, we find fear, jealousy, and embarrassment. To these I
would add shame and envy as highly personal emotions unlikely to turn up in
spectator mode.
Fear I have already treated in both modes: I can fear that the bear will take
me for her lunch; or I can fear that she will take the child for her lunch. In the
first case, it would be pathological for me to go into spectator mode, as if I were
hovering above myself, watching the bear prepare for her lunch on my body.
This fear is truly about me, and I should feel it that way. In the second case,
however, I can be in purely spectator mode—if, that is, I have no connection to
the child and am unable to act. I am confined to a wheelchair; I have no voice
and no cell phone. I can do nothing. At the same time, the child is not related
 Spectator Emotions 69

to me; I have no children of my own; I do not even like children very much. So
I am merely a spectator.
As I observed at the start, spectator mode is a matter of degree. If I am
unable to act but the child is my own, I am much more directly engaged, and
consequently less in spectator mode—like the audience member who is dis-
tracted from Hamlet by knowing that his former lover is playing the title role.
This observation leads to a resolution of the two-species question. There is no
bright line between spectator emotions and more personal ones. The directness
of emotional engagement is a matter of degree.
Jealousy, envy, shame, and embarrassment are more personal than fear and
therefore less likely to be felt in spectator mode. These emotions seem to entail
a watching relationship that is not detached and therefore undermines specta-
tor status. But not completely. Suppose I am watching a fellow student who
has been called to the blackboard to solve a problem in math. He fumbles and
stumbles; he is badly hung over and has not studied the material. The instruc-
tor keeps him at the board for an excruciatingly long time. I am acutely embar-
rassed for him, yet I am clearly in spectator mode. There is nothing I can do to
save him, short of setting off the fire alarm. I have never liked him, I do not
know him well, and I do not identify with him. I do recognize him as a fellow
student, however, and as a fellow human being, so I am not totally detached.
The next student is called up to the board to help the first one. She performs
brilliantly. As the instructor praises her I think I see an angry flash of jealousy
light up the eyes of the man with the hangover. I am not jealous of her; I could
do as well as she is doing, and I know the instructor knows this and thinks well
of me. But in my capacity for empathy I feel a flash of the hung-over student’s
anger. How could she show him up so blatantly? I would be angry in his place,
on my own behalf. As it is, I am a little angry on his behalf. He is after all, one
of us, and she has made him look bad. Again, I am not totally detached, but I
am experiencing jealousy as a spectator emotion. If such responses were not
possible, theater would be impoverished.
I conclude that spectator emotions are not a distinct species, although the
more personal emotions tend to resist spectator mode and occur only in watch-
ers whose detachment is not complete.

The Action Problem

The action I feel like taking, as a result of emotion, needs to be specific. In
a full-fledged case of fear, I feel like running away from that particular bear
that is advancing toward me looking as if she has just caught sight of a long
deferred luncheon. Fear tells me which way to run. Panic might make me run
into the bear’s jaws, but panic is not fear; indeed, it is not an emotion at all on
70 on emotions

my account. Running into the bear’s maw would not be an action, because no
one in a panic does that sort of thing by choice, and an action is an event that
proceeds from the choice of an agent. If panic defeats your ability to choose, it
defeats your ability to act as well.
The action does not have to be doable in the circumstances; this goes for the
most engaged emotions such as fear for one’s own life and therefore, a fortiori,
goes for spectator emotions as well. In fear, I feel like running away even when
I am trapped at cliff ’s edge and cannot do so. In regret, I feel like reliving and
repairing an episode in my life, even though I know that I cannot turn back
the clock. In anger at the other side, during a game, I may feel like running out
on the field and hobbling the enemy quarterback, even though I know this is
forbidden.
The action requirement, then, does not discriminate between spectator
emotions and the more personal or directly engaged ones. Both modes of expe-
rience allow for our feeling like taking an action even when circumstances pre-
vent us from taking that action.

The Engagement Problem

Because your emotions are subjective, and because they make you feel like
taking action, you must be engaged when you experience an emotion. A feel-
ing that does not really engage you—a feeling from which you feel totally
detached—could not count as a genuine emotion. But so-called spectator
emotions by definition occur to people who are in spectator mode, and who
therefore are detached in a crucial sense from what they are watching. In the
examples I have given, such as that of Claudius, a failure of detachment moves
the experience out of the realm of spectator emotions. When Claudius loses
his detachment—when the show he is watching becomes personal—then he
ceases to be a spectator.
I must solve the problem of engagement if I am to preserve the intuition,
shared by Solomon and me, that emotions are judgments we have to be able to
make if we are to understand our world. Suppose you are able to watch the death
scene in the Laramie Project without emotion.16 Solomon and I would agree (I
think) that in this case you do not understand the terrible thing that you see,
because you are unable to make the judgments that understanding requires.
Nussbaum’s view is similar: “if the emotion is not there we are entitled to
say that the judgments themselves are not fully or really there.”17 Nussbaum
also holds a strong version of the engagement requirement, which she calls

16
Kaufman et al. (2001).
17
Nussbaum (2001, 44).
 Spectator Emotions 71

“eudaimonism”: “Emotions . . . view the world from the point of view of my own
scheme of goals and projects, the things to which I attach value in a concep-
tion of what it is for me to live well.”18 She recognizes that this requirement
raises a problem for emotions such as wonder, love, and compassion—what
we might call spectator emotions. She accommodates these through a gener-
ous understanding of what counts as a person’s own scheme of goals.19 When
I experience emotions on behalf of other people, on her view, I must be seeing
those people as connected with my own goals. That is why the people on whose
behalf we experience emotions must be either related to us or seen as similar
to us. Nussbaum’s theory requires such openness; otherwise, the theory would
be unable to allow that we can have judgments in the full sense from a specta-
tor position.
If you cannot have emotions about events you merely watch, then you will
be limited to a second-class kind of understanding of these events, whether you
are watching a real traffic accident, a recreation of Matthew Shepard’s death in
1998, or Ophelia’s totally fictional mad scene. If you are really able to have a
first-class understanding of these events, then (on Solomon’s theory of under-
standing and on my theory as well) you must be able to experience genuine
emotions while in spectator mode.
So how to solve the problem of engagement? I have already suggested the
solution in defining spectator emotions. They do not call for detachment, and
they do not even call for less intense degrees of engagement. The mark of spec-
tator emotions is that they engage the subject less directly (not less intensely)
than nonspectator emotions. By this I mean that the intentional objects of
spectator emotions generally do not include the subject.
When I watch the bear approach the child, I feel intense fear for the child—
not for me. If this is not my child and I have no child of my own, I may still feel
intense fear, and the fear is plainly not directly linked to my own interests.
Moreover, because I am far away, I can only feel like rescuing the child. There is
nothing that I can actually do. I know I must watch, and only watch, but I am
not detached. I am as engaged as if the bear were charging at me. But I am not
mixed up about who is in danger. I do not identify with the child. I know the
bear is not charging at me, and I do not fear for myself.
An anomalous case: suppose I am the bear’s intended luncheon, but instead
of fearing for myself directly I see myself in my mind’s eye as if from a great
distance. Then I experience spectator fear and may not be directly motivated to
save myself. The anomaly is that I take myself as what I have called the second,

18
Nussbaum (2001, 49).
19
Nussbaum (2001, 49–56). Spectator emotions plainly show that the eudaimonism of emotions
must be attenuated, as Nussbaum recognizes. A full treatment of this difficulty, and Nussbaum’s
proposed solution to it, would take us beyond the scope of this essay.
72 on emotions

or indirect, object of the emotion. Even such odd spectator emotions are indi-
rectly engaging; that is, they engage us for indirect objects.
There is no engagement problem. Spectator emotions are engaging enough,
just not directly engaging.

The Knowledge Problem

The intentionality of emotions appears to commit us to saying that emotions
entail beliefs that are true or believed to be true. Often we experience emotions
that arise from false beliefs—even, in the case of some fiction, from beliefs
that we entertain while knowing that they are false. False belief is straightfor-
ward when we do not know that it is false, so I will start with that.
Suppose I claim to fear a poisonous snake in the grass, but there is no snake
in this grass. Then my fear has no object in the world, and therefore one of
the four crucial questions must remain unanswered. Such an experience, then,
could not count as a genuine emotion. This is the problem at the center of the
book that prompted Solomon’s thoughts about spectator emotions. But we shall
see that emotions require neither true belief nor knowledge.20
To see why this is a false problem, let’s return to our central example of a
spectator emotion: my horror at watching a terrible traffic accident from the
roadhouse window. Suppose that the accident was a fake, staged by expert
stunt drivers for a fi lm being shot without my knowledge. I was horrified at
an accident, but there was in fact no accident. The belief behind my horror was
false. Ex hypothesi I feel exactly the same intense horror as I would if I saw a real
accident. I have the same belief, and I make the same judgment with respect to
that belief. Is my horror in this case any less a genuine emotion? To deny that it
is an emotion would be counterintuitive.
Why should my claim to having an emotion be undercut by facts that I can-
not possibly have known? Admittedly, a claim to knowledge is truly undercut
by facts that are not available to the putative knower. But an emotion is not a
kind of knowledge; it is a kind of judgment. And judgment, as we have seen, is
neutral with respect to truth. My judgment that there is a poisonous snake in
the grass may be false on either of two points: there is no snake in the grass;
or there is a snake but it is harmless. But a false judgment may still be well
grounded, so a third way I could have gone wrong is by grounding my judgment
poorly on the evidence available to me. Nevertheless, I have still made a judg-
ment: the judgment that the snake in the grass poses a danger to me. And that
judgment is about something—the same something that is the object of the
emotion I claim to have—the toxic snake I believe to be in the grass. If genuine

20
See Nussbaum (2001, 46, n. 42) on the disconnect between emotion and knowledge.
 Spectator Emotions 73

judgments can have as their objects things that do not exist, then it would
seem that emotions can do so as well.
Judgments are neutral with respect to the ontological status of their objects,
so long as I believe that those objects exist. But what if I do not believe that
they exist? What of events that I believe have been made up, as in novels, plays,
or fi lms? I believe, probably correctly, that Ophelia has no basis in history, yet I
feel compassion for her in her suicidal mania while I watch the play.
Let us distinguish between fiction and mimesis, following Aristotle. Fiction
is making up characters or events (Greek: poiesis); mimesis is presenting char-
acters or events in such a way that they affect us as if they were real.21 A docu-
mentary play, such as the Laramie Project, is entirely historical and contains no
fiction. The artists have arranged documentary material and presented it so
effectively that we in the audience react as if we were watching actual events.
Shakespeare’s history plays are not documentary, but they are at least partly
historically true, albeit mixed with material that the poet has concocted. Prince
Hall is an historical character, for example, Poins is fictional, and Falstaff is in
between.
First, documentary drama: while watching the Laramie Project, I believe (cor-
rectly) that Matthew Shepard was murdered. Owing to mimesis, I experience an
emotion that appears to entail the judgment that he is dying at the time I watch
the play, although I know that this is false; he was murdered horribly in 1998
and this is 2011. How can I make a judgment that I know to be false? This is not a
problem about fiction, since it arises for history. This problem is about mimesis.
We need to be clear about the judgment entailed by my experiencing horror
at this murder. The judgment should be parallel to the other two judgments
we have accounted for: “this accident is ghastly” (in the case of the faked acci-
dent); and “this snake is dangerous” (in the case of the imagined snake). Both
were about nonexistent objects. The present emotion is about a real object—
the death of a young man—which is not actually present at the time: “the mur-
der of Matthew Shepard is atrocious.” The effect of mimesis is to allow us to
respond to this event as if it were present; such a mimetic effect is similar to the
impact of a powerful memory, as when Achilles responds to Priam as if Priam
were his own father, reminded by the old man’s presence of the reality of his
own father (Iliad, 24). Such reminders are common outside of the formal exer-
cise of mimesis: Achilles knows Priam is not his father, but he feels compassion
for his own father and therefore for Priam. If memory does the same thing as
mimesis, then we have no special puzzle in the case of documentary drama.
Objects of emotion do not need to be present to be occasions for emotion.

21
The distinction is clear in Aristotle Poetics, Chapter 9. See Woodruff (2008, 137–140)
and Woodruff (1992). Apologies to Walton (1990), who takes fiction entirely into the realm of
mimesis.
74 on emotions

Now for Ophelia: our judgment in her case is that “this girl is suffering.” We
would make the same judgment, and feel the same, if she were our neighbors’
daughter. As in the earlier cases, this judgment is neutral with respect to the
ontological status of the girl. Because judgments are not ontologically commit-
ting, for the reasons we have stated earlier, we do not judge “this girl is suffer-
ing” while also believing that this judgment is false. “This girl is suffering” is
compatible with a range of ontological theories that could account for our judg-
ments of nonexistent objects. I conclude that the intentionality of emotions is
neutral with respect to the ontological status of the objects of emotion. I offer
no theory here for nonexistent objects of fiction and imagination. Such things
must be substantial enough to serve as objects for judgments and emotions,
even though they could not be objects of true belief. I know from experience
that nonexistent objects are part of our emotional experience; let other think-
ers address the ontological issues.
One question remains: does false belief compromise the quality of the judg-
ment involved? Consider what you would think of my judgment if, on see-
ing the accident described, I laughed. I would be showing terrible judgment,
whether the accident was real or compellingly faked. If the accident were
faked, I would be having exactly the same visual experience as I would if it
were real, and my beliefs would be the same (on this thought experiment). To
laugh at something horrifying shows bad judgment whether the something
horrifying is real or not.
Consider Ophelia’s mad scene. Audiences are often horrified by this, as well
they should be. She is exhibiting suicidal mania. To laugh at such a scene is
to show appalling judgment. In the same way, if I observed a friend’s young
daughter in a similar state and laughed, I would be showing bad judgment. Part
of the art of theater is to help audiences to arrive at good judgments of what
they have seen. The art of theater may show actual events, or it may use mime-
sis to invite its audience to have emotions just as intense as those triggered by
actual events.
Because the intentionality of emotions is neutral with respect to the onto-
logical status of intentional objects, there is no knowledge problem and no
solution is necessary. Solomon was right: attention to detailed examples of
spectator emotions leads to dissolving the problem Boruah was writing about.
Twenty tears ago I wrote Bob Solomon a letter promising a discussion of
spectator emotions. Now, finally, I have kept my word.

Acknowledgment
I dedicate this essay lovingly to Robert Solomon, who taught me most of what I
know about the philosophy of emotions. More than that, he gave me the courage
 Spectator Emotions 75

to take on issues outside the usual run of Anglophone academic philosophy. He

supported my work through many discussions of issues on which we disagreed.
He would not have swallowed all of this paper, I expect, but we would have had
a fine conversation about it.
I am also grateful to John Deigh for his comments on a draft of this paper,
which is now substantially different from the one presented at the conference.
I am fortunate to have been able to discuss this with someone whose under-
standing of this subject is deeper than my own.

Works Cited
Boruah, Bijoy H. 1998. Fiction and Emotion: A Study in Aesthetics and the Philosophy of Mind.
Oxford: Clarendon Press.
Deigh, John. 1996. “Love, Guilt, and the Sense of Justice.” In John Deigh, ed., The Sources of
Moral Agency. Cambridge: Cambridge University Press, 39–64.
Carroll, Noël 1990. The Philosophy of Horror. New York: Routledge.
Kaufman, Moises, and members of the Tectonic Theater Project. 2001. The Laramie Project.
New York: Vintage.
Meineck , Peter, and Woodruff, Paul. 2007. Sophocles: Four Tragedies: Ajax, Women of Trachis,
Electra, and Philoctetes. Indianapolis: Hackett Publishing Company.
Nussbaum, Martha C. 2001. Upheavals of Thought: the Intelligence of Emotions. Cambridge:
Cambridge University Press.
Solomon, Robert C. 1977. The Passions: The Myth and Nature of Human Emotion. New York:
Anchor Books.
Solomon, Robert C. 1990. Review of Boruah 1988. Review of Metaphysics 43, 620–621.
Walton, Kendall. 1990. Mimesis as Make-Believe. Cambridge, MA : Harvard University Press.
Woodruff, Paul. 1992. “Aristotle on Mimesis.” In Amélie Rorty, ed., Essays on Aristotle’s
Poetics. Princeton, NJ: Princeton University Press, 73–90.
Woodruff, Paul. 1997. “The Paradox of Comedy.” Philosophical Topics 25, 319–335.
Woodruff, Paul. 2008. The Necessity of Theater: the Art of Watching and Being Watched. New
York: Oxford University Press.
Woodruff, Paul. 2011. The Ajax Dilemma: Justice Fairness, and Rewards. New York: Oxford
University Press.
 4

Comic Amusement, Emotion, and

Cognition
Noël Carroll

Introduction

An Irishman, named Pat, walks into a bar in New York and orders three
shots of Jameson’s. He drinks them down and orders another three
shots. And so on. Finally, the bartender asks Pat why he always orders
his drinks in sets of three. Pat tells him that he likes to make-believe
he’s drinking with his two brothers who are far away. Presently, Pat
says, his older brother is in Dublin and the younger one is in Sydney.
Well, in no time, Pat becomes regular at the bar and, every day, as soon
as he enters the bartender lines up three shots. But on one day, as Pat
bellies up to the bar, he says “Only two shots today.” “I’m sorry for
your loss” says the bartender. “What loss?” asks Pat. “Well, you’re only
ordering two shots. Has one of your brothers passed?” “Naw” says Pat
“Don’t ya know it’s me; I’m on the wagon.”

For those of you who found this joke funny, let us call the state in which you
found yourself “comic amusement” where the object of comic amusement is
humor. In this essay, I’d like to defend the thesis that comic amusement is an
emotional state while also illuminating the relevance of cognition to the mobi-
lization of that state.
On some conceptions of comic amusement, the question of whether or not
comic amusement is an emotional state is a no-brainer. The theory I have in
mind especially here is perhaps the most ancient one. It is often called the supe-
riority theory. Forerunners of this view can be found in Plato and Aristotle. In
his Philebus, Plato identifies comic amusement as a form of malice we direct at
those who fail, as Socrates might have it, to know themselves—that is, those

76
 Comic Amusement, Emotion, and Cognition 77

who think they are smarter, stronger, more agile, better looking, and so on
than they are. Aristotle repeats the theme of malice in his Poetics where he sug-
gests that comedy began as invective, perhaps as Greek versions of the African
American rituals of insult known as The Toast, The Dozens, and Yo Mama (e.g.,
“Yo mama’s so fat, she’s got two zip codes”).1 Of course, the best-known version
of the superiority theory belongs to Hobbes, who identifies comic amusement
as “sudden glory,” which is “the passion which makes those grimaces called
laughter; and is caused either by some sudden act of their own, that pleases
them, or by the apprehension of some deformed thing in another, by compari-
son whereof they suddenly applaud themselves.”2 Here comic amusement is
readily identifiable as a specific emotion, namely, contempt. On this theory,
with respect to the preceding joke, we laugh at Pat because he is either incred-
ibly stupid or stupendously self-deceived.
Nevertheless, for reasons I will address shortly, the superiority the-
ory is problematic in a number of ways. A more promising view is what is
called the incongruity theory. On this view, comic amusement is the enjoy-
ment of absurdity; we laugh along with the folks in Hyde Park at their yearly
Latke-versus-Hammentaschen debate because of the sheer incongruity of the
prospect of world-class scholars, like Martha Nussbaum, treating this topic
with the seriousness they bring to their scholarly research. Likewise, when we
laugh at Pat in the preceding joke, it is the contradiction that tickles our fancy.
We need not look down at Pat; some of us may even hold that there is some low
cunning in his behavior.
Nevertheless, whereas it is clear that comic amusement is an emotional
state on the superiority theory, it is less certain that it belongs in that category
on the incongruity theory. My aim is to show that comic amusement is an emo-
tional state even on the incongruity theory. In order to do that, I will first have
to spell out and defend the version of incongruity theory that I have in mind.
And then I will have to establish its claim to being an emotional state against
various criticisms of that view.

The Incongruity Theory of Comic Amusement

The incongruity theory of comic amusement arose as an explicit reaction to the
superiority theory. Specifically, it was Francis Hutcheson’s response to Hobbes’s
account of “sudden glory.” Hutcheson pointed out that the feeling of superior-
ity could hardly be a sufficient condition for the relevant sort of amusement,
since, despite our superiority to oysters, we never laugh at them.

1
Aristotle, Poetics (London: Penguin, 1996), p. 7.
2
Thomas Hobbes, Leviathan, Part I, chapter 6.
78 on emotions

Nor is superiority a necessary condition for comic amusement. We may

be pleasantly amused by the sight of uncommon and glaring juxtapositions
in everyday life as when we observe a Morris Minor parked next to a stretch
Hummer, but there is no one here available for us to lord it over. Also, we are
made merry by the word wit parlayed by the likes of an Oscar Wilde, but we
recognize that folks like him are far cleverer than we are. Defenders of the
superiority theory at this point often claim that when we laugh at the bon mots
of someone like Wilde, we are in effect applauding ourselves for being intel-
ligent enough to appreciate his wit. However, pace the superiority theory, this
explanation leaves unexplained the fact that we are also acknowledging our
inferiority to the wit who crafted the witticism.
Similarly, when in the movie Our Hospitality the Buster–Keaton-character
uses the rope that shackles him in order to swing to the rescue of his beloved as
she is about to go over the crest of a waterfall, our amusement is not based upon
feeling superior to Keaton. Rather, we marvel while witnessing this astounding
display of ingenuity and dexterity in his lightening reversal of the dire situa-
tion. His performance mixes an understanding of the physics of things with the
athleticism of an Olympian gymnast in ways that few can match, but still the
audience as a whole is convulsed with the laughter of comic amusement. And
finally we often laugh at ourselves when we find ourselves in the very process of
doing something foolish, like brushing our teeth with the shaving cream.
In fact, the superiority theory ill suits the laughter with which we respond
to verbal jokes insofar as most verbal jokes are practical jokes played at the
expense of the listener. Th at is, the typical joke ends with a punch line that
itself is so puzzling that it prompts the listener to come up with an inter-
pretation that renders said punch line ostensibly intelligible—for example,
a preacher asks a thief, “Why do you rob banks?” to which the robber replies,
“Because that’s where the money is.” But note that a trick has been played
upon the listener here, since one initially assumes the preacher is asking why
the thief is embarked upon a life of crime and not why he robs banks rather
than bakeries. The jokester misleads us up a verbal garden path, and then we
fall for the ruse. Jokesters usually make their audiences butts or, at least,
straight men of sorts in this way. Nevertheless, we laugh precisely at the
moment when we are at the mercy of the joke teller who has just outdone us.
According to the Hobbesian superiority theorist, we should not be chortling
when we fi nd ourselves outmatched in this fashion. But we are laughing just
the same.
Moreover, inasmuch as Hobbes’s theory is framed in terms of laughter, it is
predictably too broad to provide us with a serviceable theory of comic amuse-
ment, since laughter accompanies many occasions that are anything but comic.
The gladiator, with his foot upon the neck of his defeated rival, laughs in tri-
umph, but not because the situation is funny. Triumphant laughter is different
 Comic Amusement, Emotion, and Cognition 79

than the sort associated with comic amusement. Hobbes’s mistake was to
attempt to assimilate the two.
A theory of laughter, if there could be such a thing, cannot afford us a the-
ory of comic amusement. For in addition to triumphant laughter, there is also
the laughter that arises from tickling, from various drugs, from a sense of
well-being or joy, from nervousness, from recognition, and from lovemaking.
Uncontrollable laughter can be an affl iction, as in the case of hebephrenia. In
fact, recent scientific research has established that most of the laughter in ordi-
nary discourse functions as conversational glue—we laugh, just as we nod our
heads, in order to let our interlocutors know that we are listening to and fol-
lowing them.3 The topic being discussed need not be witty. To confirm this take
heed of all those small bursts of laughter that the interviewers weave into their
remarks on National Public Radio.
In short, the objects of laughter are various and most of them are not humor-
ous. The laughter that we care about today is directed at humor. Yet what is the
object of comic amusement? In response to Hobbes, Hutcheson identified it as
incongruity, or, perhaps better, perceived incongruity. Maybe the germ of this
proposal was already present in the classical superiority theories of the ancients
who thought that the proper objects of comedy were people who are worse
than average and, therefore perforce, deviations from the norm. However, this
suggestion is too narrow, since, as we have seen in cases like Oscar Wilde and
Buster Keaton, comedy may issue from excellence. What is key to comic amuse-
ment is a deviation from some presupposed norm—that is to say, an anomaly
or an incongruity relative to some framework governing the ways in which we
think the world is or should be. That is, the crux of comic amusement is the
failure of its object to be congruent or accordant with the ways we believe the
world is or should be.4
Sometimes this idea is stated in terms of a subversion of expectation. Yet,
this may be misleading. For, often we are comically amused by outcomes we
anticipate with mounting mirth—as when the policeman in a slapstick com-
edy, distracted by a passing beauty, walks straight into an open manhole.
Furthermore, when listening to a joke we generally do not have any inkling—
that is, any specific expectations—about where it is headed. When the punch
line is delivered, it is not as though it displaces some other specific thought that
we had in mind. We are surprised, if the joke is effective—although as we’ve

3
R. Provine, Laughter: A Scientific Study (New York: Viking, 2000).
4
Robert Latta argues that all incongruity theories of comic amusement fail because they do
not specify that which the object of comic amusement is incongruous or discordant. My candidate
is that humor, the object of comic amusement, is incongruous or disagrees with how we think the
world is or should be. See Robert Latta, The Basic Humor Process: A Cognitive Shift Theory and the
Case against Incongruity (Berlin: Mouton de Gruyter, 1999).
80 on emotions

seen from our previous example (of the slapstick policeman) our surprise is not
necessary for comic amusement.
If we want to employ the language of expectations with respect to comic
amusement, then we should not be thinking of specific expectations, like
“what exactly will Pat in the opening joke respond to the bartender.” Rather,
we should be thinking of our global expectations about how the world is or
should be. Comic amusement emerges against a background of presumed con-
gruities or norms. Moreover, because we assume so many congruities or norms
in order to navigate our way through the world, there are an indeterminately
large number of things that are potentially perceivable as incongruous.
Schopenhauer, for example, thought of the object of comic amusement
along the lines of a category mistake or an absurdity as in the case of Pat, the
Irishman, who understands the concept of “being on the wagon” to be consist-
ent with downing two drinks “for his brothers,” so long as he refrains from
tossing back one “for himself.” Likewise, the incongruity is a logical error, when
the rotund customer in the pizzeria who when asked whether he wants his pie
sliced into eight pieces or four, responds, “Four; I’m on a diet.” Because they
deviate from and disrupt norms, violations of logic, deductive and inductive,
formal and informal are incongruities and, therefore, staples of comedy.
But we abide by many other laws than those of logic and, consequently, the
ambit of humor is much broader than Schopenhauer’s very elegant theory sug-
gests.5 Comic amusement can erupt wherever sense is problematized—wher-
ever our sense of how the world is or should be is disturbed. We operate with
norms of morality, prudence, and etiquette. Hence it is unsurprising that much
humor involves immorality, reckless and extravagant behavior, and gaucherie.
In addition to logical, moral, and prudential norms, there are many conven-
tional norms, including not only the codes of politeness but the laws of language.
Violations of the latter provide the source of a great deal of humor; Groucho
Marx’s puns frequently derail in a single blow the rules of semantics and gram-
mar as well as the maxims of conversation.6 And, speaking even more broadly,
norms of appropriateness govern almost every aspect of our lives, opening up,
thereby, the possibility of humor with respect to sexual behavior, cleanliness,
attire, and much else. We presuppose norms regarding human intelligence

5
Violations of physical categories can also function as a source for comic amusement. Examples
of this include many of the incongruously delightful mixtures of features of the Muggle World
and the Wizard World in the Harry Potter series. For example, in Harry Potter and the Chamber
of Secrets we are told that when the mandrakes get over their acne, they’ll be ready for action,
thereby perpetrating a category mistake that yokes the vegetable with the human.
6
An example of the way in which the violation of a conversational maxim can provoke comic
amusement is the well-known, albeit sophomoric, witticism: Question—“Do you know what time
it is?”; answer: “Yes” (which answer transgresses the Gricean maxim of quantity). See Salvatorre
Attardo, Linguistic Theories Humor (Berlin: Mouton de Gruyter, 1994), pp. 272–292.
 Comic Amusement, Emotion, and Cognition 81

and physical condition.7 Hence, it is no accident that so many clowns are inhu-
manly stupid and exceedingly fat or skinny. These features and more can func-
tion as topics for comic amusement inasmuch as they afford opportunities for
incongruity, including, for instance, portly men like Fatty Arbuckle and Benny
Hill wearing ladies’ apparel or Ernie Kovacs’ dancers performing Swan Lake in
gorilla costumes.
Nevertheless, the very fact that incongruity covers such an indefinitely
large territory may itself pose a problem. It is a very baggy concept; perhaps
it is too loose to perform the work that incongruity theorists intend for it. In
the nineteenth century, Alexander Bain, for example, argued that incongruity
could not supply a sufficient condition for comic amusement, since there are
many incongruities that engender other sorts of feelings.8 Maybe, most pat-
ently, instead of levity, incongruities are just as likely or even more likely to stir
up fear and anxiety. Bela Lugosi done up in his Dracula outfit may cheer us in
Abbott and Costello Meet Frankenstein, but, in other contexts, he is more apt to
engender horror. Similarly, psychologists have observed that when a familiar
caregiver puts on a “funny face,” a child is likely to be amused and even giggle,
whereas when a stranger does the same thing the child becomes anxious and
may wail. So, even if incongruity were a necessary condition for comic amuse-
ment, it is not sufficient.9
One factor that needs to be added to incongruity, then, in order to approach
adequacy is the recognition that for comic amusement to take off, it must occur
in a context from which fear for ourselves and those we care about—including
fictional characters—has been banished. Comic incongruities, in other words,
must be nonthreatening, or, at least, what is potentially threatening, fright-
ening, or anxiety producing about them must be deflected or marginalized.
When someone is killed in a joke—as so many lawyers are—we are not treated
to the gruesome details of their demise. As Aristotle notes, comedy should not
involve pain or destruction,10 or, perhaps more accurately, it should not draw
focused attention upon it. The suffering is kept offstage, so to speak. Also, the

7
Th is would include what might be thought of as developmental incongruities. For exam-
ple, there is a room in the Frick Museum in New York where panels by François Boucher show
very young children at work in very adult roles (for instance, the panel The Arts and the Sciences:
Chemistry shows a veritable infant at his cauldron cooking up some formula). The joke here is the
incongruous mismatch between the child and the adult occupation. Nor is this variety of humor a
thing of the past. Presently on TV, there is a commercial encouraging personal stock trading that
presents an infant—whose mouth has been digitally jigsawed to form adult words—recommend-
ing that the process is so easy that a child can do it.
8
Alexander Bain, The Emotions and the Will (London: Longmans and Green, 1975).
9
Different types of incongruity are explored in John Morreall, “Funny Ha-Ha, Funny Strange
and Other Reactions to Incongruity,” in The Philosophy of Laughter and Humor (Albany: State
University New York Press, 1987).
10
Aristotle, Poetics, trans. Malcolm Heath (London: Penguin, 1996), p. 9.
82 on emotions

victims in jokes and the like are typically people we don’t care about and maybe
even people we are encouraged to dislike—and even to imagine to be deserving
of whatever they get (think lawyers, again).
Of course, those whom we care about can come in for a pummeling in humor,
such as the figures of slapstick, like Bob Solomon’s beloved Three Stooges,11 but
it is significant that these figures are usually clowns, beings, that is, who are not
quite human—creatures who can take a hit on the head with sledgehammers
and who, then, after a brief swoon, are back in the game, almost immediately.
They are not, in short, in real danger; so we do not feel anxiety on their behalf.
Bergson referred to this tolerance for comic brutality by means of the mem-
orable phrase “the momentary anaesthesis of the heart.” But I don’t think that
this should be understood to mean that comic amusement is alien to emo-
tion altogether but rather only that certain emotions—such as sympathy—
are disengaged, either by distracting our attention away from that which
might enlist our sympathies for the characters in question by deemphasizing
the apparent degree of danger and pain that threaten them or by portraying
them as anti-pathetic or by portraying them as other-than-normal-humans or
clown-like or moronic and not subject to the injuries to which flesh is heir.
Often the anaesthesis of sympathy can be secured by features internal to
the joke or story, as in the case of the clown or through the mobilization of ster-
eotypical characters we love to hate (again, think of lawyers).
Likewise, a fantastical setup can cue humor—as in the case of a cartoon set
in hell or one in which there are talking insects.
Other internal factors that neutralize our sympathies here involve the way
in which anything that might cause anxiety in the audience is marginalized
by either not being mentioned at all or by rushing past them and not dwelling
upon features like pain and suffering that might provoke anxiety.
But external factors to the joke or story world can also play a role. When
framed with the introduction “Did you hear the one about x” or when a change
in the intonation of the speaker shifts key or maybe only by a wink or a devilish
smile, we are alerted to the proposition that now is the time to adopt what is
called comic distance. That is, characters are about to be beaten, blown apart,
defenestrated, and so on, but we should not worry about them. In other words,
these framing devices tell us that the imaginary beings in jokes and other comic
forms are not quite like us ontologically, and, therefore, what happens to them
should not be a matter of our concern.
One apparent counterexample to my conjecture that humorous incongrui-
ties be divested of elements that would recruit our anxieties for the victims of

11
See Robert Solomon and Jon Solomon, “Are the Th ree Stooges Funny? Cointainly! (or When
Is It Okay to Laugh?),” Paper delivered at the Society for Popular Culture, Toronto, November
1990.
 Comic Amusement, Emotion, and Cognition 83

comic mayhem may appear to be many instances of black humor, such as dead
baby jokes.12 These jokes involve incongruities, if anything does, reveling as
they do in the torture and/or deaths of infants. Who can deny that the mas-
sacre of innocents is inappropriate?
But these jokes can amuse many of us. At least two factors seem to come
into play here. First, like all jokes, these specimens are framed as jokes, thereby
invoking comic distance. But second, and perhaps more importantly, the dead
babies are not really the butts of these jokes. The butt of dead baby jokes, and
of black humor in general, is those sanctimonious people we imagine will be
outraged by the joke. Those are people we are disposed to regard antipatheti-
cally already. Our merriment is grounded in recognizing how these dead baby
riddles will fluster them.
As we have already seen, in general, verbal jokes are on a continuum with
practical jokes. There is an element of trickery about them. The jokester, that
is, tricks the listener of the joke into finding the absurd conclusion of the joke
to be somehow compellingly intelligible, when it is not. With respect to black
humor, a trick is also being played upon the uptight audience whom we imag-
ine will become apoplectically indignant upon hearing such a joke. As André
Breton, the discoverer of black humor, notes, black humor “is the mortal enemy
of sentimentality. . . . ”13 That is, black humor is a satire of conventional pieties.
It is another way of outraging the bourgeoisie. Or, as my friend Joan Acocella
likes to say, “It’s a way of driving your mother insane.”
In addition to not producing anxiety, the relevant incongruities should not
be annoying. If you are the kind of person who fetishizes the rules of etiquette,
you will find disturbing the presence of the salad fork on the wrong side of the
plate. Others, however, are more likely to find such mistakes mildly amusing,
and some wags may even go out of their way to perpetrate such incongruities.
Of course, another way in which the incongruities that spark comic amuse-
ment differ from the anxiety-producing sort and the annoying sort is that we
enjoy them. We do not fret about them. They give us pleasure. Nevertheless,
acknowledging this is still not enough to fend off Bain’s objections. To say comic
incongruities are non-threatening is insufficient, since there are nonthreaten-
ing, enjoyable incongruities that are not comic. Here I am thinking particularly
in terms of the pleasure one derives from grappling with puzzles and anomalies,
whether theoretical, empirical, or practical. It may appear strange to attempt
to distinguish sharply between the pleasures experienced when confronting
puzzles and problems, on the one hand, the comic amusement that results from
humor, since a large portion of comic amusement is inspired by jokes, whether
of the riddle variety or the narrative kind, and these typically involve puzzles,

12
A typical example of a dead baby joke: “What is brown and gurgles? A baby in a casserole.”
13
André Breton, Anthology of Black Humor (San Francisco: City Lights Press, 1997), p. xix.
84 on emotions

otherwise known as punch lines whose anomalousness gives birth to interpre-

tations that are ostensibly designed to dispel their incongruity.
For example:

The lone survivor of an airplane crash is marooned on a deserted island

in the middle of the Pacific Ocean. After many years, he is rescued by
a passing ocean liner. As the doctor is examining him, he says, “You’re
in great health but tell me one thing. Why are there two synagogues
on the island?” The survivor answers, “The one on the north side of the
island is my synagogue. The other one I wouldn’t step into.”

Here the punch line explains the puzzle of why there are two synagogues
on an island with one inhabitant, but it does so at the cost of compounding the
absurdity rather than resolving it. For the joke invites us to imagine—against
anything that the principle of charity would recommend—a man so improb-
able that he would build a structure for the sole purpose of not entering it. In
this joke, an answer is supplied to the puzzle, yet the answer itself is an absur-
dity or an incongruity that nevertheless jollies us.
In contrast, when we are engaged in genuine puzzle solving, our pleasure
revolves around finding or trying to find an actual answer to our question—
one that accords with how the world is or should be. We’re looking for congru-
ity. On the other hand, with respect to jokes, one puzzle or incongruity gives
way to another, and we leave it at that. The result of a joke is a derangement of
sense. When we engage in authentic puzzle solving—from crossword puzzles to
mathematical theorems—we aim at finding the right answers and take pleas-
ure in that, whereas with jokes, we are happy with the wrong answers, indeed,
with outrageously wrong answers.
Gathering together these observations, then, let us say provisionally that
someone is comically amused if and only if (1) the object of her mental state
is a perceived incongruity (2) which she regards as neither seriously threaten-
ing to herself nor anyone she cares about, nor does she regard it as otherwise
anxiety producing, (3) which she does not find annoying, (4) which she does not
approach with a genuine, puzzle-solving attitude, but (5) which she enjoys.14

14
“Provisionally” here is meant to signal that a further condition will be added in the next
section, one that hinges on a notion of levity (to be defi ned).
One worry about this theory is that it may appear to make comic amusement seem too punc-
tal—like the response to a joke. But, it may be objected, comic amusement can be sustained over
a long period of time, for example, throughout the evening’s performance of a comedy. Th at is
true, but it is not a problem for this theory. For, in the fi rst place, an evening of comic amusement
may be sustained because a series of punctal comic incidents (e.g., one-off jokes, gags, bon mots)
succeed in inducing a sustained comic mood that suff uses the evening. And second, because there
are long-term comic structures such as running gags and the equivocal plot—as discussed in my
article “Two Comic Plots” in my Art in Three Dimensions (Oxford: Oxford University Press, 2010).
 Comic Amusement, Emotion, and Cognition 85

Humor, the perceived incongruity, then is the response-dependent property

that affords comic amusement.
This sort of theory has, of course, been advanced previously and has been
challenged by certain recurring types of criticism. So before turning to the dis-
cussion of comic amusement as an emotional state, let me attempt briefly to
dispose of those objections.
The first, which we owe to Roger Scruton, is that perceived incongruity is not
a necessary condition for comic amusement. Consider the case of caricature.
A caricature is supposed to bring about some insight about the person carica-
tured. For example, a political caricature might portray former vice president
Dick Cheney dressed in Darth Vader’s regalia in order to suggest that he’s “gone
over to the dark side.” Caricatures of Richard Nixon exaggerated his five o’clock
shadow to underscore his thuggishness. Supposing that these caricatures
reveal something true about their subjects—that is, something corresponding
to the way things are—you might be tempted to say that what we take pleasure
in here are congruities, rather than incongruities.
A second example, in a similar vein, observes that we may be comically
amused when someone acts in a way that is hypertrophically in character—for
instance, a miser who is so cheap that, despite his great wealth, walks fifteen
miles rather than pays $1.50 for public transportation. Again, the point here is
that this behavior is extremely congruous with the miser’s character, a truth,
so to speak, about his very nature. Therefore, incongruity is not a necessary
condition for comic amusement.15
I am very suspicious about these counterexamples. I think that ultimately
they rely upon playing fast and loose with our operative understandings of
“incongruity” and “congruity.” That is, equivocation is afoot here. In the case of
caricature, the alleged “congruity” is “truth with respect to the subject’s char-
acter.” Yet this sort of congruity, if that’s what you want to call it, is perfectly
compatible with the sort of incongruity that is relevant to visual caricature—
specifically, deviation from the appearance of the subject, most often in terms
of exaggeration (as is apparent in terms of the outsized ears in caricatures of
Lyndon B. Johnson, George W. Bush, and Barack Obama).16 One doubts that
there would be comic amusement without these perceived incongruities, since
revealing, realistic self-portraits, like those of Rembrandt, do not evoke comic
amusement. Why not? There is nothing visually incongruous about them.
A similar strategy may help us dissolve examples like that of the miser.
The miser’s behavior is said to be revelatory of his true character. Yet this is

15
The preceding two lines of criticism come from Roger Scruton, “Laughter,” in The Philosophy
of Laughter and Humor, ed. John Morreall (Albany: State University New York Press, 1987).
16
Of course, the incongruity need not always be an exaggeration; it may just be something
spectacularly unlikely, like Vice President Cheney done up as Darth Vader.
86 on emotions

consistent with the behavior being incongruous due to its exaggerated thrifti-
ness. It goes way beyond any reasonable norms of prudence. The behavior is
hyperbolic, and hyperbole is one form of incongruity.
Another set of counterexamples is intended to show that incongruity is not
sufficient for comic amusement. Both charge that incongruity theories fail to
differentiate comic amusement from certain of the kinds of pleasure that we
may take from engaging with artworks.
One of the pleasures to be had from watching and thinking about Orson
Welles’s fi lm Citizen Kane rests upon trying to interpret the central contra-
diction in the story: is a human life, like Kane’s, unfathomable or can it be
explained by clues as pregnant as “Rosebud.” Isn’t a large measure of the pleas-
ure we derive from Citizen Kane a function of the hermeneutic play in which we
immerse ourselves when tracking the meaning of the narrative? But even if we
call this “amusement,” it is not comic amusement. So the incongruity theory of
comic amusement is too broad.17
Also, many Surrealist images, such Dali’s melting timepieces, intrigue us by
means of their incongruity. Yet these paintings do not prompt comic amuse-
ment. They are often far too ominous. So, once again, the incongruity theory is
not tight enough.18
Undoubtedly, these objections have a great deal of force against simple
incongruity theories of comic amusement—that is to say theories that sim-
ply define the state of comic amusement as a response to perceived incongru-
ity. Nevertheless, the version of the incongruity theory that I am defending
possesses resources capable of containing these objections. The Surrealist
incongruities are intended to be unsettling. Unlike jokes, they do not even
counterfeit a patina of intelligibility. They defy intelligible explanations, nor
do they support even faux intelligible explanations. They are designed to dis-
turb—to elicit a haunting sense of enigma or mystery. And, in this regard, they
are excluded from the order of comic amusement on my account, because they
are anxiety producing.19
Responses to artworks that involve interpretive play are, on the other hand,
typically enjoyable. Reflecting upon Citizen Kane and discerning a theme that
may unify its disparate elements can be pleasurable. But, notice that here the
pleasure is connected to our adoption of a genuine, problem-solving attitude;
it is grounded in working out the interpretation. Thus, it is excluded by the
present definition from counting as comic amusement. Hence, even if some

17
Th is line of objection may be found in M. W. Martin, “Humour and the Aesthetic Enjoyment
of Incongruity,” British Journal of Aesthetics, 23, pp. 74–84, 1983.
18
Ibid.
19
The impossible illustrations of Escher may involve both the puzzle-solving response and the
disturbance response. At fi rst, we try to solve the puzzle—to fi nd the point at which the picture
has gone haywire—but then failing to fi nd it it becomes disturbing.
 Comic Amusement, Emotion, and Cognition 87

incongruity theories of comic amusement are vulnerable to the aforesaid

examples, the version I am defending is not.
So, presupposing that comic amusement is the sort of mental state I’ve indi-
cated, let me address the question of whether comic amusement, so construed,
is an emotion.

Emotion and Cognition

As indicated earlier, it is easy to count comic amusement as an emotional state
under the superiority theory, since, from that perspective, comic amusement is
a form of contempt. However, I am defending an incongruity theory. Is comic
amusement an emotional state on that view?
Clearly, there are a number of correlations between comic amusement and
paradigmatic emotional states. First, both are directed. Fear is directed at
things perceived to be threatening; comic amusement is directed at the humor-
ous, which, on the theory under examination, is that which is perceived to be
incongruous. The object that fear is directed at is intentional; it need not exist.
It is something that is perceived as threatening. Likewise, the object of comic
amusement is perceived to be incongruous but need not be. Amusement comes
in degrees, as do the paradigmatic emotions, and it is subject to a degree of
self-control; I can suppress my amusement to a certain extent, just as I can
control anger and fear somewhat. I can be self-deceived about the source of my
enjoyment with respect to a joke, just as I might be self-deceived with respect
to my hatred of a coworker of mine who happens to be of another race.20
Another correspondence between comic amusement and paradigmatic emo-
tional states is that comic amusement would appear to have a formal object,
that is, a criterion of appropriateness that governs the mobilization of the
state. A necessary condition for being in the state of fear is that one perceives
the particular object in question as dangerous or harmful. If I am of sound
mind and body, I cannot be in the state of fear, unless I perceive the object of
that state to be harmful. It may not be dangerous, but I must perceive it to be
so. Likewise, on the superiority theory of comic amusement, the formal object
of my state is that which is inferior to me, while, on the incongruity theory, the
formal object is that which appears to me or that I apprehend as incongruous.
Also, once a paradigmatic emotion is up and running, it cognitively takes
over the situation in which we find ourselves and spotlights features of the con-
text that reinforce the animating affective state. If we are angry with our lover,

20
These comparisons between comic amusement and paradigmatic emotions are drawn
from R. A. Sharpe, “Seven Reasons Why Amusement Is an Emotion,” Journal of Value Inquiry, 9
(1975), pp. 201–203. I have not cited all of Sharpe’s analogies because I don’t fi nd them all to be
convincing.
88 on emotions

the anger will enable us to find more things to be angry about (think of all those
arguments with your lover). Similarly, when we are comically amused, we will
be inclined toward finding more absurdities in the circumstances. One bon mot
elicits others. Remembering one aspect of a silly event or character from the
past, calls forth the remembrance of other silly aspects, either from ourselves
or from other listeners who are privy to the same history. Sometimes emo-
tions can inaugurate mood states—mental attitudes that calibrate perception
and memory to process everything under their aegis. When joyful, for exam-
ple, everything takes on a happy cast. The grumpy old man next door appears
quaint. And so on. Analogously, sustained comic amusement can put us in a
comic mood, one where we perceive something incongruous about everything
that comes our way.
In addition, comic amusement is contagious as are many of the emotions. We
laugh longer and harder when we watch a fi lm like Hangover with an audience
just as we are more prone to scream at a horror fi lm like Aliens when everyone
else screams. That is why laugh tracks were originally attached to television
comedies in the 1950s.
Yet, despite these correspondences, a number of objections have been
advanced against the idea that comic amusement is an emotion and, it seems
fair to say, these objections have been made with the incongruity theory of
comic amusement in mind.21 One objection is that the emotions involve the
alteration of one’s bodily state, but it is not clear that comic amusement does.
This may sound strange. If any state is associated with a change in our bodies,
surely it is that comic amusement is typically marked by explosive laughter. But
although laughter is a frequent correlate of comic amusement, it is not invari-
ant. Sometimes a witticism will tease no more from us than a slight smile or a
mild tug of the lips. But these are still bodily transformations.
And even where these are not in evidence, I suggest there is nevertheless
at least typically in creatures like us, a palpable phenomenological sensation
that accompanies comic amusement, which we may call levity and may describe
phenomenologically as a feeling of lightness, a tendency to quicken and then
relax. This feeling involves a rhythm of tensing or being set aback when ini-
tially confronting an absurdity, but then that sensation gives way to settling
back or relaxing when we realize that no genuine challenge is afoot. (Moreover,
I would argue that this sense of levity be added to the list of necessary and
jointly sufficient for comic amusement advanced in the preceding section).
Another objection to the notion that comic amusement is an emotion is that
the emotions paradigmatically require beliefs in order to ignite, but nobody
believes that there exists anyone as dumb as the moron. And related to the idea

21
In responding to these objections, I have been very influenced by Robert C. Roberts, “Is
Amusement and Emotion?” American Philosophical Quarterly, 25 (July 1988), pp. 269–274.
 Comic Amusement, Emotion, and Cognition 89

that genuine emotions are connected to beliefs is the view that the emotions
are action guiding. They motivate us behave in certain ways. Fear prepares us to
fight, flee, or freeze. However, comic amusement does not motivate us to move
in one direction or another. When comically amused, we enjoy what appears to
us to be a harmless incongruity but feel no pressure to do anything about it.22
These objections both fail in two ways: in their presumptions about the
nature of the emotions in general and in their presumptions about comic
amusement in general. With respect to the emotions, not all emotions require
beliefs. Emotional responses to fictional characters and events need not rest
upon beliefs. I can feel sorrow for Anna Karenina, although I do not believe
that she exists, since I know that she and all the events that befall her where
made up by Leo Tolstoy. Emotions need not be rooted in beliefs, understood
as propositions held in the mind as asserted. Genuine emotions may issue
from imaginings, construed as propositions entertained or held in the mind as
unasserted. Nor are fictions the only examples of emotions born of imagining.
One can certainly engender anxiety in oneself by imagining cutting off your
finger in the meat-slicing machine, even though you do not believe that it is
happening.
Consequently, because genuine emotions can issue from imaginings, the fact
that much humorous incongruity, as in the case with most jokes, is fictional or
made-up provides no reason to dismiss the claim that comic amusement is an
emotional state. Nor would it help matters to speculate that when the paradig-
matic emotions are discharging their paradigmatic function, they are tethered
to beliefs. For certainly the fact that our forebears were capable of being fright-
ened by contrary-to-fact stories about what could happen to them—as opposed
to what they believed was happening to them—performed an advantageous
evolutionary function. It kept many children out of places where they might
otherwise perish without ever having the opportunity to reproduce.
However, the no-belief argument against the suggestion that comic amuse-
ment is not an emotion falters not only because the emotions do not require
beliefs. It also supposes that comic amusement may never rest upon beliefs.
But when watching Monty Python and laughing at John Cleese performing one
of his “funny walks,” I don’t imagine that he is walking incongruously. I believe
that he is walking incongruously. And when I see an outlandish hat and smile,
I believe the hat is outlandish and don’t just think it so.
Furthermore, the fact that our comic amusement can be connected to our
beliefs has ramifications for the claim that comic amusement motivates noth-
ing. Satire is an important source of comic amusement. The San Francisco
Mime Troupe and the Bread and Puppet Theater, in the tradition of Bertolt

22
Objections along these lines can be found in Scruton, “Laughter”; Morreall, “Humor and
Emotion.”
90 on emotions

Brecht, employ satire in order to change our beliefs. They use a trope of comic
incongruity—specifically hyperbole—in order to influence our beliefs about
the establishment in the hope that that will dispose us to certain forms of politi-
cal action, if only electoral action. Indeed, satirizing sexism, racism, and/or
homophobia may alter our beliefs in ways that may change our own behaviors.
Of course, the no-motivation argument also depends upon a faulty general-
ization about the emotions. Not all emotional states incline us toward action.
Especially as an academic, I may feel sorrow that the great library of ancient
Alexandria was destroyed. But that sadness over the loss of the library at
Alexandria does not motivate me to do anything. After all there is nothing I
can do about it but grieve. Nor does this sorrow dispose me toward even wish-
ing that the library had not been destroyed, since I realize that that would have
shifted the course of history in unforeseeable but potentially nasty ways. Thus,
my grief neither prompts me to act nor even to wish I could act. Not all emo-
tional states are motivating. Consequently, even were it the case that comic
amusement never motivates us to action, that would fail to show it is not an
emotional state.
A somewhat more complicated objection to regarding comic amusement as
an emotional state claims that with the emotions, properly so called, when
we find the emotional state in question pleasurable, we simultaneously find
the object of the emotional state to be pleasing. When we are in the state of
requited love, our overall condition is pleasurable and we are pleased by our
lover. When we are frightened of the snake, our state is one of distress and we
are distressed by the snake. But this kind of symmetry can be lacking in cases
of comic amusement. As John Morreall puts it:

Amusement . . . is pleasant not in that the thing which amuses us is

pleasing to us—often amusing objects are unattractive and even
potentially repulsive—but in that our observing it or thinking about
it is a pleasant experience. The gaily painted house with the odd win-
dows, for instance, may strike us as grotesque; what is pleasant is see-
ing it among the other houses with which it obviously doesn’t belong.
What amuses us is the incongruity of the house in this setting.”23

We can call this the symmetry argument. Like the no-belief and the
no-motivation arguments, however, it presupposes a generalization about the
emotions which does not hold across the board. Contempt is an emotion, but
clearly one can enjoy feeling contempt for something—one can be pleased
by looking down at it—without being pleased by the object of our contempt.
Morreall feels amused by the gaudy house in his example. Someone less tolerant

23
Morreall, “Humor and Emotion,” p. 215.
 Comic Amusement, Emotion, and Cognition 91

than Morreall might feel contempt in the same circumstances. The experience
of contempt can be quite gratifying. Many enjoy being members of cliques that
disparage outsiders and their taste. There is no question that contempt is an
emotional state, although there need be no symmetry between the potential
pleasure of the experience of feeling contempt and the aversion or repulsion
felt by those who are contemptuous toward those whom they find contempt-
ible. Consequently, there should be no problem where there is an asymmetry
between the experience of comic amusement and its objects.
Roger Scruton denies that comic amusement can be an emotion on the
grounds that it lacks a formal object. Of course, I have maintained that per-
ceived incongruity, appropriately qualified, is the formal object of comic amuse-
ment. Scruton challenges this for several reasons. First, he maintains that
incongruity is not a necessary condition for comic amusement. We encountered
his arguments in the last section where he alleged that comic amusement often
issues—as in the case of caricature—from congruity rather than incongruity.
And second, he suggests that rather than being a pattern of thought that per-
ceives things in an incongruous light, comic amusement is a devaluing mode
of attending to things, where Scruton suspects we will not be able to turn the
notion of a devaluing pattern of attention into a formal object. Scruton never
shows why this cannot be done. His evidence that comic amusement always
devalues is that no one enjoys being laughed at.
As I hope I have already demonstrated, Scruton’s rejection—on the basis of
alleged comic congruities—of incongruity as the formal object of the relevant
variety of amusement relies upon an equivocation. In the case of graphic cari-
cature, the equivocation is between congruity as correspondence to the nature
of its subject and incongruity as lack of correspondence to the appearance of
its subject. Surely, these are not incompatible. Both may obtain with respect to
the same figures. Moreover, if only the former variety of congruity occurred
without the discrepancy in appearance, there would be no comic amusement.
Think of those photographs of Winston Churchill as a mountain of smoldering
determination. They reveal the resoluteness in his character, but they are not
funny. They are not caricatures, because they are not visually incongruous.
Scruton’s rival to the conjecture that comic amusement involves perceiving
things in an incongruous light is the idea that it involves seeing things in a deval-
uing light. Comic amusement is a form of attentive demolition. Scruton doesn’t
think that this conception will yield a formal object because there are so many
different ways of thinking about things so as to devalue them. I am not sure why
this preludes there being a formal object in the neighborhood, since there are so
many different ways to fear things, to be angered by things, and, for that mat-
ter, to find something incongruous about them. However, my deeper reserva-
tions concerning Scruton’s proposal is that the competitor to incongruity that he
offers us does not fare as well as incongruity when weighed against the data.
92 on emotions

That is, supposing that Scruton’s congruity arguments have not scotched the
incongruity theory from the contest, incongruity is more comprehensive than
Scruton’s candidate. For, as we have seen, sometimes comic amusement accrues
when we encounter something that deviates from the norm but in a way that
excels. When a Buster Keaton manipulates the world with unsurpassed inge-
nuity, grace, and wit, our comic amusement and giddiness is not a matter of
attentive demolition but of attentive admiration. Scruton’s evidence for the
hypothesis of attentive demolition is that no one wants to be laughed at. But
I’m not completely convinced of this. Certainly professional comics want to be
laughed at; they would be out of business otherwise. And, as I recall from my
own experience, class clowns are quite happy to do anything for a laugh. Of
course, Scruton may protest that by being “laughed at,” he just means being
“disparaged or devalued.” But “being laughed at” surely may have the more
innocent meaning of “being the particular object of comic amusement” as in
the case of Buster Keaton performing some ingenious feat. However, in that
event, it may be that Scruton is begging the question. He has not proven that
comic amusement is tantamount to belittling attention but merely that some
instances of “laughing at someone” are derisive. Yet that is hardly enough upon
which to mount a theory.
A more recent kind of objection to the notion that comic amusement is an
emotional state might be that it is too cognitive. It presupposes something
like a cognitive theory of the emotions. On the incongruity theory, the com-
ically amused subject marshals categories, attempts to subsume the object of
amusement under said categories and then enjoys the way or ways in which
it slips out of them. Yet on neo-Jamesian views, of the sort that are gaining
traction nowadays, there seems to be too much thinking going on here. For
the neo-Jamesian, the emotional state is much more like a perception. It is a
noncognitive affective appraisal that then elicits a physiological change. Where
cognition enters the picture is after the original response, when it functions to
monitor and modify the earlier and ongoing stages of our response. Were comic
amusement an emotional state, it would not be as intellectual as the incongru-
ity theory of comic amusement assumes. Consequently, comic amusement, if it
is as analyzed by incongruity theorists, is not an emotional state.
Although I have not encountered this objection in the literature yet, permit
me to launch a preemptive strike. First, even though I accept the scientific evi-
dence that something like the aforesaid scenario describes the most frequent
pattern of emotional response, I do not see why neo-Jamesians are so convinced
that emotional states never take the trajectory hypothesized by earlier cogni-
tive theories of the emotions. On those accounts, some cognitive-evaluative
interlude precedes the physiological-feeling stage of the emotion. Aren’t some
emotions like that, such as academic envy? Furthermore, it seems to me
that various neo-Jamesian attempts to accommodate these more apparently
 Comic Amusement, Emotion, and Cognition 93

cognitively driven emotional states are suspiciously ad hoc. According to

Jenefer Robinson, where deliberation precedes affective appraisal, there is
an intervening stage during which our emotional memory system matches
our deliberations with previous scenarios or situational types that then elicit
bodily appraisals.24 And this is where the emotion proper starts.
But, to me, there seems to be an extra step here. Why can’t deliberation
alone arrive at an appraisal sufficient to provoke a physiological change? Why
do we have to be rerouted through emotively charged memory types? It strikes
me that only the commitment to a thoroughgoing noncognitivism makes the
move tempting, which, in this context, would amount to presupposing what
needs to be solved.
Moreover, inasmuch as the neo-Jamesian supposes that cognitive states (in
the monitoring phase) can modify or even halt the pertinent feelings of phys-
iological perturbation, the question arises as to why cognitive states cannot
cause said feelings. If cognitive states can interact with these feelings so as to
evaporate them, why can’t they ignite them?
So, given these problems, for the moment at least, I think that it is fair to
assume that some emotional states are structured in the way that earlier cog-
nitive theories of the emotions suggested and that at least some cases of comic
amusement can be understood on this model.
But what of the remaining cases? Here I think that the incongruity theory
of comic amusement need not be thought to be as categorically inhospitable to
neo-Jamesianism as I have suggested thus far. Consider the two-stage struc-
ture of most jokes.

Sally calls her boyfriend Philip from the highway on her cell phone.
Sally tells Philip to be careful because the radio says there is a nut
on the highway driving in the wrong direction. Philip says, “Yeah, there
are hundreds of them.”

Stage one of the joke is completed when the punch line arrives. Punch lines
are by nature puzzling. They call for interpretations. Stage two occurs when
the listener reaches the interpretation. But the interpretation is generally as
incongruous or as absurd as the punch line. In the preceding joke, the punch
line pushes us to ask how could there be hundreds of people on the highway
driving in the wrong direction. Then we realize that Philip is the nut who is
driving in the wrong direction. But it is nearly unfathomable—it strains the
principle of interpretive charity to the breaking point—to imagine that there
is someone so dim-witted that he fails to see that he is driving the wrong way

24
Jenefer Robinson, Deeper than Reason: Emotion and Its Role in Literature, Music, and Art
(Oxford: Oxford University Press, 2005), pp. 70–75.
94 on emotions

on a crowded highway. Such a person would be as nonsensical as a moron. In

fact, Philip is a moron.
Th is scenario, however, fits very neatly with the neo-Jamesian charac-
terization of the emotions. At stage one, there is an affective appraisal.
Following Robinson, let us say it amounts to something like the apprehen-
sion that “Th is is strange and unexpected,” or maybe even “Th is is incon-
gruous.” Th at is, the affective appraisal stage involves pattern recognition,
which, of course, will also be sensitive to deviations from the pattern. Th is
may lead to the physiological response that we can call “being taken aback,”
which is then followed by cognitive monitoring that establishes that this
is harmless nonsense. Since it is harmless, the emotion that evolves is not
fear. Since it is nonsense, we do not hunker down to dismantle, dissect, or
to explain this anomaly. Instead, we enjoy it, and that enjoyment is signaled
either by laughter or a feeling of levity as our initial sense of being taken
aback is dispelled.
Although the incongruity theory of comic amusement appears to depend
upon the kind of cognitive theory of the emotions that neo-Jamesians eschew,
it need not. In fact, it can be segued nicely with neo-Jamesianism. So, if
neo-Jamesianism is true, that is no reason to discount as an emotional state
comic amusement understood in terms of incongruity. On the other hand,
if some cognitive theory of the emotions or some combination of a cognitive
theory and neo-Jamesianism is true, then the incongruity theorist of comic
emotion has no cause for anxiety.
The last objection to the conjecture that comic amusement is an emotional
state observes that the emotions involve vital human interests. Fear protects us
from harm and anger from injustice. Jealousy warns us that we may be losing an
important source of affection. Sorrow alerts us to loss. But if comic amusement is
an emotion, to what vital human interest is it connected? Again, the superiority
theory of comic amusement has an answer. The laughter of sudden glory rein-
forces our sense of social standing. But if the incongruity theory of comic amuse-
ment is correct, what vital interest does taking pleasure in incongruity serve?
One provocative suggestion has been offered by the polymath Jonathan
Miller—the neurologist, opera director, and former member of Beyond the
Fringe. Consonant with the incongruity theory of comic amusement, Miller
cites the scene in which Chaplin eats his boot in The Gold Rush, remarking that
it constitutes a “jarring discrepancy in which an object is suddenly forcefully
reclassified by being taken out of the category of the radically inedible and
placed into the category of the finely, the wonderfully edible. . . . This scene reju-
venates our sense of what these everyday categories are.”25 By playing with our
categories and concepts in this way, Miller argues, “we have prevented ourselves

25
Jonathan Miller, “Jokes and Joking: A Serious Laughing Matter,” in Jonathan Miller et al.,
eds., Laughing Matters (Essex, England: Longman Group UK, 1988), p. 15.
 Comic Amusement, Emotion, and Cognition 95

from becoming slaves to the categories we live by”; in this regard, humor is the
“rehearsal and re-establishment of concepts.”26
Expanding on this theme, Miller notes, “In all procedures of life there are
rules of thumb which enable us to go on “automatic pilot. . . . We depend on the
existence of these categories in order to go about our everyday business. Jokes
allow us to stand back from these rules and inspect them.”27 As a result of this,
Miller contends being comically amused enables us to rehearse and revise the
categories we live by thereby, according to Miller, restoring us to what he calls
“more versatile versions of ourselves.”28 Thus, is comic amusement connected
to the serious business of serving our highest interests.
Miller’s proposal is very congenial to the incongruity theory of comic amuse-
ment. And although I concur with Miller that the service comic amusement per-
forms concerns cognition through play with our concepts, I am not convinced
that humor has much to do with the production of new and better concepts,
as Miller seems to think. I suppose that you could say that comic amusement
frees us from everyday norms and concepts. But it doesn’t give way to higher
sense; it leads to nonsense. Apart from certain philosophical counterexamples,
it is difficult to come up with many examples of comic amusement leading to
genuine concept revision. But Miller is not talking about philosophical coun-
terexamples. He’s talking about everyday comic amusement.
A more promising conjecture can be found in The Society of the Mind by Marvin
Minsky, cofounder of the Artificial Intelligence Laboratory at MIT.29 Like Miller,
Minsky emphasizes the connection between comic amusement and cognition.
And like Miller, Minsky maintains that the primary function of humor is to dis-
rupt the heuristics we deploy in everyday life. For Minsky, we are possessed of
a plethora of rules of thumb that we use in order wend our way through life. Of
course, these heuristics are imperfect reasoning routines. They work quickly, but
not always accurately. Indeed, these heuristics can go wrong, often in more ways
than one. According to Minsky, the function of comic amusement is to apprize
us of the many ways that normal thinking and reasoning can go wrong. The
function of humor is to signal the existence of “cognitive bugs.”30 Analogizing

26
Ibid.
27
Miller, p. 16.
28
Ibid.
29
Marvin Minsky, The Society of the Mind (New York: Simon and Schuster Paperbacks, 1988),
especially pp. 274–281. See also Marvin Minsky, “Jokes and Their Relation to the Cognitive
Unconscious,” in J. Hintikka et al., eds., Cognitive Constraints on Communication (Amsterdam:
Reidel, 1981).
30
For an example of what Minsky has in mind, consider this humorous anecdote from ancient
China: A sage encounters a man, a peasant walking in front of his horse. The sage asks the man
why he is walking the horse rather than riding it. The peasant answers, “Because six legs are faster
than four.” Here, a generally reliable heuristic such as “more legs, greater speed” misfi res in con-
text. It is Minsky’s contention that the function of comic amusement is to locate bugs like this in
our cognitive system as a means of rooting out mistakes.
96 on emotions

his theory to Freud’s, Minsky claims that via humor, we are able to build up
what amounts to an unconscious cognitive censor that polices everyday think-
ing and guards us against the wealth of errors our heuristics can induce. There
are so many jokes just because our heuristics mislead us in so many ways. Joking
might be thought of as a course in informal logic.
I think that Minsky’s approach to comic amusement is superior to Miller’s,
since Minsky is not under the illusion that comic amusement is the source of
better thinking. Its function is to call our attention to flawed thinking. Yet
Minsky’s analogy to Freud’s notion of unconscious censorship does not suc-
ceed. For Freud, jokes elude the censor. In contrast, Minsky maintains that
jokes construct the censor—piece by piece, heuristic misfiring by heuristic
misfiring. Nor is the disanalogy an insignificant one. The source of the pleasure
in jokes on Freud’s account is lifting the censorship. That is, Freud explains the
pleasure in joking in terms of its service to an unconscious id. But Minsky has
no cognitive id—probably for the simple reason that a cognitive id would be a
contradiction in terms. However, that leaves Minsky with no role for pleasure
in his account.
In fact, I think that the problem here is worse than this. The fact that we
take pleasure in comic nonsense would seem to predict the opposite of the con-
struction of a cognitive censor. Since we enjoy comic incongruities, wouldn’t
that encourage us to perpetrate them? The more the merrier, so to speak. We
don’t typically suppress what gives us pleasure. Since comic incongruities make
us happy, wouldn’t that dispose us to commit more rather than fewer absurdi-
ties? Perhaps this is borne out by the readily observed phenomenon that one
joke leads to another.
In this regard, contra Minsky, we would predict that comic amusement
is the natural enemy of cognition rather than its benefactor. Plato certainly
worried about this with respect to the education of his Guardians. He recom-
mend that they should not be encouraged to laugh, because those habituated to
laughter are inclined to say or to do anything in order to indulge it, no matter
how alien to reason. If humor is linked to “cognitive bugs,” as Minsky and the
incongruity theorist agree, then what service does comic amusement perform
for cognition? Wouldn’t the pleasure it engenders reinforce our production
of errors rather than diminish them. In short, Minsky’s notion of a cognitive
unconscious censor leaves unexplained the role of comic enjoyment in service
to the interests of cognition.
Although it is true that one typically does not spend a great amount of
time dwelling on a joke or other form of cognitive incongruity, it is quite
common to run the punch line and its interpretation through one’s mind a
few times in order to savor it. Often people repeat the punch line out loud a
few times in order to replay, in a manner of speaking, the joke or the bon mot
a second or third time around. Like the moment when the interpretation of
 Comic Amusement, Emotion, and Cognition 97

the punch line fi rst dawns on us, these replays are suff used with pleasure.
The function of the pleasure here, it seems to me, is to encourage us to focus
intently upon the comic incongruity. Translating this into Minsky’s lan-
guage, the pleasure component of comic amusement rivets our attention on
the cognitive bugs in play and induces us to scrutinize them closely in order
to know them better so as to avoid them in our activation of the panoply of
heuristics, rules-of-thumb, norms, and concepts that we deploy in everyday
life. Evolution rarely attaches pleasure to something unless it has cause to.
Pleasure is connected to comic incongruity to alert us to the pitfalls to which
our mental equipment is irresistibly prone, as is shown by how easily we,
as they say, “fall for a joke.” Thus, comic amusement, conceived of in terms
of the enjoyment of incongruities, can be seen to be, like the other emo-
tions, serving vital human interests, namely, our interests in cognitive well
functioning. 31
Comic theorists, like Max Eastman, have often attempted to assimilate
humor and comic amusement as a form of play. This is clearly insufficient,
since there are so many different kinds of play other than comic amusement.
However, if play is protected practice, perhaps some aspects of the play the-
ory of humor can be connected to the preceding account of the evolutionary
value of comic amusement with respect to cognitive incongruities and mis-
takes, namely, that the function of humor is to afford practice with debugging
the kinds of mistakes and absurdities to which our reliance upon heuristics,
rules-of-thumb, schemas, and the like makes us prone. Comic amusement is
pleasurable, and that pleasure inclines us to indulge it frequently in play as
a means of sharpening our attentiveness to the errors to which our optimal
rather than fully rational, imperfect, down and dirty reasoning procedures lay
us open.
Moreover, if comic amusement is existentially serviceable for the manner
in which it draws our attention to local flaws in our ways of thinking, it also
at least has the potential, global significance of disclosing to the reflective stu-
dent of humor the frailty of human thinking. In comic amusement, we learn
how readily we err. Rather, pace the superiority theory, comic amusement does
not and should not serve as source of human arrogance but instead as an occa-
sion for humility, for an appreciation of our human weakness.32

31
The theory propounded above was fi rst presented on February 7, 2009, at a conference on the
philosophy of the emotions that was held in memory of Robert Solomon at the University of Texas
at Austin. Since then, a similar, more highly developed theory along the same lines as Matthew M.
Hurley, Daniel C. Dennett, and Reginald B. Adams, Inside Jokes: Using Humor to Reverse-Engineer
the Mind (Cambridge, MA: MIT Press, 2011).
32
The suggestion that the experience of comic amusement involves a humbling dimension is
also advanced, though through a different set of considerations, in Solomon and Solomon, “Are
the Th ree Stooges Funny?”
98 on emotions

Concluding Remarks
Before concluding, let me air two worries that some of you might have about
what I’ve said. I’ve proposed an analysis of comic amusement. This is some-
thing that wiser philosophers than me, such as Ted Cohen, the dean of the
philosophy of humor, have said repeatedly is impossible. With respect to my
own proposal, you might complain that the central notion of incongruity is still
insufferably vague. Perhaps. Nevertheless, I still think that it is a valuable exer-
cise to attempt to define comic amusement, since every failed attempt tells us
more about the phenomenon. Moreover, the notion of incongruity, even if too
elastic, is a valuable heuristic. It can be used to pinpoint the pertinent features
in a wide range of comic examples from narratives, to limericks, riddles, sat-
ires, and even practical jokes.33 And working through these examples under the
guidance of the notion of incongruity we can simultaneously gain more insight
into what is going on—what structures are in play—with respect to humor
while also sharpening our understanding of the pertinent kinds of incongrui-
ties. That is, by embracing the incongruity as a heuristic, we can begin to gather
the kind of data we need for a better theory.
A second objection to this paper is that the very project is quixotic. Who
cares whether or not comic amusement is an emotion? What difference does it
make if it is not an emotion? Admittedly, probably not very much. Yet again, the
value here is heuristic. We are starting to learn a lot about the emotions. Comic
amusement is still very mysterious. Thus, by contemplating comic amusement
in terms of current thinking about the emotions, we may be able to put our-
selves in a position where we can parse the structures of comic amusement
with greater clarity than our predecessors.

33
For an example of what I have in mind by using the incongruity theory as a heuristic, see my
“Two Comic Plots,” in Art in Three Dimensions (Oxford: Oxford University Press, 2010).
 5

Intellectual and Other Nonstandard

Emotions
Michael Stocker

Th is work is mainly about intellectual emotions—briefly, emotions about

intellectual matters. It is also about some other nonstandard emotions,
to rely on William James’s categorization of emotions in his “What Is an
Emotion?”1
We philosophers are familiar with intellectual emotions—perhaps as
familiar and familiar in the same, even if unknowing, ways that M. Jourdain
in Molière’s Le Bourgeois Gentilhomme was familiar with prose. It is only a bit
of an exaggerations to say that we talk about these emotions all the time
(even if we do not talk about them under the rubric of “intellectual emo-
tions”) and have done so from our earliest writings: To remind ourselves of
this, we could start by considering the etymology of “philosophy,” itself, fol-
lowed by reading the description of Socrates’ life and works in the Phaedo
and throughout many other dialogues; the discussion in the Republic of the
methods, contents, and goals of the guardians’ education; the steps on and
ultimate goal of the ascent to love of wisdom and goodness in the Symposium.
So too, we could read Aristotle about contemplation and the pleasures of
inquiry and of knowing (discussed below); Aquinas about these, especially
when turned to God; Hume on how philosophy and hunting have much the
same pleasures and emotions; Mill on the value, the need for, intellectual
culture; and so on and on.
Nonetheless, some thirty years ago, some of us complained, I think with
complete justification, that our philosophical contemporaries often denied,
explicitly, or implicitly by ignoring them, the importance or even the existence

1
Mind, 9, 1884, 188–205; widely reprinted.

99
100 on emotions

of intellectual emotions.2 And often enough, I still am asked whatever could I

have in mind by “intellectual emotions,” how could there be emotions where
intellect is concerned? (Sometimes I am asked whether I am concerned with the
cognitive content of emotions—for example, that fear of F requires thoughts
or beliefs that F is dangerous. This is not my present concern.) I mention this
to make a point about the history, especially the recent history, of our subject:
What needs to be explained is how and why, during (roughly) the first part
of the twentieth century, a major concern of earlier philosophers came to be
forgotten. As shown by some of the theorists I discuss below, intellectual emo-
tions are no longer forgotten.
To help characterize these various emotions—both intellectual and other
nonstandard ones—and to set the stage for my discussion of them, it may help
to mention some contrasts used by philosophers and other theorists in their
discussions of emotions.
The contrasts are doing and thinking; body and mind; activity and passiv-
ity; practical and theoretical; personal and interpersonal. These are important
contrasts, but like many other contrasts—such as reason and emotion, a con-
trast I will not examine—they have often been misused in ways that hinder
our understanding of emotions.
I want now to present several somewhat long passages of texts important for
this work, presented in chronological order: the first by Aristotle; the second by
William James; the third by Théodule Ribot; and the fourth by two psycholo-
gists, both theorists of emotions, Nico H. Frijda and Louise Sundararajan. The
main goals here are to show that intellectual and other nonstandard emotions
are well-known in our tradition, to show some of the variety they come in, and
to begin to answer some questions and issues they raise.

Aristotle
In Book X of the Nicomachean Ethics, Aristotle writes that “complete happiness
is a contemplative activity . . . the activity of God, which surpasses all others in
blessedness, must be contemplative; and of human activities, therefore, that
which is most akin to this must be most of the nature of happiness.” And also:

So if among excellent actions political and military actions are distin-

guished by nobility and greatness, and these are unleisurely and aim

2
See, for example, Israel Scheffler, “In Praise of the Cognitive Emotions,” in In Praise of
the Cognitive Emotions and Other Essays in the Philosophy of Education (New York: Routledge,
1991), originally published in Teachers College Record, 72, 1977, 171–186; and Michael Stocker,
“Intellectual Desire, Emotion, and Action,” Explaining Emotions, Amélie O. Rorty, ed. (Berkeley:
University of California Press, 1980).
 I n t e l l e c t u a l a n d O t h e r N o n s t a n d a rd E m o t i o n s 101

at an end and are not desirable for their own sake, but the activity of
intellect, which is contemplative, seems both to be superior in worth
and to aim at no end beyond itself, and to have its pleasure proper
to itself (and this augments the activity), and the self-sufficiency, lei-
sureliness, unweariedness (so far as this is possible for man), and all
the other attributes ascribed to the blessed man are evidently those
connected with this activity, it follows that this will be the complete
happiness of man, if it be allowed a complete term of life (for none of
the attributes of happiness is incomplete).

I suggest that we take what Aristotle says about contemplation to show the
value he attaches to intellectual activity and emotions. To be sure, what we
mean by “contemplation” is different enough from what Aristotle means by
(what is here translated as) “contemplation,” namely, theoria. Nonetheless, I use
this passage from Aristotle to show that he, too, recognizes the existence and,
more importantly, the high value of thinking that is not a constituent of action.3
I recognize that here and elsewhere, Aristotle does count thinking as activity,
as in “the activity of intellect” in the passage just quoted. So I should have said
that Aristotle recognizes the high value of thinking that is not a constituent of
action, as “action” is often understood, such as in our contrasting pairs thought
and action or thinking and doing. Further complicating my use of this contrast, I
continue to hold what I have long argued for in “Intellectual Desire, Emotion,
and Action,” for example, that thinking can be an action, a doing. Thinking is,
after all, part of our job description, part of what we are hired to do.
It should, further, be noted that contemplation as understood in this pas-
sage from Aristotle almost certainly does not involve emotions. When they
are contemplating, his Gods and the men of perfect or best eudaimonia may
well not be excited, interested, enthralled by, curious about, in awe, and so on.
Aristotle considers contemplation to be an example of perfect or total activity
and thus no passivity, no pathē, and thus no emotions, although, as just seen,
he does allow for contemplation and other acts of thinking to be pleasurable.
In these ways, it might seem misleading to use Aristotle on contemplation to
introduce intellectual emotions.
However, to be a useful introduction, there need not be the sorts of similari-
ties that, as just suggested, our contemplation and thinking do not have with
contemplation as understood by Aristotle. I suggest that if we think about what
Aristotle is saying, we can be (and I was) naturally, even though not neces-
sarily, led to thinking about our intellectual activities, including at least some
instances of what we call contemplation, that do or can involve emotions.

3
For a similar use of “contemplation” and a defense of using it this way, see Peter Goldie,
“Virtues of Art,” Proceedings of the Aristotelian Society, Supplementary Volume, 82, 2008, 179–195.
102 on emotions

In Book X we also read, “Now the activity of the practical excellences is exhib-
ited in political or military affairs.” This quote in conjunction with the earlier
quote from Book X might suggest that on Aristotle’s view, intellectual activity
is to be divided exclusively and without remainder between practical thinking
and contemplation. But this suggestion is clearly mistaken. In the Metaphysics,
Aristotle writes:

All men by nature desire to know. An indication of this is the delight

we take in our senses; for even apart from their usefulness they are
loved for themselves; and above all others the sense of sight. For not
only with a view to action, but even when we are not going to do any-
thing, we prefer sight to almost everything else. The reason is that
this, most of all the senses, makes us know and brings to light many
differences between things.

And in Book X of the Nicomachean Ethics: “those who know will pass their time
more pleasantly than those who inquire,” suggesting once again that inquiring
can be pleasurable.
It must also be remembered that Aristotle was keenly interested in gaining,
deploying, and contemplating knowledge about all sorts of matters and that
many of these matters allowed for intellectual delight even though they are
not the subject matter of contemplation nor are they for the sake of the politi-
cal or military action. Biology and the constitutions of various city-states come
immediately to mind.
There would be a further problem in delimiting practical thought that is
“exhibited in political or military aff airs” so that practical thought—even
when limited to those aff airs—does not allow for intellectual emotions. One
very ruthlessly constraining way to try to do this would limit the practical
thinking or its emotional uptake to what goes into and is essential for the
practical activity. To this end, it might be held that the only emotions about
an episode of military strategizing would be about its effectiveness and effi-
ciency in producing the activity or its goal, such as winning. Th is would be
to hold that the cleverness of the strategizing, its profundity, its fiendish-
ness, its beauty, . . . could not be sources of delight or upset; nor even could
the thought that it succeeded against all odds; nor could there be awe at
its audacity, delight in what the victory will allow, and so on—unless these
in turn are understood just in terms of effectiveness and efficiency. Th is
would allow no room for anything like the emotional refi nement and savor-
ing discussed by Frijda and Sundarajan below nor, of course, for pleasures
of philosophizing about the nature of military courage as in the Laches or
in the Nicomachean Ethics —unless, of course, these are understood simply
as bearing on efficiency and effectiveness. I think such excessive practical-
 I n t e l l e c t u a l a n d O t h e r N o n s t a n d a rd E m o t i o n s 103

mindedness is a caricature even of Spartans. It would be more than simply

strange to apply it to Aristotle.

William James
For a use of the contrast between mind and body as that is relevant to emo-
tions, we could hardly do better than quote at length from James’s “What Is
an Emotion?”

I should say first of all that the only emotions I propose expressly to
consider here are those that have a distinct bodily expression. That
there are feelings of pleasure and displeasure, of interest and excite-
ment, bound up with mental operations, but having no obvious bod-
ily expression for their consequence, would, I suppose, be held true
by most readers. Certain arrangements of sounds, of lines, of colours,
are agreeable, and others the reverse, without the degree of the feel-
ing being sufficient to quicken the pulse or breathing, or to prompt to
movements of either the body or the face. Certain sequences of ideas
charm us as much as others tire us. It is a real intellectual delight
to get a problem solved, and a real intellectual torment to have to
leave it unfinished. The first set of examples, the sounds, lines, and
colours, are either bodily sensations, or the images of such. The sec-
ond set seem to depend on processes in the ideational centres exclu-
sively. Taken together, they appear to prove that there are pleasures
and pains inherent in certain forms of nerve-action as such, wherever
that action occur. The case of these feelings we will at present leave
entirely aside, and confine our attention to the more complicated cases
in which a wave of bodily disturbance of some kind accompanies the
perception of the interesting sights or sounds, or the passage of the
exciting train of ideas. Surprise, curiosity, rapture, fear, anger, lust,
greed, and the like, become then the names of the mental states with
which the person is possessed. The bodily disturbances are said to be
the “manifestation” of these several emotions, their “expression” or
“natural language”; and these emotions themselves, being so strongly
characterized both from within and without, may be called the stand-
ard emotions.
Our natural way of thinking about these standard emotions is that
the mental perception of some fact excites the mental affection called
the emotion, and that this latter state of mind gives rise to the bodily
expression. My thesis on the contrary is that the bodily changes follow
directly the PERCEPTION of the exciting fact, and that our feeling of
104 on emotions

the same changes as they occur IS the emotion. Common sense says,
we lose our fortune, are sorry and weep; we meet a bear, are frightened
and run; we are insulted by a rival, are angry and strike. The hypoth-
esis here to be defended says that this order of sequence is incorrect,
that the one mental state is not immediately induced by the other,
that the bodily manifestations must first be interposed between, and
that the more rational statement is that we feel sorry because we cry,
angry because we strike, afraid because we tremble, and not that we
cry, strike, or tremble, because we are sorry, angry, or fearful, as the
case may be. Without the bodily states following on the perception,
the latter would be purely cognitive in form, pale, colourless, destitute
of emotional warmth. We might then see the bear, and judge it best
to run, receive the insult and deem it right to strike, but we could not
actually feel afraid or angry.4

Many, if not most, commentators take James to be claiming an identity between

feelings of bodily changes and emotions—that’s what emotions are. It is gener-
ally thought that this identification is obviously wrong, indeed wrong-headed
because it fails to recognize, and even denies, that emotions involve cognitive
content, not just feelings of bodily changes; and more particularly that it fails
to recognize and even denies that emotions are to be characterized, and to be
differentiated from other emotions, in terms of such content, not or not just
these feelings of bodily changes.
But there is another way to understand James’s claim. This is to follow an
interpretation offered by Charles Young.5 Young suggests that James does not
identify emotions and those feelings and that he does recognize that emotions
have cognitive content including thoughts, desires, and values. Young suggests
that James holds that these feelings are, indeed, a differentiating feature of
emotions: perhaps serving to differentiate emotions from other emotions but,
far more importantly, serving to differentiate an emotion from its own constit-
uent thoughts, desires, and values. As just read, James holds that without the
bodily feelings, those other constituents are “pale, colourless, destitute of emo-
tional warmth.” This is to take James as holding that these feelings of bodily
changes are what make emotions emotional. These feelings are what make for
affectivity.
I agree with James that emotions require affectivity and on the need to
differentiate affectivity from the other constituent elements of emotions.

4
James, pp. 189–190, italics in original
5
In his review of Sarah Broadie’s Ethics with Aristotle, Journal of the History of Philosophy, 31,
1993, 625–627, 625. Young’s suggestion is discussed briefly in my and Elizabeth Hegeman’s
Valuing Emotions (Cambridge, UK: Cambridge University Press, 1996), pp. 25–26.
 I n t e l l e c t u a l a n d O t h e r N o n s t a n d a rd E m o t i o n s 105

However, I do not agree that affectivity can be characterized in terms of feeling

of bodily changes.6 In the passage just quoted, James suggests that while the
emotionality of the standard emotions is to be understood in terms of bodily
feelings, the emotionality of nonstandard emotions is not. As will be discussed
below, somewhat later in “What Is an Emotion?” James significantly modi-
fies this view about the nonbodily nature of the emotionality of nonstandard
emotions.
A further point against taking James to identify emotions or emotional-
ity with feelings of bodily changes—and, not incidentally, against the charge
of failing to see that thoughts, desires, and values are constituents of emo-
tions—is found in claims that, for there to be an emotion, these feelings of
bodily changes must be “interposed between” mental states. If both the bodily
feelings and the mental states are constituents of the emotion, emotions can-
not be just the feelings. The feelings cannot be the emotion—that is, the whole
of the emotion.
If, on the other hand, those mental states are not constituents of the emo-
tion, but only important, even essential, elements of a series—mental states,
feelings of bodily changes, mental states—the consensus view that James
identifies emotions with bodily feelings turns out to be almost merely termi-
nological, turning on the question of whether in fact or on James’s view the
term “emotion” applies to the whole series or just part of it. I say almost entirely
because, as just said, James holds that the feelings are needed to make the
second lot of mental states emotional rather than “pale, colourless, destitute of
emotional warmth.”
The eminent contemporary theorist of emotions, the psychologist James
Averill, puts these points well.7 He says, correctly in my view, that James
defined emotions, not as identical to bodily feeling but rather as “subjective
experience dependent upon bodily change” (216).8
What is perhaps of most importance in this passage from James’s “What Is
an Emotion?” for present concerns is that he restricts his discussion of these
issues about bodily feelings to what he calls the standard emotions. As we read,

6
See Valuing Emotions and my “Psychic Feelings: Their Importance and Irreducibility,”
Australasian Journal of Philosophy, 61, 1983, 5–26.
7
“Analysis of Psychophysiological Symbolism,” in The Emotions: Social, Cultural and Biological
Dimensions, Rom Harré and W. Gerrod Parrott, eds. (London: SAGE, 1996) from which the follow-
ing is taken, originally in Journal for the Theory of Social Behavior, 4, 1974, 147–190.
8
Averill notes that Carl Lange, also a nineteenth-century emotion theorist, in work originally
published in 1885, did identify emotions with bodily changes or bodily events. James’s theory and
Lange’s theory were often paired—perhaps confl ated—giving us what is now called the “James–
Lange Theory,” not the “James and Lange Theories.” Th is pairing might be explained by James and
Lange being listed as coauthors of the still well-known book, The Emotions, which includes both
Lange’s 1885 work and James’s “What Is an Emotion?”—such as the 1922 edition (Baltimore:
Williams and Wilkins, 1922).
106 on emotions

he states that he does not intend it to apply—indeed, that he intends it not

to apply—to some other emotions. I will call these nonstandard emotions—a
term I do not think he uses. He includes among these nonstandard emotions,
first, some intellectual emotions; second, some other nonbodily emotions; and,
I think, third, some other emotions that do involve the body but are not, pri-
marily, bodily, and do not involve, are not manifested by, bodily disturbances.
(Whether or not he had Darwin in mind in making these suggestions about
nonstandard emotions, James might be read as supplementing and perhaps
correcting Darwin’s very brief comment on intellectual activity: that it is often
expressed by a frown and down cast eyes.9 James says that he was concerned to
supplement and perhaps correct Darwin on standard emotions: “But not even
a Darwin has exhaustively enumerated all the bodily affections characteristic
of any one of the standard emotions” [191].)
James characterizes these nonstandard emotions by means of briefly
sketched examples with some very brief commentary. On intellectual emo-
tions, he writes, “It is a real intellectual delight to get a problem solved, and
a real intellectual torment to have to leave it unfinished” gives the examples,
about which he says that they “seem to depend on processes in the ideational
centres exclusively.” For other nonstandard emotions that do not involve feel-
ings of bodily changes, he writes, “Certain arrangements of sounds, of lines, of
colours, are agreeable, and others the reverse, without the degree of the feeling
being sufficient to quicken the pulse or breathing, or to prompt movements of
either the body or the face.” (These, of course, are also some of the objects of
pure pleasures in Plato’s Philebus.)
Although he does not say this, I take the last quoted passage to suggest that
James holds that there is a third group of nonstandard emotions. These involve
still other arrangements of sounds, lines, and colors that (perhaps in addition
to be being agreeable or otherwise) do have a degree of feeling that is sufficient
to quicken the pulse or breathing, or to prompt movements of the body or face.
The second and perhaps also the third set of examples are “either bodily sensa-
tions or the images of such.” (These, too, are discussed in the Philebus, a bit ear-
lier.) Nonetheless, bodily feelings that may be associated with these are not said
to be what makes these emotions emotional. Rather, it is strongly suggested
that these feelings, if any, are not what makes these emotions emotional.
In another of his articles, aptly named “Intellectual Emotions,” Averill sug-
gests that hope is an intellectual emotion.10 As I understand Averill and James,
hope does not fit completely well into any of James’s categories of nonstandard

9
Darwin, The Expression of the Emotions in Man and Animals (New York: Appleton, 1898), 222.
10
The Emotions: Social, Cultural and Biological Dimensions, Rom Harré and W. Gerrod Parrott, eds.
(London: SAGE, 1996); originally in C. D. Spielberger et al., eds., Stress and Anxiety (Washington,
DC: Hemisphere, 1991), v. 14, pp. 3–16.
 I n t e l l e c t u a l a n d O t h e r N o n s t a n d a rd E m o t i o n s 107

emotions. It seems most like James’s second (and perhaps also the third) non-
standard emotion, not his first, which he calls intellectual.
But it is of only terminological interest whether Averill uses the very same
categories James does. James uses only a few examples with very brief com-
mentary—in my view, too few examples and far too little commentary—to tell
us what he will not be concerned with. He did not give, and did not set out to
give, a complete typology of what he was not concerned with. I wholeheart-
edly endorse Averill’s closing claim that what is important is that there are any
number of different emotions and different sorts of emotions:

Most contemporary theories do not recognize hope and other ‘intel-

lectual emotions’ as true emotions. I believe this says more about the
nature of our theories than about the nature of hope.
For the most part, our theories of emotion have been erected on a
very narrow base. Fear, anger, and to a lesser extent, love and grief
have been the primary emotions investigated. Yet, literally hundreds
of emotions are recognized in ordinary language. (36)

As just noted, James gives hardly any attention to the nonstandard emo-
tions—at least in the passages just quoted—but only the brief examples and
commentary just quoted. His own lack of discussion may help explain why
these emotions, or even the fact that James thinks there are such emotions,
are absent from at least many of the commentaries on him or on “What Is an
Emotion?”
I want to make two very brief comments about James on these nonstan-
dard emotions. First, as he recognizes, they are a heterodox lot, with few, if
any, common features—other than the negative one of not being a standard
emotion. In these passages, he further suggests that there need be no feelings
of bodily changes in intellectual emotions or in emotions of the second group
of nonstandard emotions, nor is their emotionality, their affectivity, to be
explained in terms of such feelings. The third sort of nonstandard emotions
may well involve such feelings, but their being emotional is said not to depend
on that. Nonetheless, there is no suggestion here that any of the emotions
of these three nonstandard sorts is “pale, colourless, destitute of emotional
warmth.” Indeed, to the contrary. (These last characterizations will be ques-
tioned below.)
Second, James gives every indication of seeing nothing strange or problem-
atic or unusual about these nonstandard emotions. He gives every indication
of confidently expecting his readers to recognize what he is talking about and
to be able to distinguish between standard and nonstandard emotions and
among the various sorts of nonstandard emotions. (This is taken up below, in
the discussions of Ribot and of Frijda and Sundarajan.)
108 on emotions

What I have reported as James’s views about nonstandard emotions is, I think,
accurate to what he says in the early part of “What Is an Emotion?” But he later
adds to that view, changing it significantly: holding that without bodily feelings,
there are no intellectual and other nonstandard emotions, but only judgments
that are not emotional but rather dry, pale, and lacking all glow.11 He writes,

Rapture, love, ambition, indignation, and pride, considered as feel-

ings, are fruits of the same soil with the grossest bodily sensations
of pleasure and of pain. But it was said at the outset that this would
be affirmed only of what we then agreed to call the “standard” emo-
tions; and that those inward sensibilities that appeared devoid at first
sight of bodily results should be left out of our account. We had bet-
ter, before closing, say a word or two about these latter feelings. They
are, the reader will remember, the moral, intellectual, and aesthetic
feelings. Concords of sounds, of colours, of lines, logical consistencies,
teleological fitnesses, affect us with a pleasure that seems ingrained
in the very form of the representation itself, and to borrow nothing
from any reverberation surging up from the parts below the brain. . . .
We have then, or some of us seem to have, genuinely cerebral forms
of pleasure and displeasure, apparently not agreeing in their mode of
production with the so-called “standard” emotions we have been ana-
lyzing. . . . Unless in them [i.e., cases of pure cerebral emotion] there
actually be coupled with the intellectual feeling a bodily reverberation
of some kind, unless we actually laugh at the neatness of the mechani-
cal device, thrill at the justice of the act, or tingle at the perfection of
the musical form, our mental condition is more allied to a judgment
of right than to anything else. And such a judgment is rather to be
classed among awarenesses of truth: it is a cognitive act. But as a mat-
ter of fact the intellectual feeling hardly ever does exist thus unaccom-
panied. The bodily sounding-board is at work, as careful introspection
will show, far more than we usually suppose. Still, where long famil-
iarity with a certain class of effects has blunted emotional sensibility
thereto as much as it has sharpened the taste and judgment, we do get
the intellectual emotion, if such it can be called, pure and undefiled.
And the dryness of it, the paleness, the absence of all glow, as it may
exist in a thoroughly expert critic’s mind, not only shows us what an
altogether different thing it is from the “standard” emotions we con-
sidered first, but makes us suspect that almost the entire difference
lies in the fact that the bodily sounding-board, vibrating in the one
case, is in the other mute.12

11
My thanks are owed to Peter Goldie for urging me to pursue this.
12
James, pp. 201–202
 I n t e l l e c t u a l a n d O t h e r N o n s t a n d a rd E m o t i o n s 109

So, James now says that the nonstandard emotions, along with the stan-
dard ones, involve and depend on bodily feelings: that without bodily feel-
ings, there is no emotionality in either the standard or the nonstandard
emotions.
I see various interpretive possibilities here: first, James no longer distin-
guishes between standard and nonstandard emotions—perhaps he now holds
that although many theorists and others do claim there is a difference in kind,
they are mistaken. Second, he holds that all emotions involve bodily feelings
but that there are differences of strength or kind between those feelings the
standard emotions involve and those the nonstandard ones do. (This probably
should be that there are such differences between the two classes of emotions:
between [most] every standard emotion and [most] every nonstandard one.)
Third, he sees the difference in terms of what the emotions are about—such as
an onrushing bear, a neat argument. Fourth, he holds that the bodily feelings
are “manifestations,” “expressions,” “natural language” of only the standard
emotions and not also of the nonstandard ones.
A full study of James’s ultimate view on intellectual and other nonstandard
emotions would require determining which, if any, of these interpretations is
right. Emotion theorists have given extensive consideration to the third and
fourth possibilities. I think we have already profited enough from these brief
comments on James not to pursue these issues or that full study.
To conclude my study of James, I will simply make a suggestion about how I
see some contemporary studies of emotions. It is that few, if any, contemporary
emotion theorists accept the details of James’s account of emotions, stand-
ard or nonstandard. However, some theorists, including the groundbreaking
researcher Antonio Damasio, do praise James and say they are inspired by his
work.13 And many do retain what to my mind is the “fl avor” of James, at least
to the extent that they divide up the emotions in ways that, again to my mind,
suggest James. We have already seen this in Averill’s criticism of his fellow
emotion theorists, for using “a very narrow base” composed mainly of “fear,
anger, and to a lesser extent, love and grief.” (Frijda and Sundarajan will be seen
to make a similar criticism.)
I am also reminded of James—in particular of his reliance on the body
and biology in his account of emotions and his bifurcation of emotions—by
the work of many who look to neurobiology for help with their accounts of
emotions. At least roughly, where James talked of bodily feelings, they talk of
neurobiological structures and happenings. As just indicated, I am thinking
here of Damasio—and his distinction between primary and nonprimary emo-
tions. I am also thinking of Paul Griffiths’s What Emotions Really Are,14 espe-

13
Descartes’ Error (New York: Putnam, 1994), pp. 129–131.
14
Chicago: University of Chicago Press, 1997.
110 on emotions

cially his distinction between basic emotions and higher cognitive emotions.
As Jason Clark writes:

According to Griffiths, basic emotions are evolutionarily old sets of

stimulus-detection mechanisms and stereotyped physiological and
behavioral response patterns rooted in dedicated neural circuits, with
a genetic basis that has evolved in response to recurrent evolutionary
challenges, and which we share with nonhuman mammals. These pat-
terns are capable of being activated by unconditioned stimuli and lead
to unconditioned, involuntary responses. . . . 15

Théodule Ribot
That these emotions, or at least the intellectual ones, were seen as everyday,
ordinary emotions—even if also seen as nonstandard emotions—around this
time can also be seen in The Psychology of the Emotions,16 written by the French
psychologist Théodule Ribot. The following passage from Ribot was written
within a decade or so of James’s work, to which he refers. Ribot writes:

I shall be very brief in treating of intellectual emotion, since it is rare,

and usually temperate in character; however, . . . [it can spring] up with
the true characteristics of intense emotion. . . . Most human beings are
not passionately eager for the search after or the discovery of pure
truth, any more than they are afflicted by privation of it; but those
possessed by this demon are given up to him, body and soul. . . . The
biographies of learned men furnish us with innumerable examples:
the perpetual physical sufferings of Pascal, Malebranche nearly suf-
focated by the palpitations of his heart when reading Descartes,
Humphrey Davy dancing in his laboratory after having made the dis-
covery of potassium, Hamilton suddenly feeling something “like the
closing of a galvanic circuit” at the moment of discovering the method
of quaternions, etc. There is no need to extend our search so far; every-
day life provides us moment by moment with examples which, though
prosaic, are none the less valuable as proofs. The instinct of curiosity is
at the root of all intellectual emotion, whether lofty or commonplace.
Does not the man who perpetually watches his neighbour’s conduct
and the thousand petty details of his life, feel when his puerile curios-
ity is baffled, all the physical anguish of unsatisfied desire? (101)

15
My thanks are owed to Clark for help here. The quotation is from an unpublished work of his.
16
New York: Scribners, 1897.
 I n t e l l e c t u a l a n d O t h e r N o n s t a n d a rd E m o t i o n s 111

I offered this passage simply to give evidence that talk of intellectual emotions
was accepted as entirely ordinary. I will not discuss it further, except to note
that, as they are put, the claims about the bodily feelings, such as physical
anguish, do not make it clear whether Ribot accepts James’s later, modified
account or Lange’s or some other account of the relations between bodily feel-
ings and emotions.

Nico H. Frijda and Louise Sundararajan

I turn now to Frijda and Sundararajan’s fascinating and illuminating “Emotion
Refi nement: A Theory Inspired by Chinese Poetics.”17 Th is work includes an
extensive bibliography of works on emotions, mainly psychological, some
philosophical, and some on Chinese aesthetics.18 Frijda and Sundarajan
write:

William James made a distinction between coarse and noncoarse emo-

tions. In the present article, we explore the nature of such noncoarse
emotions, which we designate as emotions with refinement. We take
our cue from the treatment of refined emotions in Chinese poetics
and philosophy. The theory and description of savoring (in Chinese,
pin-wei) points to several features of emotion experiences and behav-
ior that are usually absent in direct emotional responses of emotional
events, such as self-reflexivity and higher level second-order aware-
ness, detachment, and restraint. Emotions with those features can be
found outside savoring and aesthetic contexts, for instance while deal-
ing with actual life events. . . .
Coarse emotions [James’ standard emotions] are characterized by
distinct bodily upset, overt behavior manifestations with brisk time
courses, and relatively simple event–emotion relationships. They are
more or less paradigmatic and are the main subject of psychological
studies of emotion (e.g., Frijda, 1986; Lazarus, 1991; Oatley, 1992).

17
Perspectives on Psychological Science, 2, 2007, 227–241.
18
Works of particular note for us include N. Frijda, The Emotions (Cambridge, UK: Cambridge
University Press, 1986); R. Lazarus, Emotion and Adaptation (New York: Oxford University
Press, 1991); K. Oatley, Best laid schemes: The Psychology of Emotions (Cambridge, UK: Cambridge
University Press, 1992); D. Keltner & J. Haidt, “Approaching awe, a moral, spiritual, and aesthetic
emotion,” Cognition and Emotion, 17, 2003, 297–314; M. Kubovy, “On the pleasures of the mind”
in D. Kahneman, E. Diener, & N. Schwarz, eds., Foundations of hedonic psychology: Scientific perspec-
tives on enjoyment and suffering, (New York: Russell Sage, 1999); L. Sundararajan, “Religious awe:
Potential contributions of negative theology to psychology, “positive” or otherwise,” Journal of
Theoretical and Philosophical Psychology, 22, 2002, 174–197.]
112 on emotions

They are best illustrated by an emotion-arousing event like meet-

ing a bear in the woods. Coarse emotions differ from emotions that
are more felt than acted upon, and thus do not obviously manifest
themselves in overt behaviors like attack, embrace, or fl ight; may not
show very pronounced physiological upset; are often about complex
events or subtle event aspects; and are not easily done justice by com-
mon emotion labels. We will refer to noncoarse emotions collectively
as . . . refined emotions. . . . We hold that they do not form a subset of
emotions. It is not that anger would be considered as a coarse emo-
tion and that love would be considered as a refined emotion. Rather,
refinement represents a mode of perhaps all emotions that language
or emotion taxonomy could distinguish. There exist refined anger,
love, and sexual ecstasy, as well as coarse, straightforward anger, love,
or sexual ecstasy. We think it worthwhile to examine emotion refine-
ment to benefit emotion theory by shedding some light on the rela-
tionship between refined and coarse emotions, on emotions that are
not done justice by simple emotion labels, and on emotions aroused by
perceiving objects of art, often called aesthetic emotions, which pose
unresolved problems for theories of emotion.
Our analysis will show how refined is an appropriate designation
for emotions that show few outward signs but still involve strong feel-
ings and that share the following features: They occur under attitudes
of detachment and restraint, their experience involves reflexive sec-
ond-order awareness, they result from and contain extensive elabo-
ration of appraisal of the eliciting events that may invest the events
with meanings far beyond their immediately given aspects, and they
include virtual states of action readiness rather than states that mani-
fest in overt acts or suppressed action impulses. [227]
Again, we get a division of emotions. And again, one sort, coarse emotions,
is characterized largely in terms of coarse bodily feelings and reactions. In
terms of James’s categories of nonstandard emotions, refined emotions fit
most easily into the fi rst sort, intellectual emotions. But not all intellectual
emotions, as discussed ever so briefly by James, fit easily into the category
of refi ned emotions. Again, this is only of the barest, merely terminologi-
cal importance. The refi ned emotions are of obvious importance, both to our
emotional lives, how we live, and also to our accounts of emotions. Once we
are—now that we have been—introduced to them by name and description,
we see that we have known them and their importance well, even if we were
rarely if ever aware that we knew them. So far as theories and accounts of
emotions are concerned, we are brought once again to wonder how it was pos-
sible to be satisfied with coarse emotions, with James’s standard emotions—
as the major emotional elements of our lives or of our accounts of emotions.
 I n t e l l e c t u a l a n d O t h e r N o n s t a n d a rd E m o t i o n s 113

Perhaps this is to be explained in terms of coarse thinking. As I said at the

outset, Frijda and Sundarajan present us with a fascinating and wonderfully
illuminating work.
This concludes my showing that we, joined by many of our philosophical
predecessors, know full well that there are intellectual emotions. I now want
to turn to two general issues about intellectual emotions. These issues deserve
far more time than I can give them here. But I hope that what I say will make a
good start on them. These issues are, in turn, First, are some or all of the intel-
lectual emotions the same emotions as found elsewhere? Are some or all of
the nonintellectual emotions found in intellectual emotions? And second, how,
if at all, are emotions, especially intellectual emotions, needed for successful
action, especially intellectual action.
Are some or all of the intellectual emotions the same emotions as found
elsewhere? Are some or all of the nonintellectual emotions found in intellec-
tual emotions? One way to start on answering these questions is by noting
that emotions of the same name occur in both. There are any number of lists
of emotions that bear this out. Consider, for example, the chapter headings of
David Konstan’s deeply informative The Emotions of the Ancient Greeks:19 anger,
satisfaction, shame, envy and indignation, fear, gratitude, love, hatred, pity,
jealousy. It is, I think, obvious that all of these can be about intellectual issues
or exercises of the intellect and in other sorts of nonstandard, mental, perhaps
refined, nonbodily emotions. Let us consider, even if very briefly, only the first
and the last, anger and jealousy.
I can be angered by a belittling review. So too, I can be angry at my own
stupidity for continuing down what I should have seen was a dead end. (Anger
in the last sentence moves away from Aristotle’s requirement of insult for
anger where “anger” is used to translate orge [pronounced orgay] as discussed
by Aristotle in Rhetoric II, 2, which in turn is discussed at length in Valuing
Emotions.20 It moves away from this to our contemporary and more capacious
notion of anger. Now jealousy: I can be jealous about my reputation: I am con-
sumed by fear that I will lose my reputation or that it will be stolen from me
(perhaps by the person who made the belittling remark, who, I now surmise,
made the remark to steal my thunder).
I recognize that having the same name, such as “anger,” is compatible
with the emotions being different, at best only analogically similar. Th is,
however, involves a very general issue, not just between standard emotions,
taken as a unified group, and nonstandard ones, taken as another unified
group.

19
University of Toronto Press: Toronto, 2006.
20
Michael Stocker and Elizabeth Hegeman (Cambridge, England: Cambridge University Press,
1996).
114 on emotions

Suppose I am asked whether intellectual courage—willingness to risk

ridicule, loss of job or freedom, loss of reputation to think along certain lines
or to publish my conclusions—involves the same sort of courage as courage
understood as a standard emotion. Is this question different from a question
we might ask about courage as treated by Aristotle: are his paradigm and non-
paradigm “sorts” of courage—courage in battle and courage in facing danger-
ous and painful medical conditions—really different sorts of courage? Or are
they only different sorts of objects or occasions for the self same emotion?
Returning to intellectual emotions, we might ask whether courage in pursuing
one’s research—facing the danger and pain of challenging one’s own or one’s
society’s deeply held beliefs—is yet another sort of courage or only a different
occasion for the selfsame courage as found in military or medical situations.
These questions about sameness can also be asked about courage-in-battle.
So we might ask whether courage in facing a cavalry charge is the same courage
as courage in facing missiles or about courage in facing a cavalry charge when
you know that your army is about to be routed and when you know that the
opposing army is about to be routed; and so on.
My point here is not that I think these questions and their answers are silly
or pointless. It is rather that, first, I think that questions and answers about
sameness and difference can come up all over the place. (Here we might turn
to the Philebus around 18a and Socrates’ warnings about going haywire about
the one and the many.) Second, I think these questions are the same questions
whether they are about just one intellectual emotion or just one nonintellec-
tual emotion, or about pairs consisting of one of each, or about suitably chosen
groups of each taken severally or jointly. I would put these two together to
suggest, third, that differing answers to the original question—“Are some or
all intellectual emotions the same emotions as found in or about nonintellec-
tual objects and situations?”—are perhaps disguised applications of differing
accounts or theories of emotion identity and individuation. (So too, of course,
for holding that these are or are not the same questions.) And I would sug-
gest, fourth, that even if it can be done we have no present need to sort out
and evaluate these different accounts and theories. All this may be little more
than showing that I don’t embrace realism about emotion individuation and
identification.
I want now to consider some very general questions of how, if at all, are
emotions, especially intellectual emotions, needed for successful action, espe-
cially intellectual action. For example, is interest, engagement, or love of truth
needed for good intellectual activity or even just any intellectual activity. If it
is, how and why is it needed? I have just suggested that the same emotions are
found in both standard and nonstandard cases. So, I will start on these ques-
tions by following up a theme from Aristotle’s discussion of courageous action.
He holds that courageous action requires, inter alia, facing the dangers of battle
 I n t e l l e c t u a l a n d O t h e r N o n s t a n d a rd E m o t i o n s 115

with or from the right emotions, including not too much and not too little,
but rather a mean amount of, fear, and also with or from the right sorts and
amount of concerns, such as an ultimate desire to do what is good, for example,
save one’s fellows and one’s city. Winning in battle is a goal of such action. (We
need not be concerned here with whether winning is only a goal or the goal nor
with what winning is, such as wounding, killing, driving off the enemy.) The
best men will, if the situation arises, act courageously. But, on Aristotle’s view,
the best men may well not be the best at winning. A mercenary—who fights
for money rather than for the good—might be better at this, and he might be
better because he has less fear, perhaps because he has less to fear, less to lose,
than a good man would.
I use this—whether or not I agree with it—to illustrate a way that emotions
can be essential to an act’s being a given sort of act and, indeed, for being a good
and successful instance of such an act.
Aristotle’s understanding of courage as differentiated from being a good sol-
dier is, of course, tied up with his concern with virtue and his rejection of what
is often seen as a utilitarian account of virtue, in terms of doing good things,
getting good things done. On such a view, a good person, a person of virtue,
is a person who succeeds at doing or bringing about what is good. Perhaps we
should call this not the utilitarian view but a utilitarian view, for some utili-
tarians reject that view and agree with Aristotle. I have particularly in mind
Mill’s claims in On Liberty and elsewhere that we should be concerned “not only
[with] what men do, but also what manner of men they are that do it.” I have
mentioned this not because I am concerned with the adequacy or otherwise of
utilitarianism but rather because I am concerned with showing that in some
cases at least, the answer to whether emotions are or are not required by vari-
ous acts and conditions can have to do with the moral views involved with the
relevant understanding of those acts and conditions.
Identifying the relevant understandings (or arguing for or against particu-
lar understandings) will be difficult and perhaps controversial. Nonetheless,
let us start by considering what professors, lecturers and other teachers, and
students of philosophy are expected to do. (I do not assume, one way or the
other, whether other disciplines are the same.) Let us start by asking what is
required of professors giving lectures on, say, moral realism to other equally
advanced professors.
Suppose we hold that they have succeeded if they present the relevant theo-
ries and accounts adequately and that they have done this if (but certainly not,
only if) their lecture is pretty much like the pages of a good textbook—perhaps
they are reading out loud the page proofs of their most recent book. What need,
it might be asked, is there for intellectual (or other) emotions in those giving
the lectures or in those listening to them? Some might think it obvious that
there is no need and perhaps even no room. But there is some controversy here,
116 on emotions

which I will only mention, not try to adjudicate. One strand of this controversy
concerns the question of whether we can understand a particular philosopher’s
position without a deep understanding of the philosopher, where that requires
something like a depth psychological understanding of the philosopher, the
philosopher’s temperament, emotional makeup, and the like. (Iris Murdoch
suggests something like this in Metaphysics as a Guide to Morals.)
Another strand of the controversy concerns the very nature of understanding
and whether to be understandable a position, claim, or whatever must answer
to our interests; or must be capable of engendering feelings of certainty (for or
against) or of uncertainty; or requires some form of commitment. (I think sug-
gestions favoring something along one or other of these lines are found in writ-
ings of Hilary Putnam and Jurgen Habermas’s Knowledge and Human Interests21
and in Bas van Fraassen’s The Empirical Stance.22)
Further, some have held that, at least at the highest level, large-scale claims
and theories—at least in the pure sciences and perhaps also in philosophy—are
to be evaluated in terms of simplicity, elegance, and power and (other) pragmatic
considerations. These claims are sometimes put as being about the content of
the theories, what they are about, and sometimes as being about theory accept-
ance and reception. It would go considerably too far afield to enter into these
debates.23 So too, it would go too far afield to explore the relations between
emotions and simplicity, elegance, power, and pragmatic considerations.
But two points can be made. First, even if understanding itself or simplic-
ity, say, are ultimately to be understood as involving emotions, that by itself
does not show that a theorist who aims to be understood or for simplicity or
judges work in terms of understandability or simplicity has those emotions.
(The point is at least somewhat like this one: a theorist may use procedures that
have been developed in light of certain mathematical proofs without being able
to do those proofs or, indeed, without even knowing about their existence.)
This may be to say little more than that the needed emotions and proofs need
not be in these theorists’ heads or minds. It will be sufficient, I think, if the
discipline, itself, incorporates those goals—if, for example, they are part of the
relevant research programs. Second, having certain emotions may certainly be
useful for those theorists. So, for example, without his interest in moral real-
ism, a given member of the audience might not have the energy or commitment
to keep his attention focused on the lecture. This, of course, can be true of some
members of the audience but not others. These others might “just see” what
the lecturer is on about. For a parallel, some people “just see” that the butler

21
Boston: Beacon Press, 1971.
22
New Haven, CT: Yale University Press, 2002.
23
For a useful and compelling discussion, see David Owens, Reason without Freedom (New
York: Routledge, 2000); see also Habermas and van Fraassen.
 I n t e l l e c t u a l a n d O t h e r N o n s t a n d a rd E m o t i o n s 117

committed the murder, while others have to struggle with the clues, which only
those deeply interested by such mysteries would do.
Let us turn briefly to lower-level lecturers. Judging from the evaluation
forms for teachers that I have seen, securing students’ understanding of the
subject is the bare minimum. It is expected, even required, that enthusiasm
or at least interest in the subject be engendered. In part, this is for budget-
ary reasons, where a university department receives funds in proportion to the
number of students in their classes, and one important factor in taking a class
is interest in the subject. (It is obviously not the only factor; classes that satisfy
prerequisites for further study may be uninteresting, even boring or worse.)
There are also other reasons for engendering interest. One such reason is
(the view) that part of what it is to be well educated, to be a person of even mod-
erate culture, to be well-rounded, a good citizen even is to be interested in some
intellectual disciplines or at least to find them interesting: perhaps poetry or
music, or perhaps one or more sciences or a social science.
Teachers of all ranks are evaluated not only in terms of the interest they
engender in their students and colleagues but also in terms of the interest they
have and show. To be judged “not interested in the subject” is a matter of some
considerable concern—for retention, promotion, and the like. In part, I think,
this has to do with (thoughts about) effectiveness in teaching and research. It
is often thought, correctly or not, that teachers who do not evince interest do
not engender interest. Theories and anecdotes about emotional “contagion” are
often cited here. So too, it may be that students are unlikely to be interested
or sustain interest in a subject unless the teachers show that they consider it
important, and it may also be that showing interest is one of the best ways,
perhaps even a needed way, to show that. (Only the greatest actor, it might be
thought, is able to be successful in simulating or faking such interest.)
These instrumental needs for interest are important. But I think there may
be more to say. Some comments on interpersonal relations and group dynamics
should help fi ll this out. Most of us know how unwelcome hostility and divi-
siveness are in a department (and, of course, elsewhere). For similar reasons,
not being interested in the subject is also unwelcome. Here we should think of
what it is like for others in a department to have one or more colleagues who
have suffered burn out, who have lost commitment and interest, whose hearts
are no longer in it. We should also think of what this may be like for the people
who have lost interest.
This has been to speak about the roles of certain intellectual emotions—
mainly interest in philosophy—in a highly particular, nonuniversal “setting”:
a contemporary philosophy department, in universities like ours, in societies
like ours, with social needs and possibilities like ours, with students like ours,
staffed with people like us, with experiences, psychologies, expectations like
ours, and so on.
118 on emotions

Some changes in such settings may make emotions, sometimes including

intellectual and other nonstandard emotions, more important or less impor-
tant. Here, instead of focusing on a philosophy lecture, we could consider other
“talks”: for example, a sermon intended to rally the faithful; a keynote address
at a political meeting; heartening troops about to enter battle; a declaration
of love; soothing an angry customer; and so on. It would misunderstand these
talks to hold that to be successful they do not need to engender emotionally
laden assent but will be counted successful if they engender only the emotion-
less assent or understanding of the sorts I suggested are all that is required by
the advanced philosophy lectures.
I want now to turn to a somewhat different issue. It is obvious enough
that intellectual emotions can be useful. A moment’s thought should be suf-
ficient to see that they can also be harmful for intellectual activity. Consider
savoring, as discussed by Frijda and Sundarajan. It is often good to stop
to smell the roses. But there are times when this is inappropriate, danger-
ous, wasteful of time, and so on. There are some people who are too eager
to get things done, too fi xated on doing and achieving (as these are com-
monly understood—doing as contrasted with thinking), who are unwilling
to spend time thinking out and reconsidering what is to be done, and so on.
So too, there are some who are too willing to stop, to digress, to be distracted
from the tasks at hand, to rethink over and over, who are unable to commit
themselves to action, and so on—who do not get beyond thinking to doing.
Some people are too enthusiastic, with never a moment of repose and still-
ness, who enthuse over and get caught up in every new fad or theory. Some
are never enthusiastic enough. These can be character fl aws or problems that
arise only occasionally or even just this once. Much the same holds for reflec-
tion and theorizing. There are times and places for engaging in these and
times and places where to engage in them would endanger the very activity
reflected on.
All this is obvious enough. But it leaves open the question of how, if at all,
intellectual emotions are important, perhaps even essential, for good intellec-
tual work. I do agree that many of our intellectual heroes were deeply interested
in what they thought and did. Ribot’s pantheon of Pascal, Malebranche, Davy,
and Hamilton make this clear, as do the intellectual biographies of Einstein
and of Watson and Crick, and so many others.
But perhaps on the other side, there is Fleming’s just happening to notice
the precursor of penicillin. As the story goes, his noticing what he did was
more attributable to his attentiveness to anomalies in his surroundings than
it was to his engagement. As well, we have to deal with science and other
intellectual inquiry done by teams of researchers, technicians, and other
workers. Here what is important for success may be more tied to their taking
part in progressive research programs (to use Lakatos’s terms) with a good
 I n t e l l e c t u a l a n d O t h e r N o n s t a n d a rd E m o t i o n s 119

division of labor, good leadership, and the like than to interest, fascination,
curiosity, and other intellectual emotions.
As I see matters, we are here confronted by at least two very difficult, if not
intractable, problems. The first has to do with the relative importance of indi-
viduals and of groups, and their interrelations, in accounts of intellectual (and
other) work. As this might be put: are we to look more to psychology or rather
to sociology and social theory for these accounts? Emphasis on intellectual
emotions seems to suggest concern with individuals. (Of course, it need not.
It can allow for group psychology and it can see individuals as being shaped
by, as expressing the concerns of, groups.) These very difficult issues have been
discussed at length, and, I think, inconclusively in the subdiscipline of the phi-
losophy of history, under the rubric of the role of the hero, the great man or
woman, in history.
I intend to avoid any explicit concern with these issues and to turn to an
issue, perhaps to be located in moral psychology, about intellectual and other
interest. At issue is the likelihood of significant intellectual work being done by
an individual or a group without intellectual interest and without excitement
about the subject, even without interest in or love for the subject.
To put this, I will sketch how I imagine medieval scribes copying docu-
ments—an activity of incalculably immense intellectual worth. (Those bothered
by my deficient imagination are invited to think of members of a contemporary
research team.) I imagine that one lot of scribes found the documents, them-
selves, or copying them exactly, deeply engaging. A second lot, however, found
the copying just a job of work, even drudgery. They were told by their superiors
to copy carefully and exactly, which they did because they were told to. A third
lot did it out of a sense of duty or a religious or a professional calling.
Were these last two sorts of scribes interested in what they were doing? We
might conclude that they were not interested in the copying as such. But it
might be replied that the second lot must have been interested in having and
keeping a job, despite the drudgery and that the third lot must have been inter-
ested in fulfi lling their duty or their calling.
The reply about these people, especially the second lot, is useful as a way to
put my worry about interest. I do see how it can be said that these people were,
must have been, interested in having and keeping the job. (I am imagining that
they were not forced, as prison inmates might be, to do the copying.) But—and
this is my worry—the considerations leading us to hold that they were inter-
ested in having and keeping their jobs are pretty much the same considerations
that could lead us (and have led some) to say that those considerations show
that they wanted to have and keep the job. Why else would they do it? The reply
might be, “To earn a salary to pay for their keep.” The reply, in turn, to this is
that this shows that they were interested in earning a salary to earn their keep
and similarly for wanting to earn a salary to earn their keep.
120 on emotions

I agree that there is a case for saying that they were interested in something—
much like the case for saying that they wanted something. Some might say that
this uses a discredited, overstretched sense of “wanting.” But I think this is a
legitimate use of “wanting,” even though it comes to little more than acting
intentionally, intending or trying to achieve what the person is said to want. I
think “interest” in these last cases is understood similarly: what in this sense
interests people is whatever they are intending or trying to achieve; whatever,
from their point of view, is the point of what they are doing.
This, or a similar, understanding of “interest”—interest understood simply
in terms of acting intentionally with a goal—is of importance for us in at least
two ways. First, some such interest of these last two sorts of scribes is not intel-
lectual interest in the copying. So, intellectual interest in copying is not needed
for good copying, even if some other interest is needed. Second, this sort of
interest may well not be (part of) an emotion. So, the claim that good intellec-
tual work requires an intellectual emotion of interest is doubly wrong.
My second point relies on the more general claim that acting intentionally
toward a goal is possible even if one does not act with or from an emotion:
intentional action need not involve any emotion. If this is right, we should
examine whether intellectual interest must involve the intellectual emotion of
interest or any other intellectual emotion. If it need not, then the claim that
good intellectual work requires the emotion of intellectual interest or some
other intellectual emotions will be wrong in three ways.
This in turn raises questions about the role and nature of intellectual inter-
est vis-à-vis being a person of intellect, in common parlance an intellectual.
An intellectual, as I understand this notion, is a person who is interested in
intellectual matters, who has intellectual interests, and who is admirable as a
person in light of having such interests. I think intellectual emotions are also
needed. It thus seems that, in these regards, an intellectual vis-à-vis intellec-
tual emotions is like Aristotle’s good man vis-à-vis the emotion(s) of courage.
In both cases, having the emotion(s) is needed.
Aristotle’s courageous person and an intellectual may also be similar in the
following way: Aristotle holds that courage is not needed to be the best and
most successful soldier; a mercenary can be a better soldier than a courageous
man. I suggested above that having intellectual emotions, not just intellectual
interests, may be necessary for being an intellectual. But someone who is merely
intelligent, careful, attentive to details, and the like may well be a better intel-
lectual worker, producing more and better intellectual products. (The issues sur-
rounding whether we should hire and admire the best intellectual worker or the
best intellectual raise similar issues about being red vs. expert, as discussed and
fought over in China in the middle decades of the twentieth century, or about
being a loyal party member vs. a skilled technocrat, as raised in examinations of
the “spoils system” entrenched in American governmental practices.)
 I n t e l l e c t u a l a n d O t h e r N o n s t a n d a rd E m o t i o n s 121

I think these issues quickly lead to very large and difficult issues, some of
which were mentioned earlier and most of which are too large and too difficult
to do more now than simply announce them. I do not think we know—I know
that I do not know—enough about the general, panhuman roles of intellectual
emotions in the production of good intellectual work. I am unable to declare
on the possibilities of intellectual success in a society or organization where
intellectual workers—most, many, a significant number of them—do not have
intellectual emotions, much like some of the scribes mentioned above. I think
that in some cases at least, other interests, other incentives, could achieve
what, in other cases, intellectual emotions would.
Th is last claim might seem to go against what I said earlier about the
roles of intellectual emotions in teaching and learning. If it does, I should
have been more specific. I said something like “good teachers of philosophy
to undergraduates need to engender intellectual emotions, at least curios-
ity about, perhaps even a love of, philosophy.” I may not have made it clear
enough that I was talking about how I think matters now stand with the
students we now have. I am thinking not of all students but those who will
not pursue studies that do not interest them, that do not engender and sus-
tain interest and other (positive) intellectual emotions. I include under the
term “studies” those as large-scale as, say, physical sciences; or, somewhat
smaller, philosophy; or, still smaller, a particular course offered by a par-
ticular teacher.
There is, of course, controversy over whether students’ following their inter-
ests is good or bad. So too, there is controversy over whether students’ fol-
lowing their interests marks a significant change in student behavior. I do not
know if it is or, if it is, how much of a change there has been in, say, the last 100
years or, again if it is, how to explain it.
I have been concerned with the question of how important intellectual emo-
tions are for good intellectual work. My comments about students following
their interests and scribes doing what they are told to do or acting from a calling
or duty raise two interrelated questions about these different motivations. One
concerns the issue of how likely it is that the different motivations could have
the same results. How likely is it that both finding the copying interesting and
feeling a calling or duty could result in equally good intellectual work, perhaps
even the same intellectual work? The other concerns the relative importance
in people’s psychic economies—their organization and budget of motives—of
pursuing what one finds interesting, on the one hand, and following a calling
or the call of duty, on the other. I think it clear that there are some people in
whom one predominates and others in whom the other does: so, in order to get
good copying done, it may well be sufficient to have some scribes who find this
of interest or some others who feel called or duty bound to do this (or, of course,
some mixture of both).
122 on emotions

I will consider both of these by examining Adam Morton’s comments on his

powerful example:

An extraordinarily well trained and malleable young scientist . . . [with]

a superlative grasp of research techniques, . . . [who has] one flaw, how-
ever. She does not care about the subject. She has no curiosity. . . . She
does not find herself wanting the truth to turn out one way rather
than another in more than . . . [a career-enhancing] instrumental way.
She does not sometimes wonder whether lines of inquiry that are, with
good reasons, disparaged by her research group might not in the end
give important clues to the underlying processes she is investigating.
This scientist may well go on to do excellent work, and make signifi-
cant discoveries. She may become eminent. But it is unlikely that she
will lead her subject in radically new directions. Nor that she will be
the one to find the new way ahead if current approaches stall, or to see
deep subtle fl aws in those current approaches, or willingly take her
work in a direction that seems to her important but risks a lifetime of
obscurity. She is rather like the child prodigy musician with rare skills
and a marvellous technique, lacking only a love of music.24

I agree with what is implicit in Morton’s example: that a person with those
skills and training who also cares about her subject is likely enough—or as
likely as anyone can be—to be curious, to wonder about disparaged lines of
inquiry, to start radically new directions of research, to be alive to subtle flaws,
and so on. What I am unconvinced of, however, is that such caring about the
subject is necessary for any or all of these. As suggested about the scribes—and
earlier, in a related way, about the advanced lecture on moral realism or physics
to other equally advanced professors in the field—I think she could do any or
all of these things without the intellectual emotion of caring for the subject. It
would be sufficient, I think, if her team leader gave her well-articulated direc-
tions: “look into lines we disparage; make bold conjectures; get your colleagues
and students to follow these out; check for subtle or other fl aws”; and so on. So
too, I think it would be sufficient if she had a sense of calling or duty.
This is to agree with Morton that something more than abilities, skills, and
training is needed to make those accomplishments realistically likely. They are,
after all, above and beyond the ordinary. It is also to agree with him that care
for the subject might supply this something. But, now disagreeing with him, it
is also to suggest that other conditions—other motivations or incentives—can
also supply that something.

24
Adam Morton, “Epistemic Emotions,” in The Oxford Handbook of Philosophy of Emotion, Peter
Goldie, ed. (Oxford: Oxford University Press, 2010), 388–389.
 I n t e l l e c t u a l a n d O t h e r N o n s t a n d a rd E m o t i o n s 123

My general point here is that as useful as intellectual emotions are for good
intellectual work, they are not part of or essential to that work, as work is
understood in the relevant text or field, e.g., a philosophy article, a physics
lecture. In this, such work differs from music as depicted by Morton. (All I
will say on Morton’s claim about music is that it is controversial, as are other,
similar views that great art requires emotions in some or all of these ways:
the artist’s having them, transmitting them, and the audience receiving and
experiencing them.)
I have said that it seems to me that Morton’s researcher does not need the
intellectual emotion of caring for her subject to achieve those great intellec-
tual goods. But, for reasons given above, I am not entirely confident that this
is right. So even at the end my inquiry, I am unable to declare with certainty
on whether—and if so, how—good intellectual work requires intellectual
emotions.
This is to understand any such need factually—for example, in terms of the
psychic economy of different motivations and whether the same work could be
done from other motivations or incentives. However, I am confident about at
least some of the evaluative needs of the one for the other. Here, I take my lead
from Aristotle’s discussion of how the value of courage differs from the value
of winning and how what makes for the one value differs from what makes for
the other.
I have given the materials for my evaluative comments throughout this
work and elsewhere.25 So, I will be very brief here. Intellectual emotions are
instrumentally valuable, whether or not necessary, for good intellectual work
and good intellectual workers. Intellectual emotions are also intrinsically valu-
able—as constituents of good societies or good aspects of societies and also as
constituent values of good lives and good aspects of lives. For both instrumental
and intrinsic, constitutive reasons, then, intellectual emotions are to be recog-
nized, respected, encouraged, and sustained. It is in part due to these emo-
tions that philosophy and philosophers (both in and outside the academy) are
valuable—indeed, valuable enough to merit social support and personal com-
mitment. These emotions can be, and often are, central to the self-conception
and positive self-appraisal of intellectuals, people who pursue and even love
what is so valuable. Having those emotions is not just a job description; it is
also a matter of accomplishment, pride, and value.
Recognizing the value and the different ways intellectual emotions are valu-
able is, I think, at least as important as being able to determine the extent to
which good intellectual work requires those emotions.

25
For example, in “The Schizophrenia of Modern Ethical Theories,” “Values and Purposes,”
and Valuing Emotions.
 6

Authenticity and the Examined Life

Jerome Neu

On one notion of authenticity, one must “to thine own self be true.”1 But that
notion seems to assume that the self is transparent to itself. The truth is, how-
ever, that we may not always be sure exactly who we are. Indeed, in addition
to the possibilities of inchoateness, confusion, and self-deception, there may
be grounds for distinguishing a “true” from a “false” self, and so there may be
difficulties first in knowing which aspects of the self fall on which side of the
divide, and then further difficulties in being true to the true self. Must every
aspect of that self be given expression—is that what “being true” in this con-
text means?
We are torn in different directions by ideals of authenticity. While many
insist we should be who we are, that we ought to act on our real desires, incli-
nations, and attitudes, and thus express our true natures, the same people
typically believe that there are other ideals that ought to be striven for and
respected, ideals that would have us change or suppress those aspects of our
nature that are violent, sadistic, cruel, bigoted, nasty, envious, and in other
ways unseemly, harmful, or immoral. So if we are to express our true natures,
is that to be only after we have pruned them in accord with social and personal
norms? Or are norms (some? all?) themselves to be distinguished as either true
or false, so that only those that are true are to be retained as a part of the self
that is to be expressed and used to guide the shaping and expression of other
aspects of the self? And by what standards is that distinction to be drawn? It is
to be remembered that when distinguishing true and false norms, all personal
norms that might provide potential standards are themselves suspect as pos-
sibly false.

1
What started as a prescription from Shakespeare’s Polonius for ensuring virtuous
fair-dealing with others—“Th is above all, to thine own self be true,/And it must follow as the
night the day/Thou canst not then be false to any man.” ( Hamlet I.iii.78–82)—has become in
modern times a condition for self-fulfi llment (Trilling 1971, 3–4, 9, 11; Taylor 1992, 15: “the
moral ideal behind self-fulfi lment is that of being true to oneself ”).

124
 Authenticity and the Examined Life 125

True and False Selves

Care must be taken when “truth” is transposed from the world of sentences,
statements, and propositions, that is, the world of facts, to the world of values,
where “true” comes to characterize emotions and desires and selves. For exam-
ple, back in the sixties, many followed R. D. Laing of The Divided Self (1960)
in distinguishing an inner (real and true) self from an outer (false and deter-
mined) self. But the separation is as false as the Cartesian split between mind
and body that it mimics—both approaches treating the real or essential self
as though it were a disembodied mind. It is, in Laing’s language, a “schizoid”
vision of the self.
While perhaps easing a distinction between the sinner and his sins, the com-
forting vision of a well-meaning (and, in the full schizoid version, all-talented
and omnipotent) inner self should be resisted. There are several protections
against the metaphysical and moral temptation to regard one’s inner or mental
life as somehow “true” and one’s bodily life (with its overt, observable actions)
as external and somehow “false.” The first is to consider carefully what “false”
might mean here. In most senses (except where it is equated from the start with
things bodily and visible) it can apply equally to things mental and physical.
That is, emotions and thoughts may be as “false” as social roles in the sense of
being, for example, undesired, unchosen, and disliked. Properly understood,
the true–false distinction cuts across the mind–body distinction rather than
running parallel to it. Mental states (mind) can be false as well as true. Bodily
states can be true as well as false—remembering that “false” here most often
means something like undesired, unchosen, and disliked while “true” most
often means something like approved and identified with.
The dual status of both mental and physical states connects with a second
major protection against the schizoid delusion: the recognition that not all
social roles are false. We build our identity partly through others’ perception
and recognition of us. Some of the social roles that make us who we are we
in fact desire and choose. Being a parent, friend, student, or lover need not
be “false” just because each is a social role involving an embodied interacting
life. Going beyond that, more sophisticated arguments aimed at rendering
social roles “false” based on a metaphysics of freedom that rejects actions in
accordance with a role as a form of bad faith are equally suspect and perhaps
confused.
Sartre, in his discussion of “Bad Faith” in Being and Nothingness, explores the
example of a waiter in a café (1956 [1943], 59–60; see Neu 1988). His concern
is with a waiter who takes the duties of his role as determining his actions.
The assumption of a role does not lift the burden of choice. We remain people
whether we are waiters or, in a morally more serious context, soldiers, and we
cannot use the demands of our position as an excuse for our actions, for we can
126 on emotions

always (if necessary) abandon our particular role or position. Sartre certainly
has a point, a point perhaps more readily visible in the case of the soldier. Given
unconscionable orders, say, to slaughter innocent children, the soldier cannot
hide behind his role and pretend he has no choice. (Following orders, as the
Nuremburg trials made explicit, is no excuse.) We always have a choice to step
outside of our role, and it is bad faith, a denial of our freedom, to pretend that
we do not. Indeed, as Sartre would insist, refusing to choose, letting events
take their course, is itself a choice. Of course, stepping back from our role may
be costly. The waiter may be fired, the soldier court-martialed or even summar-
ily shot, but we must not deny the contrast between facticity and transcen-
dence. We are more than our given (or chosen) social roles, and even within the
constraints of a particular role we have choices. What does not follow is that
social roles are inherently inauthentic, inherently forms of bad faith. All roles
need not be a kind of pretense, a bad faith playacting and denial of freedom.
To see this, aside from remembering the point already made, that roles can be
desired and come to form an essential part of our chosen identities, consider
what the avoidance of bad faith, the project of sincerity, would require if acting
in accord with the requirements of our roles were disallowed. Would we always
have to reject or fail in our duties? Certainly no social role expresses everything
that we are or might be, but does it follow that good faith requires some sort
of dramatic role distancing or a life of endless irony? Surely, we can live out
and through our social roles while all the time recognizing that we are more
than any given role and that we may sometimes be called upon to abandon or
transcend the dictates of a given role. But not every moment is an emergency
or morally fraught.
Even the varied and changing requirements of etiquette, which can include
scripted roles that are no doubt socially imposed and so may seem the epitome of
self-alienated conformism, can form the basis of essential character traits—it
may be one’s true nature to be mannerly (Miller 1993, 30–31; 2003, 40–41).
Feelings against social demands of any sort may be understood in terms of a
romantic, Rousseauean and Rilkean, idolizing of the solitary as the originary
and ultimate home of authentic living (Guignon 2004). But while an emphasis
on individual creative and artistic aspects of the self and self-making are no
doubt part of the appeal of authenticity, it is to be remembered that interper-
sonal recognition and social aspects of the self may themselves be essential
both to creativity and (more broadly) to identity, indeed, to meaningful life.
Returning once more to the distinction between true and false selves
that depends on a schizoid split of mind and body, the inner and the outer, a
third remedy comes from consideration of what constitutes a “mental state.”
Philosophers such as Gilbert Ryle (1949) have emphasized the behavioral
aspects of intelligence, “knowing how,” vanity, and so on. And some of the indi-
vidual characteristics that we regard as most distinctive and valuable, such as
 Authenticity and the Examined Life 127

being witty or charming, depend on their public manifestation and their social
recognition. Can we be funny if nobody laughs? To cite Wittgenstein’s expan-
sive formulation, “The human body is the best picture of the human soul”
(1953, 178). The self inevitably becomes empty if it is regarded as disembodied
because the attribution and existence of many psychological states and indi-
vidual characteristics depend on their bodily expression.
The moral comfort of a retreat to a well-meaning inner self can be bought
only at the cost of gross distortion of just what it is that makes us who and
what we are. Sinners cannot shed their sins by a simple metaphysical shift in
identity. And the problem of the nature of the self that authenticity demands
be fulfi lled cannot be settled by detaching the self from the body and embod-
ied social roles nor, for that matter, as we shall see, by a notion of freedom and
self-expression that turns to unconstrained and arbitrary choices as the mark
of authenticity.

Real Authenticity
The truth of propositions has often been understood in terms of some sort of
correspondence with reality. Being true to oneself has also been seen to involve
some sort of correspondence. Lionel Trilling limns some of the relevant reso-
nances: “Who would not wish to be true to his own self? True, which is to say
loyal, never wavering in constancy. True, which is to say honest: there are to
be no subterfuges in dealing with him. True, which is to say, as carpenters and
bricklayers use the word, precisely aligned with him. But it is not easy” (Trilling
1971, 4). Some of the difficulty has to do with the precise nature of the rele-
vant correspondences, and the contrasting disharmonies, misalignments, or
failures of correspondence to be avoided. What does it take to be “authentic”?
While less broad in application than “real,” “authentic” fulfi lls a similar
function and is one of those words that is, in the terminology of J. L. Austin,
both substantive-hungry and a trouser-word (Austin 1962, 62–77). Austin has
taught us that such words depend for their significance on the nouns they mod-
ify and on what specifically is being excluded, that is to say, on contrast (“it is
the negative use that wears the trousers”; ibid., 70). There is no essential prop-
erty that all real things (be they real whipped cream or real diamonds) have in
common, and the same for authentic things. Turning the adjective into a noun,
going from “real” to “reality” or “authentic” to “authenticity” does not yield
a common essence that marks an ideal of existence. Even when one narrows
one’s interest to an “authentic life,” there are as many alternative attributes as
for “real life” (e.g., “reality TV” may be unscripted—though it is not always even
that—but that is not what college students are talking about when they speak
of leaving school for “real life”).
128 on emotions

To understand what authenticity requires, one does best to look to what

is meant to be excluded, the forms of inauthenticity that are being avoided.
Here I believe that what Stuart Hampshire has to say about ideals of sincerity
can equally well be said of authenticity. In “Sincerity and Single-Mindedness”
(1972), Hampshire distinguishes sincerity of statement from sincerity of a
state of mind. The first requires a correspondence between what one says and
what one thinks. The virtue aimed at amounts to honesty and the danger to be
avoided is lying.2 Such correspondence of what one says and what one thinks
is relatively easy of achievement. Spontaneity should be enough. In effect, one
could achieve such sincerity or authenticity by following the creed of the Beats,
Ginsberg and Kerouac: “First Thought, Best Thought.” One mustn’t edit one-
self; one must live without fi lters. Or, as French moralists from Chamfort to
Gide would have it, one must not watch oneself living. Just be spontaneous: say
what you think.3
But sincerity of a state of mind would appear to require greater effort, because
the virtue to be achieved (what Hampshire refers to as “single-mindedness”)
requires a correspondence, not between what one says and what one thinks but
between what one says and what one does. Failure here is commonly referred
to as hypocrisy, and to avoid having what one says undermined and belied by
what one does, spontaneity is not enough—at least not if one takes seriously,
first, the peculiarities of reflexive knowledge and, second, the place of the
unconscious in human life.
As Hampshire, following Spinoza, emphasizes, given the fact that our beliefs
about the sources of our states of mind are themselves a part of those states

2
Irony too may involve a mismatch between what one says and what one thinks, indeed
what one says may be the very opposite of what one thinks. Still, irony need not involve the
intention to deceive or (more broadly) mislead of lying. Even complex irony, where what gets
said is different from or other than what is thought without being precisely contrary, may
make insight into the speaker’s mind more diffi cult without amounting to lying (Nehamas
1998, 46–69). In the end, irony need not be incompatible with sincerity. One’s meaning can
still come through. There remains room for mystery and confusion wherever meaning is less
than transparent—ironical, and fi gurative, and even literal speech, in certain circumstances,
can obscure. But lying and deceit are moral fl aws; irony is a rhetorical mode and (in the work
of Richard Rorty and others) a tentative and skeptical attitude.
3
Of course, such spontaneity does not guarantee consistency over time or a stable and
steady self. The complexities emerge particularly tellingly in Diderot’s (1966) account of
Rameau’s Nephew, which interested Hegel for its picture of a “disintegrated, alienated, and
distraught consciousness,” and whose implications for ideals of authenticity are explored by
Trilling (1971, 26–52, quote above 47) and Williams (2002, 172–205, quote below 189): “The
model of sincerity as uninhibited spontaneity even applies to his [Rameau’s nephew’s] fl attery
and deceit. In those activities, as he makes clear, the expectations of his audience make him
actually become, to order and for a while, what they require; his availability to them takes the
form, not of a systematic misreporting of his states, but rather of an instantaneous imperson-
ation, a[n] improvisation of another short-lived personality.”
 Authenticity and the Examined Life 129

of mind, a certain complexity requiring self-awareness and self-watching is

necessary if we are to give a full account of any given (or evolving) state of
mind. The peculiarities of reflexive knowledge, knowledge of our own men-
tal states, will not allow us to look away and unselfconsciously achieve full
self-understanding. If we look, we may sometimes discover that we are inclined
towards incompatible ideas about the sources of our current state of mind, and
these confl icting thoughts can be enough to make our current state of mind
itself confl icted and confused. Our states of mind may be subject to a kind of
indeterminacy. As Hampshire puts it, “Insofar as the subject is confused and
uncertain, when he asks himself what he believes, what he wants, and what
he feels about something, his actual beliefs, desires, and feelings are confused
and unformed also” (1972, 242). And this is before one comes to dynamically
unconscious thoughts—that is to say, forcefully resisted as opposed to merely
neglected and unsurveyed thoughts. That there are such thoughts seems beyond
doubt. There are forces at work within us of which we are not always or readily
aware, and unobserved they can end up determining what we do despite our
best (or, at any rate, our declared) intentions. We must engage in self-watching
if our statements are to be reliable guides to our actions.
Socrates famously insisted “the unexamined life is not worth living” (Plato,
Apology 38a). He had in mind the elenctic challenging of one’s own beliefs with
the implications of those beliefs, a constant testing and retesting for consis-
tency and coherence that was the hallmark of his own life and philosophical
practice. As Nietzsche put it, what is required is not the courage of one’s con-
victions but rather “the courage for an attack on one’s convictions” (Nietzsche
1967, 226 n. 3). It is only through the survival of merciless self-criticism
that one can have any conviction that one’s beliefs are worth living and (as
in Socrates’s case) dying for. But there may be another reason why the unex-
amined life is not worth living: for without suitable self-examination, we can
never be sure whether it is in fact our own life that we are living. That is, if one
admits the possible existence of hidden, subterranean, unacknowledged forces
that may to some extent govern one’s life, that may determine one’s actions,
without suitable self-examination it may never be clear whether it is such alien
forces that are living their life through us while we passively surrender to their
impetus. Those forces may be of the sort that Freud uncovered in the uncon-
scious, that Marx found in social structure, that the modern descendants of
Darwin see in one’s genes, or they may involve the situational cues that social
scientists attend to, but whatever their precise character, if we are unaware of
them and they are in fact what move us, we may have to be regarded as their
playthings rather than as independent agents. We may think we are doing what
we want when what we take as “our” wants are the manipulated and manipulat-
ing product of forces we would repudiate were we but aware of their existence.
The problem may sometimes emerge in especially disturbing form on occasions
130 on emotions

when we realize that we are not managing to do what we think we want because
we are constantly undermining ourselves, that is to say, tripping over ourselves
because of unacknowledged inner confl ict, and thus falling into hypocrisy and
other forms of inauthenticity.
I have been arguing that what effort is required to achieve authenticity or
sincerity depends on (following Hampshire) what correspondence is being
maintained and (following Austin) what contrasting condition is being avoided.
I would now add a third correspondence and a third contrast to the two already
mentioned:

(1) thinking-saying vs. lying

(2) saying-doing vs. hypocrisy
(3) thinking-doing vs. faking

Where talk is of authentic things, whether an authentic Rolex or an authen-

tic Vermeer, the sense that comes to the fore is spelled out in the Shorter OED:
“Real, actual, genuine; original, first-hand; really proceeding from its stated
source, author, painter, etc.” Where the source is misrepresented, the object is
denominated a fake. Concern about correspondence to a source also emerges
when speaking about the lives and actions of persons, though now typically
the source is an inner one. There are myriad ways one may be, more or less
self-consciously, “faking it” where the “it” may be a role, an experience, a per-
formance, a disposition, a belief, an emotion, or a character trait. Ultimately,
the possible contrast of what one thinks with what one does or dissembles
takes us back to an inner self that emerges unfi ltered, rather than conform-
ing to external ideals in a fake way. The notion of “doing” in the third pairing
includes but is broader than just speech acts (which are at the center of both the
first and second correspondences and contrasts).
The varieties of Faking It are very fruitfully (and amusingly) explored in
William Ian Miller’s book of that name (2003). What is the role of self-watching
in this arena? Is one inevitably faking it insofar as one is conscious of what
one is doing and deliberately aiming at effects a more spontaneous and less
self-conscious agent might not even be aware of? This is among the compli-
cations our attitudes towards our own attitudes and motivations provide in
relation to the virtues and our efforts at virtuous living. The uncertainties are
among the risks of living an examined life. The fundamental theoretical axis of
Miller’s book is the tension between role immersion and role distancing. As in
Erving Goff man, the self tends to be regarded as the sum of its public roles. But
this externalist view of the self leaves little room for the inner agent that inhab-
its, that directs, and that feels comfortable or uncomfortable within its roles.
That is the self that gets tangled in twisting deceptions and self-deceptions
even as it seeks to do the right, the appropriate, the worthy thing for the right
 Authenticity and the Examined Life 131

reasons. Both the inner and the outer self change over time, and one’s develop-
ment over time may itself seem a problem, ultimately a problem of identity.
Freud tells a joke (one of his more awkward ones) about the voicing of labor
pain by a German Baroness whose cries of pain move from “Ah, mon Dieu, que
je souff re!” to “Mein Gott, mein Gott, what terrible pains!” to a preupwardly
assimilated “Aa-ee, aa-ee, aa-ee!” It is only the last, the inarticulate primal
screams (in other tellings of the tale, the cries are in Yiddish: “Oy vay iz mir”),
that signal for the waiting doctor the need for action. Miller extracts from the
joke a picture of true identity as historical, the notion that the authentic core
is reached when the masks of later identity are stripped away and one ceases to
attempt to pass as something else than what one began as (2003, 132–140). In
the same vein, Robert Paul Wolff tells the story of an encounter in the sixties
between the very cultivated Lionel Trilling, later author of the valuable study
Sincerity and Authenticity, which we have already had occasion to refer to, and
Sidney Morgenbesser, a Columbia University philosophy professor who in his
speech and manner famously retained his New York Jewish roots. Wolff writes,
“Trilling was a New York Jewish boy who actually went to the high school [De
Witt Clinton] at which my father taught for a while before spending his entire
career at Columbia. Despite his origins, he affected a cultivated manner that,
I imagine, he thought would be appropriate in an Oxford Senior Common
Room. . . . One day, Sidney went to a cocktail party, at which he spotted Trilling
holding forth in his best Oxonian style. Sidney walked up and said, in a loud
voice, ‘Ah, Lionel. Incognito ergo sum, eh?’”4
I know of a similar encounter during the same period at Oxford itself. A very
New York Jewish friend of mine named Marshall was told, when he first came
to Oxford, that his tutor (that is, supervisor for graduate studies) at Balliol
College would be a politics don (“don” is of course the Oxford word for teacher
or professor—it does not have its usual Mafia connotations in this context)
named, if I recall correctly, Weinstein. Weinstein was in fact an American.
Also a Jew from New York who had however been at Oxford for eight years and
become very assimilated—including dropping his New York accent in favor of a
rather pompous Oxford accent. When Marshall walked into his office for their
first meeting, his words (in the heaviest New York Jewish accent he could mus-
ter) were: “Hiya, Weinstein. I unnerstan’ yur from Brooklyn.” Weinstein was
of course left speechless—what could he possibly say in reply in his acquired
bogus Oxford accent?
As I’ve indicated, these stories suggest that true identity is historical. But
is an earlier, or perhaps the earliest, self somehow more authentic? People do

4
http://robertpaulwolff.blogspot.com/2010/05/memoir-volume-two-hapter-three-ninth.html,
accessed 5/21/2011. Other aspects of identity and “faking it,” including in particular relations to
origins, are engagingly explored by Richard Todd (2008).
132 on emotions

talk about not forgetting their roots and not denying their past. But are we all,
whatever we have become, in the end, at bottom, really, our infantile selves? Is
all education and development ultimately a denial of our primitive selves? Why
should the primordial be regarded as providing the test for authenticity? Why
can’t we really be what we have become, what we (given the conditions of life)
have made ourselves? Surely we do not regard all later roles as more alien or less
congenial than all earlier roles. Or is the point that we are our most disrepu-
table, most despised and polluted selves?

The Nature of Human Nature

The issues here go beyond questions of individual identity to connect with
broader issues of the nature of human nature and the essential (see Neu 2000,
3–13). To get to the essence of his piece of wax, Descartes stripped it of its
changeable properties. To get to the essence of human beings, political theo-
rists have often imagined a “state of nature,” expecting the nature of humanity
would be revealed in such stripped-down circumstances. These thought experi-
ments generally imagine people in very extreme conditions. Hobbes strips
individuals of the authority and protection of the sovereign—the restraints
of society—and sees a war of all against all, a state of nature in which life is
nasty, brutish, and short. Locke imagines a more genteel scene: in the state of
nature, aside from a few marauding renegades, the loss of the forms of state
and of civil society would leave people pretty much as they are. Rousseau sees
a very different scene indeed. What for Hobbes are necessary protections, for
Rousseau are distorting chains. They warp the free and happy individual of the
state of nature into the miserable creature of society. The noble savage or happy
ape in his state of nature looks rather different from the scurrying and claw-
ing rat of Hobbes’s world. The bestiaries are different, the masterless men look
different, but the experiment is similar. People will often suggest that one strip
things away, that one look to extreme conditions (absence of organized society,
scarcity of food and means of life) in order to see the true nature of humans.
Typically, it is suggested that their mean egotistical motives will be revealed.
A modern version of the experiment can be found in William Golding’s novel
Lord of the Flies (1954), in which boys left to themselves revert to primitive
savagery. The stories of cannibalism in lifeboats are also occasionally cited.
Sometimes, of course, the opposite presumption is made. Mencius, in ancient
China, argued that “the tendency of human nature to do good is like that of
water to flow downward. There is no man who does not tend to do good; there
is no water that does not flow downward” (de Bary 1960, 89).
But one should note how odd the stripping procedure is. We rarely if ever
follow it in our efforts to understand other natural kinds. One does not seek
 Authenticity and the Examined Life 133

to discover the nature of a rose by observing it under arctic conditions. If one

did, one would no doubt conclude the rose to be, in its essence, really, a bare
twig with thorns. Rather, we put the rose in the best possible conditions, allow
it to flourish, and suppose we have discovered its nature only when it has been
fulfi lled. This approach is not without partisans in the history of political the-
ory. Edmund Burke suggested one could see human nature by looking at people
as they are. There is no need to pull away the decent draperies. Of course one
would discover nasty things if one scraped away the thin veneer of civiliza-
tion. But men are what they make themselves—and they make civilization.
The accretions of culture do not hide human nature, they express it. The senti-
ment that nature is artifice is perhaps echoed in Oscar Wilde’s assertion that
“the first duty in life is to be as artificial as possible.”
So we have different experiments and different conclusions even within the
same sort of experiment. One approach to the search for human nature seems
to yield basic motives, the other fulfilled form. Yet neither approach is satisfac-
tory. In stripping things away, how can one be sure that one is discarding only
the inessential or distorting conditions? (Consider the rose.) The experiment
seems to presuppose its results. (And Mencius’s argument could similarly be
readily reversed. Why not say, “The tendency of human nature to do evil is like
that of water to flow downward”?) And the other approach does not seem to
allow sense to “the essential” at all. In accepting everything as it is, nothing is
distinguished, and no allowance is made for unfulfi lled potential.
Calls for authenticity are often calls to fulfi ll potential, to become oneself.
But how are we to do that? Some have advocated that we return to origins,
that we give up the discontents of civilization in favor of the polymorphously
perverse pleasures of childhood. Some have advocated that we move beyond
good and evil and return to the presumed innocence of childhood. What must
be recognized, however, is that such moves would have their costs. Childhood
is in fact itself mixed, full of contrary and destructive instincts, so reversion
to it (without the controls of adulthood) would lead ultimately to the destruc-
tion of the supposedly liberated individuals and those around them. In a world
governed by infantile impulses combined with adult powers (and without adult
restraints) the darker side of childhood is likely to dominate. We see once more
the problem of how ideals of self-knowledge and self-expression are to be rec-
onciled with ideals of restraint and self-control for the sake of social cohesion.
The abandonment of fi lters and control might take one back to a primal past,
but it would likely already be a polluted past, troubled in ways that make the
path of self-examination and self-criticism, hard as it is, inviting and even
unavoidable.
While different balances might be reached, we may ultimately have no
choice. Given that one’s eyes have been opened, perhaps one does not have the
option to close them. Our nature (despite existentialist denials that there is
134 on emotions

such a thing until we have made ourselves through our choices and actions)
places limits on what we can be and plays a role in establishing the conditions
of our happiness. Ideals of authenticity that insist on the preservation of indi-
vidual roots or a quest for societal states of nature in effect call for a return to
early stages of development. Childhood is supposed somehow to be maximally
authentic. But while things start there, they do not end there, and critical anal-
ysis may make all the difference as to the value of the life lived.

Sexuality and Authenticity

Part of the virtue of authenticity is to get us to question the norms of the soci-
ety we happen to find ourselves living in. Questioning does not always mean
rejecting. Sometimes, perhaps often, perhaps most often (depending on what
society one is in), societal norms are reasonable, even necessary. But the point
remains that the examined life calls on one to question both oneself and one’s
social environment: one’s self in the world.
In addition to whatever universal human nature there is (the study of biol-
ogy, anthropology, psychology, history, and the like can help give us insight
into what is special to and characteristic of our species) we have our individual
natures, original even if naturally endowed features of our selves that it may
take what Mill (1861) called “experiments in living” to uncover. These are very
much a part of what ideals of authenticity call upon us to be true to, to make
manifest. These include the natural or unimpeded activities of the soul that
Aristotle links to individual happiness. And whatever our special talents and
interests, the unique gifts we may bring to the human panorama, sexuality is
something common to all yet unique to each. It provides an example of the type
of diversity that authenticity encourages and calls for conformity may stifle.
And most (though not all) forms of diversity are of value.
While some ideals of authenticity seek to escape the body and embodied
social roles, some desires, in particular sexual desires, insist that one’s true
self is embodied. When E. M. Forster says “only connect” in the famous epi-
graph to Howards End, he is talking of connecting our public, conventional
faces with our internal, transgressive erotic desires, as appears more fully
later in the book: “Only connect the prose and the passion, and both will be
exalted, and human love will be seen at its height. Live in fragments no longer.
Only connect, and the beast and the monk, robbed of the isolation that is life
to either, will die” (Forster 1921, 214). Salvation for Forster involves having
the courage to be what one is, sexually speaking, being true to one’s desiring
self. It is a message that rings through Blake and D. H. Lawrence and many
other authors, a call specially needing heeding in times of sexual repression—
which is, of course, connected with political and other forms of repression
 Authenticity and the Examined Life 135

(a point seen as far back as Pausanias in his speech in Plato’s Symposium but
emphasized in modern times by Freud and his sexual liberationist followers,
among them Herbert Marcuse and Norman O. Brown).
This aspect of authenticity makes the needs and difficulties of self-knowledge
especially salient. Freud has taught us about the unconscious and the multiple
forces of repression, manifested often in inner conflict, making access to self-
understanding and (undistorted) self-expression a difficult achievement—not
simply a matter of letting go or acting out. And once issues of self-knowledge
and self-understanding are dealt with, obstacles to self-expression remain. The
limits to freedom marked by Mill’s “harm to others” principle of course apply to
sexual as well as other activities (so rape, child molestation, and the like may be
prohibited and punished by the state, no less than robbery, assault, and other
harmful activities, despite whatever would-be perpetrators’ desires might be
pressing for fulfi llment in both sexual and nonsexual cases). But beyond that,
even in America, the land of liberty and authentic self-expression, there have
been social and legislative attempts to deny and contain. One striking instance
is the military policy of “Don’t Ask, Don’t Tell,” which for an extended period
(now happily over) insisted that those who would serve in the military neither
reveal nor act on homosexual or lesbian or other disfavored desires. A homo-
sexual or lesbian could serve in the military on the explicit condition that they
kept their desires secret (they had, of course, long served precisely by keep-
ing everything inexplicit). The military, by not asking, was supposed to help in
that effort. But such an enforced alienation of one’s personal from one’s profes-
sional life, even if it allowed clandestine action on one’s desires, had its costs.
There were, aside from the cost to the nation of the services of those who could
not or would not keep their sexual inclinations secret, the costs of individual
hypocrisy, suppression, alienation, and fear to those who sought to stay in the
military by conforming to the mandate, and the costs to society of maintaining
invidious and discriminatory attitudes towards a class of its citizens. It was the
legislation of inauthenticity. It enacted in the public sphere the sort of conflict
and mixed principles that fight within individuals against living a truthful life.
“Only connect.”

Authentic Values: The Source of Value

What kind of life is most worth living cannot simply be read off the nature of
human nature, nor can it be discovered by direct introspection of one’s per-
sonal desires. We may have reason to value authenticity in its various forms
and to avoid lying, hypocrisy, and faking in their various forms. But unthink-
ing spontaneity is unlikely to yield an admirable life or ultimate satisfaction—
there is simply too much else that we value.
136 on emotions

Some would claim that a commitment to authenticity at least secures one

from embarrassment and shame. In today’s shameless society, that might
appear an unalloyed gain. But that would be to dismiss the many ideals that
we noted at the beginning, ideals that sometimes go against unconstrained
self-expression. After all, shame and embarrassment, like many other
self-reflexive emotions, are the affective embodiment of such ideals, giving
felt significance to the transgression of boundaries and the falling short of
aspirations.
In a New York Times article on plagiarism in 2010, it was reported that
“Helene Hegemann, a German teenager whose best-selling novel about Berlin
club life turned out to include passages lifted from others,” did not apologize
when her dishonesty was uncovered. Instead, she insisted, “There’s no such
thing as originality anyway, just authenticity.”5 When the stealing of oth-
ers’ words and presenting them as one’s own is defended as “authentic,” one
must wonder what the standards of authenticity have become. Is it enough
that one agrees with another’s words—the match now shifting from what
you say corresponding to what you think to what you think corresponding
to what someone else has said? Even such a loose standard distorts insofar
as it fails to acknowledge or denies the original source (the correspondence
between what you say and what someone else has said gets occluded). It
turns copying from a form of fraud or faking into self-expression. Explicit
quotation used to be a way of both signaling agreement and showing respect.
Unacknowledged quotation (where the intent is to deceive and appropriate
authorship) in claiming authenticity overturns honor and degrades both
respect and self-respect.
Getting back to sources—how is it to be done? One straightforward reading
of Polonius’ claimed necessary or inevitable connection (“it doth follow, as the
night the day”) between being true to oneself and being true to others is that
truth (whether to self or others) amounts to the refusal of masks (“the absence of
dissimulation or feigning or pretence”—Trilling 1971, 13). What we have seen,
however, is that spontaneity does not ensure transparency. It may be enough to
avoid deliberate deceit, but self-deception and hidden forces may still cover and
conceal the truth, masking it from both self and other. To get to the true self, an
examined life is required. And even then, simply acting on what one truly feels
may not be enough for a good life because, whatever the virtues of authenticity,
unconstrained self-expression is not the only thing of value.
While Kantian morality specifically excludes qualities of feeling and tem-
perament that are not under an individual’s control from the realm of moral
value and judgment (all that matters is what one wills, one’s intentions and
what one tries to do), virtues and vices often center on just such matters

5
http://www.nytimes.com/2010/08/02/education/02cheat.html?hp, accessed 8/2/2010.
 Authenticity and the Examined Life 137

of feeling and temperament. As Thomas Nagel puts it: “A person may be

greedy, envious, cowardly, cold, ungenerous, unkind, vain, or conceited, but
behave perfectly by a monumental effort of will. To possess these vices is to
be unable to help having certain feelings under certain circumstances, and
to have strong spontaneous impulses to act badly. Even if one controls the
impulses, one still has the vice. An envious person hates the greater success
of others. He can be morally condemned as envious even if he congratulates
them cordially and does nothing to denigrate or spoil their success” (Nagel
1979, 32–33).6 Even if Kantian commitments keep us from condemning bad
character accompanied by good action as morally defective, can the commit-
ments of authenticity elevate our evil dispositions and feelings into aspects
of the self worthy of outward expression? Does the difference between vir-
tues and vices of character cease to make a moral or any consequential dif-
ference? Should we be “true” to all aspects of our selves, even those aspects
which might properly be regarded as despicable? There have been times when
some have seen sincerity as a redeeming virtue in those who act on their
basest impulses (“At least Hitler did what he truly believed in doing . . . ”),
but it is difficult to see why. Are feelings and impulses to be regarded as
valuable just because they are ours? Is uncritically accepting and acting on
all our inclinations in itself some kind of strength of character? The eleva-
tion of the arbitrary (not to say, once again, the despicable) is, in the end, an
abandonment of moral value and, ultimately, the offering of an incoherent
substitute.
The problems of giving ultimate value to self-fulfi llment, self-choice,
or (as some would have it) self-indulgence, go beyond the fact that we
have—as a matter of fact—multiple values and that some of these may call
for self-restraint or otherwise confl ict with our unhindered self-regarding
self-expression. There is a conceptual issue built into the very notion of
value. As Charles Taylor argues, significance has conditions of intelligibility
and certain background assumptions are needed for particular valuations to
be intelligible: “The ideal of self-choice supposes that there are other issues

6
Just as fl aws in a crystal can make latent structure manifest, vices can be more revealing
than virtues. It has been argued that stable virtues such as honesty do not exist. Th at is, that
character traits are not robust and predictive. Instead, they are rather narrowly circumscribed
and highly situation dependent: an individual may be honest in business dealings but not
in intimate relations, charitable when in a good mood, but oblivious to the needs of others
when even slightly piqued or in a hurry (see Doris 2002). On the contrary, however, I believe
that whatever may the case about honesty and similar virtues, bad character is highly stable,
robust, and predictive. We can know with confidence that the ill-inclined will behave char-
acteristically, given half the chance. Just as alcoholics will (perhaps even sincerely) express
regret for their lapses, time after time, and yet do it again, the mean-spirited, conniving,
nasty, and venal will regularly behave as expected. Th e stability of bad characteristics may
help explain why the polluted and primal are so often seen as revealing the authentic self.
138 on emotions

of significance beyond self-choice” (1992, 39). If all that mattered were

self-determination, any arbitrary valuation might stand (my goal in life is
to keep my toes curled or to wax my toothpicks to a high gloss), but it is only
against a background of things that are assumed to matter that certain valu-
ations can make sense.
Self-realization as an ideal often amounts to an amalgam of relativism and
individualism, what Taylor calls “the individualism of self-fulfi lment”:

Everyone has a right to develop their own form of life, grounded on

their own sense of what is really important or of value. People are
called upon to be true to themselves and to seek their own self-fulfi l-
ment. What this consists of, each must, in the last instance, determine
for him- or herself. No one else can or should try to dictate its content.
(1992, 14)

Even on this view, it must be admitted that some individuals may value group
or communitarian solidarity and achievement. But more strongly, as Taylor
argues, “there is something self-defeating in a mode of fulfi lment that denies
our ties to others” (41). This is not just because our identities and values are
formed in dialogue and relation to others, but because of the constraints on
intelligibility that put significance and arbitrariness in confl ict. The problem is
ultimately one that Socrates brings out in his dialogue with Euthyphro about
piety, about turning to the gods as the ultimate source of moral authority (esp.
Euthyphro 10d). Assuming one could get around questions concerning the exis-
tence of God and problems of knowledge (whether there be many gods, who
might conflict, or one God, who might be taken to offer different revelations
open to varying interpretations), why follow an authority that follows no stan-
dard but its whims? Without reasons, divine injunctions (assuming we could
know them) are arbitrary. And the problem persists if one substitutes oneself
for God or the gods as the source of value. Why follow your own thoughts if
they have no argument or support beyond the mere fact that you happen to
think them? Why should you follow arbitrary thoughts even when they happen
to be your own?
Sartre has a notion that values are inevitably arbitrary, that we must always
choose without guidance from standards, and that a chosen option “has value
only because it is chosen” (Sartre 1957 [1946], 21). But Sartre’s notion that
we create value through our choices, that our choices are exemplary, is in fact
an incoherent extrapolation from Kant’s emphasis on universalization in his
Categorical Imperative. While Sartre insists that our choices are exemplary,
that when we choose we choose for all mankind, he is unable (unlike Kant) to
explain why, and the doctrine is in fact incompatible with his own insistence
on the pervasiveness and anguish of freedom.
 Authenticity and the Examined Life 139

Sartre is concerned with the burden of deciding in a world without preexist-

ing moral guideposts:

Man is forlorn, because neither within him nor without does he find
anything to cling to. . . . If existence really does precede essence, there
is no explaining things away by reference to a fi xed and given human
nature. In other words, there is no determinism, man is free, man is
freedom. . . . If God does not exist, we find no values or commands to
turn to which legitimize our conduct. So, in the bright realm of val-
ues, we have no excuse behind us, nor justification before us. We are
alone, with no excuses. . . . Man is condemned to be free. (Sartre 1957
[1946], 22–23)

The burden is particularly heavy because, for Sartre, our choices have wider
implications: “When we say that a man is responsible for himself, we do not only
mean that he is responsible for his own individuality, but that he is responsible
for all men” (16). Our choices create universal values: “In creating the man that
we want to be, there is not a single one of our acts which does not at the same
time create an image of man as we think he ought to be” (17). If all this meant
was that good reasons must be general, it would be acceptable. Differences in
principles must indeed (if they are to be moral) be based on relevant differences
in persons. But it is simply false that “nothing can be good for us without being
good for all”—as Sartre goes on to say (17). And Sartre means this sort of uni-
versalization quite literally:

If I want to marry, to have children; even if this marriage depends

solely on my own circumstances or passion or wish, I am involving all
humanity in monogamy and not merely myself. Therefore, I am respon-
sible for myself and for everyone else. I am creating a certain image of
man of my own choosing. In choosing myself, I choose man. (18)

Serious choices produce anguish because we are legislating for all mankind,
and that is a vast responsibility. Sartre is, in a way, picking up on the Kantian
idea of the will legislating universal law in all its maxims, but Sartre’s view
is an upside-down and incoherent transformation of Kant’s theory of moral
value. It misunderstands the point of universalization. For Kant, asking
whether the principle of one’s action could be a universal law (a version of
the question, “What if everyone did that?”) is a test of the moral character of
one’s action, in particular, it is a test of whether one is making an exception
of one’s self (what Kant mockingly refers to as the “dear self”). The concern
is with fairness, with justice. If the principle of one’s action cannot coher-
ently withstand the test of universalization and serve as a rule for all (based,
140 on emotions

ultimately, on rationality), it is not a moral rule. For Kant, we must univer-

salize before we decide (it is a matter of testing for fairness). For Sartre, we
universalize after we decide.
For Kant, moral laws are a priori and unconditional, fi xed in human nature,
the nature of reason. People do not invent their own moral laws, they sim-
ply adopt the existing moral law for themselves—the moral law is not arbi-
trary, reason fi xes its content. For Sartre in his existentialist mode, people
make their own morality, simply choose their principles, without any guidance
or standards. There is no human nature or objective moral order. Sartre tries
to preserve Kant’s notion of moral autonomy (legislating for oneself) while
separating it from commitment to a moral law (independent of oneself and
determined by reason). He wants to insist on the anguish of legislating freely,
without guidance, for all mankind—the burden of a vast responsibility. But
since every member of mankind must legislate, since each person carries that
responsibility for himself, it is not clear how anyone else can carry it for another
individual and so have a wider burden. And it remains a mystery, on Sartre’s
account, why one must universalize at all.
Sartre insists all acts are exemplary: “At every moment I’m obliged to per-
form exemplary acts. For every man, everything happens as if all mankind
had its eyes fi xed on him and were guiding itself by what he does” (20). But
why? Sartre insists that it is bad faith to deny our responsibility. But I believe
that the notions of responsibility and choice have become empty in Sartre’s
universe—far from committing all mankind by our choices, we do not commit
even ourselves. For Sartre, universalization is a gesture made after one has
made one’s subjective unconstrained choice, and yet it should be clear that
one’s choice does not bind even oneself. For it must be recalled that one is con-
demned to be free always and everywhere. When the time for action comes,
one cannot regard oneself as bound by any prior conditions (that would be bad
faith), whether they are the demands of one’s upbringing, the requirements
of the state, or even one’s own prior choices. Thus, one must wonder what
the force of a “choice” that does not bind even oneself in the future is sup-
posed to be. If one is truly condemned to be free, it appears even choice cannot
constrain later choice (one must choose again at the later time, and one may
change one’s mind even if circumstances have not changed), and so “choice”
becomes empty. It is hard to regard acts as exemplary, as creating standards
binding on all mankind, if they do not bind even the chooser himself in the
future. And it is difficult to see how they can, if his future decisions are to be
truly free, that is unconstrained by values (even those he once chose). The slo-
ganeering doctrines that man is condemned to be free and that each man’s acts
are exemplary and he is responsible for all mankind are ultimately incompat-
ible. And the notion that you should universalize after you decide, in a world
with no prior guidance or standards, neither reason nor fairness, is without
 Authenticity and the Examined Life 141

significance or sense. The value of authenticity, of being true to one’s stan-

dards when those standards are themselves arbitrary, remains unmoored.

Authentic Emotion
We have been considering the authenticity of selves, desires, and values, among
other things. Let us conclude by focusing on emotions. What does it take for
an emotion to be true, or real, or authentic? Sometimes the authenticity of a
passion, as of a desire, is connected to its stability (as when Pausanias in Plato’s
Symposium contrasts long-lasting Heavenly Love based on moral and spiritual
aims with the fleeting Common Love that is typically based on passing physi-
cal attraction). But of course, phobias and other unrealistically based, and so in
a sense “false,” passions can be remarkably persistent. Pervasive moods, good
and bad, or long-term anger, boredom, and other affective attitudes need not
be more authentic or otherwise valuable than briefer episodes of emotion. As
before, looking to broader and perhaps deeper correspondences and contrasts
can be clarifying. What is the flaw that makes an emotion inauthentic? While
it might fall under the general rubric of “faking it,” there is always the possi-
bility of pretense or make-believe, and there is an interesting suggestion that
the possibility is inevitably realized when the object of emotion is fictional.
Is pretense simply a matter of intentional mismatch of outward expression
and inner feeling? That is doubtless too simple. (Some of the complexities are
spelled out by J. L. Austin in “Pretending,” 1970.) Are the conditions different
for actors and audience at a play?
Certainly the actors in a play typically strive to appear to have whatever emo-
tions are required by the script. Method actors may actually work themselves
up into a state, the thought being that the best way to put on the face of anger
or fear is to feel real anger or fear, working from the inside. There are counterar-
guments, going back at least to Diderot (1773), that the most effective pretense
can be achieved precisely by not being in the state affected, that inner turmoil
may not yield the appropriate outward appearance, especially not night after
night in the theatre and especially not if rapid changes of emotion are called
for. To appear to be in a certain state, it may sometimes be best not actually to
be in that state—that is the “Paradox of Acting.” But whatever the technique
needed to produce the appropriate observable expression for an emotion, it is
the representation rather than the reality that counts in the theatre, and even
where the technique used is Stanislavski’s, the actor’s emotion is not true or
fully real. Because of the context, the occasion for the feeling arising (if it does),
the realistic expression of emotion does not amount to authentic emotion. The
emotion is called up for a purpose—call it an artistic purpose—and in life an
emotion called up for a purpose is not an authentic emotion, it is a pretense
142 on emotions

or (to use the more precisely relevant and telling label) acting. (At its root, its
Greek origin, “hypocrisy” is acting and a “hypocrite” is an actor.) Whether the
purpose is artistic, as in the theatre, or commercial, as with the forced smile of
a fl ight attendant, an ulterior motive undermines authenticity.
Of course, Sartre’s theory of emotion holds that all emotions serve a pur-
pose, are chosen in an attempt, typically an ineffective and in that sense a mag-
ical attempt, to change the world when circumstances are too difficult to deal
with realistically. That might make all emotion a pretense, a kind of playacting.
We don’t cry because we are sad (as most people would suppose most of the
time), and we aren’t sad because we cry (as William James’s theory of emo-
tion as the perception of physiological change would have it), but we are sad
and cry because it may save us from having to talk in a difficult situation or
otherwise give us a path through a situation we have trouble dealing with (at
least that is Sartre’s purposive view of emotion). It is part of Sartre’s general
insistence that we are condemned to be free, that we have more choices than
we wish (in our bad faith) to admit to. While sometimes an illuminating way to
look at emotions, as a general theory of emotion it has serious flaws. In addi-
tion to underplaying the passive and overwhelming aspects of emotional expe-
rience, it seems unable to account for happy emotions in response to happy
circumstances. And while fainting may sometimes be a magical attempt to
escape danger (or embarrassment or other unpleasantness), running out of fear
need not be, as Sartre would have it, “a fainting which is enacted.” Running
may sometimes be an effective way to get out of harm’s way. (See Sartre 1948
[1939], 63; Neu 2000, 25–32; Solomon 2003; Neu 2005.) Even where emotional
expressions are clearly chosen, that is to say, deliberate, the emotion that is
being expressed need not have been deliberately chosen. Actors in the theatre,
choosing what to feel (if they do), are an exception.
What of the audience? Kendall Walton has argued that when the object of
an emotion is known to be fictional, the emotion cannot be real. Fear at a hor-
ror movie, despite whatever physiological upset may take place, is, according
to him, make-believe fear. The inauthenticity here is not a matter of ulterior
motives, but (Walton argues) missing conditions of real fear. What are those
conditions?
We might start with the physiological upset that Walton allows may be pres-
ent. On one model of pretense (the doubtless too-simple model mentioned a
moment ago), that is precisely what is missing when emotion is pretended to:
the outward shell is present without the inner feeling. There are at least two
problems with focusing on the felt physiological upset as what is missing in
fear of fictions. First, as just noted, physiological upset may in fact be felt in
abundance while watching a horror movie. Indeed, that may be the attraction
of horror movies for some, just as it surely is for the thrill seekers who ride
roller coasters. Such rides offer the adrenalin rush of fear without having to
 Authenticity and the Examined Life 143

confront real danger. (While roller coasters may occasionally fail and produce
real accidents with real injuries, few riders desire or anticipate that.) Walton
dismisses such physiological upset (in the case of Charles at a horror movie,
“his muscles are tensed, he clutches his chair, his pulse quickens, his adrena-
lin flows”—1978, 6) as merely “quasi-fear” because Walton thinks other con-
ditions for real fear are absent. But there is a second problem with regarding
physiological upset as essential to real emotion. While it might seem a tautol-
ogy to say that emotions are feelings, that depends on using “feelings” in a
broad sense (where the equivalence may well hold). But the Feeling Theorists
who insist that emotions are feelings don’t mean to be making an empty claim,
they typically use “feelings” in a narrow sense, where it refers to bodily sensa-
tions, that is, felt physiological upset. And the problem with Feeling Theory in
that sense, as has long been argued, is that it is simply false that each emotion
has a unique and distinctive set of bodily sensations. The differences among
(for example) regret, remorse, embarrassment, guilt, shame, and other emo-
tions in that unhappy range are not to be found in characteristic sensations
but in the situations that provoke them and the thoughts that constitute
them. There are many other problems with Feeling Theory, but the point to see
here is that real fear does not require feelings of fear, any more than any emo-
tion is necessarily characterized by peculiar and unique sensations. My fears
concerning (say) nuclear proliferation, global warming, and the thwarting of
democracy (at home and abroad) may be embodied in inclinations or disposi-
tions to action, without physiological upset. Such “calm passions” point us to
a second essential element of real emotion that might seem to be missing in
fear of fictions.
Walton insists that the person gripped by quasi-fear sensations as
all-devouring slime appears to be coming towards him from the movie screen
lacks the inclinations natural to real fear. He “does not have even an inclina-
tion to leave the theater or call the police” (1978, 8). But that need not be
so: fi rst because the range of natural inclinations may be wider than Walton
acknowledges; and second because the inclinations may be present but inhib-
ited. On the fi rst point, there is no short list of fear behavior or anger behav-
ior because, given appropriate context and beliefs, just about any behavior
could express just about any emotion. (For one example, I might punch you
to express not my anger, but my gratitude: perhaps my gratitude to another
party who hates you, or perhaps to you because you derive pleasure in strange
ways or at any rate because I believe you derive pleasure in strange ways.)
Even leaving out such contingent beliefs, an inclination to freeze in place
may be as natural to fear as an inclination to flee. And the absence of observ-
able action need not signify the absence of inclination. Walton himself allows
that a person with an irrational fear of flying nonetheless has a real fear.
Such a person might even get on planes, though “if he must board a plane he
144 on emotions

becomes nervous and upset” (1978, 8–9). I don’t see why the person gripped
by a horror movie could not be frozen in place with fear or, alternatively, be
inclined to flee or scream but inhibit the inclinations precisely because he is
aware that he is watching a movie and that the slime is not real.
Walton suggests that deliberate screaming, as opposed to a spontaneous
shriek, would only confi rm the pretended character of the fear (1978, 8 n.
4). But this may undermine his picture of the horror movie viewer, whom he
calls Charles, as “an actor impersonating himself,” like a child playing a game
of make-believe with his father who pretends to be a monster whom the child
flees screaming yet also smiling. Such a child is not really afraid (1978, 13,
15–16). But in a truly comparable situation, where Charles was playacting
fear for a girlfriend in the next seat, his screams and clutches really would be
a pretense. When Charles alone screams, or is inclined to scream but inhibits
the inclination, he is not performing for himself. He is precisely not imper-
sonating himself, dissembling his real feelings, or pretending. But now we
must confront the beliefs of the physiologically upset or quasi-fearful viewer,
for Walton emphasizes the absence of real belief in danger as the crucial
missing feature in fear of fictions, the feature that keeps such emotions from
being real.
The thought of danger is no doubt essential to fear, no less than the thought
of loss is essential to jealousy. Such thoughts need not amount to beliefs,
but they must be present in some form and at some level if the distinctive
distinguishing features of those particular emotions are to be present. The
difference between beliefs and thoughts (both of which play explanatory as
well as phenomenological roles in our self-understanding, and so need not
be explicitly conscious) may be enough to undermine Walton’s position (Neu
2005). Coleridge’s “willing suspension of disbelief” may be enough to put us
as much in touch with the worlds of fiction as we are with past worlds and
distant worlds. There is no reason to doubt the reality of our feelings (in the
broad sense) for past events or distant peoples. Immediate presence is not
required. Nor is single-minded full-blown belief. As we have noted, Walton
himself allows that a person’s irrational fear of flying may be real despite self-
conscious awareness of the irrationality of the underlying belief. He writes of
a person who hates flying:

He realizes, in one sense, that airplanes are (relatively) safe. He

says, honestly, that they are, and can quote statistics to prove it.
Nevertheless, he avoids traveling by air whenever possible. He is bril-
liant at devising excuses. And if he must board a plane he becomes
nervous and upset. I grant that this person believes at a ‘gut’ level that
flying is dangerous, despite his ‘intellectual’ belief to the contrary. I
grant also that he is really afraid of flying. (Walton 1978, 8)
 Authenticity and the Examined Life 145

If a bifocal vision of flying as dangerous and not-really-dangerous is enough for

an admittedly irrational fear of flying to be a real fear, why shouldn’t a similar
bifocal vision of movie slime as dangerous and not-really-dangerous be enough
for fear of an acknowledged fiction to be a real fear? In both cases the evidence
for a bifocal vision can be found in a mixture of deliberate actions, inclinations
to action, and physiological upset—in Charles’s case, the symptoms of what
Walton calls quasi-fear, “his more or less automatic, nondeliberate, reactions:
his pulse rate, his sweaty palms, his knotted stomach, his spontaneous shriek”
(1978, 8). Whatever the thoughts required for fear, they may be present when
objects are at a temporal, physical, or even fictional distance. One may fear for
oneself in the face of horrific movie dangers or for a made-up character in a
book or other medium in whose story one is caught up.
Even Sartre, for whom emotion is a magical attempt at transformation of
the world, says, “True emotion is . . . accompanied by belief”—with the “serious-
ness” of belief marked by the degree of physiological upset (1948 [1939], 73).
He also speaks of “sincere emotion,” presumably with similar criteria in mind.
But suspension of disbelief or letting oneself go as one gets caught up in a story,
bifocal vision or conflicted thoughts may be sufficient belief (Walton calls it “gut”
belief) for real emotion. Emotions may, of course, be inappropriate or irrational
in myriad ways, as when one is afraid of nonexistent objects, such as ghosts, or
angry at a person who has done one no wrong and caused one no harm, or proud
of what is in fact a despicable act. But in such cases one typically is in no way
aware that one’s belief is false, and no one would claim that emotions founded
on such errors are themselves necessarily inauthentic or otherwise less than
fully real.
This is not to say that our state of mind is whatever we think it is. Our beliefs
about our own state of mind, like our beliefs about the external world, may be
mistaken or confused—reflexive knowledge is not infallible even if it may be
in some ways privileged (that is, we may have forms of access to what we think
that are unavailable to others) and there is importantly room for self-deception,
or what Sartre would call “bad faith,” about what we feel. Indeed, we are some-
times the last to know what we feel, as when our friends inform us, despite
our obliviousness or even our denials, that we are in love or jealous. And such
flaws in our self-understanding can give a sense to the falsity, insincerity, or
inauthenticity of emotion that is different from simple lying, and different too
from the many other ways in which we have considered that selves, or social
roles, or states of mind in general may fall short—whether because they are
undesired, unchosen, or disliked, distant from our roots or origins, or place us
in a better light than we deserve (or think that we deserve). And in all cases and
in all senses, the surest route to the self-understanding that truth, sincerity,
and authenticity require includes self-examination. So far as those are of value,
we have yet further reason for believing that the unexamined life is not worth
146 on emotions

living. (Th is is not to deny that some illusions may themselves be valuable,
perhaps even necessary.)
It might seem that if we can describe an emotion based on an irrational
belief as an irrational emotion, we can equally describe an emotion based on
make-believe as a make-believe emotion. But it is a false parallel, at least when
questions of the realness of the psychological state come into the picture. An
emotion based on irrational belief remains real, and an emotion based on
make-believe does too. The object of an emotion may not exist, but that does
not undermine the claim to existence of the psychological condition. As we
have seen, following Austin, the criteria for “reality” ultimately depend on the
type of thing one is talking about and, in particular, on the relevant excluded
contrasts. When the thing talked about is an emotion the gaps and defects
pointed to by claims of unreality, inauthenticity, pretense, and the like have to
do with the constituents of emotion and the context for their occurrence. And,
as I have argued, those constituents are as much in place when we talk about
underlying make-believe thoughts as when we talk about underlying irrational
thoughts.

The Project of Authenticity

There is another level of difficulty, beyond moral and societal constraints, as
one strives for sincerity and authenticity in one’s life. Just as there is a risk of
pride in the pursuit of humility, hubris may trip and trouble a commitment
to sincerity. There is grave danger of self-deception in remorseless truth-tell-
ing that emerges in Molière’s Alceste in Le Misanthrope (Trilling 1971, 17), in
Ibsen’s Gregers Werle in The Wild Duck, and in O’Neill’s Hickey in The Iceman
Cometh (Neu 2000, chap. 15); the motives of would-be undeceivers such as
Gregers and Hickey are often shielded under the banner of pure allegiance to
the truth in the relentless examination and unmasking of others, and need
self-examination to be revealed. Sartre asserts the project of sincerity is inevi-
tably entrapped and entangled in the pitfalls of bad faith. He sees in the ten-
sion of facticity and transcendence the impossibility of a project of sincerity:
“What can be the significance of the ideal of sincerity except as a task impos-
sible to achieve, of which the very meaning is in contradiction with the struc-
ture of my consciousness” (1956 [1943], 62). For Sartre, there can no more be
a stable self than there is a given human nature, we are and cease to be our
ever-evolving acts as we move into an open future. Certainly, authenticity and
overcoming self-estrangement or alienation is difficult. Further, as Andre Gide
has said, “one cannot both be sincere and seem so” (quoted in Trilling 1971,
70) There is a problem in both pursuing the project of sincerity and project-
ing sincerity, putting one’s effort on display risks becoming the playing of an
 Authenticity and the Examined Life 147

alienated role. Miller’s complexities of self-consciousness come into play. But

we have seen that the vices of inauthenticity are many and various, includ-
ing: lying, hypocrisy, faking, denying roots, conformism, and pretense. And
the virtues of authenticity can be found in the efforts at creative self-making
and critical questioning of self and society that make for and are made by an
examined life.

Acknowledgment
I dedicate this essay to the memory of Bob Solomon. I believe it is very much in
his spirit, even when my views do not coincide with his.

References
Austin, J. L. 1962. Sense and Sensibilia (Oxford: Oxford University Press).
Austin, J. L. 1970. “Pretending,” in Philosophical Essays, 2d ed., ed. J. O. Urmson and
G. J. Warnock (Oxford: Oxford University Press), pp. 253–271.
de Bary, Wm. Theodore, ed. 1960. Sources of Chinese Tradition, Vol. 1 (New York: Columbia
University Press).
Diderot, D. 1957 (1773). The Paradox of Acting, trans. Walter Herries Pollock (New York: Hill
and Wang).
Diderot, D. 1966. Rameau’s Nephew, trans. L. W. Tancock (Harmondsworth: Penguin Books).
Doris, John. 2002. Lack of Character: Personality and Moral Behavior (Cambridge, UK:
Cambridge University Press).
Forster, E. M. 1921. Howards End (New York: Alfred A. Knopf).
Golding, William. 1954. Lord of the Flies (New York: Coward-McCann).
Guignon, Charles. 2004. On Being Authentic (London: Routledge).
Hampshire, Stuart. 1972. “Sincerity and Single-Mindedness,” in Freedom of Mind and Other
Essays (Oxford: Oxford University Press), pp. 232–256.
Laing, R. D. 1960. The Divided Self (London: Tavistock).
Mill, J. S. 1961 (1861). “Utilitarianism,” in The Philosophy of John Stuart Mill, ed. Marshall
Cohen (New York: Modern Library).
Miller, William Ian. 1993. Humiliation (Ithaca, NY: Cornell University Press).
Miller, William Ian. 2003. Faking It (Cambridge, UK: Cambridge University Press).
Nagel, Thomas. 1979. “Moral Luck,” in Mortal Questions (Cambridge, UK: Cambridge
University Press), pp. 24–38.
Nehamas, Alexander. The Art of Living: Socratic Reflections from Plato to Foucault (Berkeley:
University of California Press, 1998).
Neu, Jerome. 1988. “Divided Minds: Sartre’s ‘Bad Faith’ Critique of Freud,” Review of
Metaphysics 42: 79–101.
Neu, Jerome. 2000. A Tear Is an Intellectual Thing: The Meanings of Emotion (Oxford: Oxford
University Press).
Neu, Jerome. 2005. “Robert C. Solomon’s Not Passion’s Slave: Emotions and Choice,” Philosophy
and Phenomenological Research 71: 741–744.
Nietzsche, Friedrich. 1967. Basic Writings of Nietzsche, trans. and ed. Walter Kaufmann
(New York: The Modern Library).
Plato. 1997. Plato Complete Works, ed. J. M. Cooper (Indianapolis: Hackett).
Ryle, Gilbert. 1949. The Concept of Mind (London: Hutchinson).
148 on emotions

Sartre, Jean-Paul. 1948 (1939). The Emotions: Outline of a Theory, trans. B. Frechtman (New York:
Philosophical Library).
Sartre, Jean-Paul. 1956 (1943). Being and Nothingness: An Essay on Phenomenological Ontology,
trans. Hazel E. Barnes (New York: Philosophical Library).
Sartre, Jean-Paul. 1957 (1946). “Existentialism Is a Humanism,” translated as
“Existentialism.” In Existentialism and Human Emotions, trans. Bernard Frechtman (New
York: Philosophical Library).
Shakespeare, William. 1969. The Complete Pelican Shakespeare, ed. Alfred Harbage (Penguin
Books).
Solomon, Robert C. 2003. Not Passion’s Slave: Emotions and Choice (Oxford: Oxford University
Press).
Taylor, Charles. 1992. The Ethics of Authenticity (Cambridge, MA: Harvard University Press).
Todd, Richard. 2008. The Thing Itself: On the Search for Authenticity (New York: Riverhead
Books).
Trilling, Lionel. 1971. Sincerity and Authenticity (Cambridge, MA: Harvard University Press).
Walton, Kendall. 1978. “Fearing Fictions,” Journal of Philosophy 75: 5–27.
Williams, Bernard. 2002. Truth and Truthfulness (Princeton, NJ: Princeton University Press).
Wittgenstein, Ludwig. 1953. Philosophical Investigations, trans. G. E. M. Anscombe (Oxford:
Basil Blackwell).
 7

Self-Knowledge and the Affi rmation of

Love
Laurence Thomas

Introduction
For all the talk about human beings being rational creatures and for all the talk
about game theoretic models intended to mirror how human beings at their
best think, there is the simply reality that nothing makes a difference for the
better in the life of a human being like the object of love. It is not just that we
need love at the beginning of our lives, when we are utterly helpless, in order
to flourish, but it is also the case that love makes a most profound difference
for the better even after we are well into adulthood. It is this insight that so
marvelously animates the work of Robert C. Solomon. While he has lots to be
say about romantic love and what he has to say is extraordinarily rich, I shall
focus in this essay upon love in general. I believe that what I shall say about
love generally serves only to underwrite Solomon’s general views about love,
including romantic love. I shall begin with a discussion of a topic that seems
rather removed from love, namely, self-knowledge. Indeed, my fundamental
argument shall be that self-knowledge does not replace the genuine affirma-
tion that comes from another. And love at its best is none other than a most
majestic form of affirmation in which the self-knowledge that we have worth
in the eyes of another is so wondrously secured.
The following words are written by Solomon in his work Love, Emotion, Myth
and Metaphor:

Love can be understood only “from the inside,” as a language can be

understood only by someone who speaks it, as a world can be under-
stood only by someone who lives in it.

The arguments of this essay shall be ever so true to this insight.

149
150 on emotions

Self-Knowledge
Part I
We can very simply define self-knowledge as having clarity with regard to the
attributes, be they good or bad or neutral, that apply to one. We are born with
a complete lack of self-knowledge, and, if all goes well, we acquire a fountain
of self-knowledge in the course of living. To be sure, with advanced age there
can be a decline in self-knowledge on the part of an individual whose life, at
an earlier point in time, was marvelously fi lled such knowledge. So, as these
considerations imply, self-knowledge is not an all-or-nothing matter. On one
hand, it is extremely difficult to imagine a psychologically healthy adult person
having no self-knowledge at all. On the other, a human being would be rather
like a god were she or he to have complete self-knowledge.
There are no restrictions to the content of self-knowledge. It can pertain to
the great strengths that a person has or to the horrendous moral defects that
are definitive of a person’s character. Or, by contrast, self-knowledge may per-
tain to a plethora of inabilities and unseemly emotional configurations, and
so on that are characteristic of a person’s life. For most individuals, of course,
self-knowledge is some mixture of these two extremes. Typically, a person
grasps that she or he has some wonderful abilities and traits, on the one hand,
and that there are also inabilities and undesirable traits, on the other, that
hold true for her or him. No reasonable person can suppose that the deliver-
ances of self-knowledge for her or him would be none other than excellences
on all fronts. Indeed, precisely what we know is that people who can be most
admirable in one area of life can be open to serious criticism in another area
of life.
In general, having self-knowledge is a good thing, for it is typically the case
that knowing both our strengths and limitations serves us well. After all, the
self-knowledge that one has certain shortcomings can enable one to take the
necessary precautions to avoid situations in which one is vulnerable or situ-
ations in which one will behave in unacceptable ways. And, of course, self-
knowledge can be a stepping stone to a variety of unseemly behaviors because
an individual knows that she or he has an asset that most people lack. For
instance, there are people who are known for exploiting their good looks, and
thus the knowledge that they appear particularly attractive physically in the
eyes of others. But to state the obvious: people can exploit their good looks in
different ways and so be open to different kinds of criticism. One person might
use his good looks to facilitate his lying effectively and so exploiting others in
effect; whereas another might exploit her good looks simply as a way of being
the center of attention. Generally, the first case will be more objectionable mor-
ally than the second one.
 Self-K nowledge and the Af fir mation of Love 151

Part II
The matter to which I shall now turn is how do human beings come to have
self-knowledge. The most obvious answer, of course, is through the experience
of doing things. There is no better evidence that one can perform a given activ-
ity or task than precisely the fact that one does just that. Now, while there is
no gainsaying the importance of experience a caveat is in order. This is because
so much of what we do is subject to the assessment of various others—our
social audience, if you will. There is one social audience for, say, our behavior
as a lecturer; typically a different social audience for our behavior at a bar; and
yet another social audience for our behavior as the member of an audience at a
distinguished event such as the recipients of the Nobel Prize. There is the social
audience of our friends as well as the social audience of our family. Finally, the
general public can also serve as a social audience.
Generally, in the absence of either verbal or non-verbal disapproval of our
behavior, then there is in effect tacit of approval of our behavior as having met
the basic standard of acceptability for the context in question. To state the
obvious, verbal expressions of disapproval are typically quite clear. However,
the reality is that nonverbal behavior, and so nonverbal expressions of either
approval or disapproval, is typically all too apparent. The reality of fl irting
behavior tells us just how attuned human beings are to nonverbal behavior,
for such behavior is rarely preceded by an announcement that such behavior
is about to take place. Moreover, such behavior is typically rather subtle. Yet
it is the rare person who is the object of such fl irting behavior who does not
to pick up on it. And when the fl irting behavior takes place in the presence of
other acquaintances the acquaintances usually pick up on it as well. In general,
insofar as people know that others find them sexually attractive this is owing
to their seeing the glances of others and not by public signs or statements by
others to that effect. To be sure, there can be various comments along the way.
The truth of the matter, however, is that comments do not in any way replace
the nonverbal behavior.
The point just made is perhaps rather sublime. We can all see images of our-
selves simply by looking in mirrors and then we can do an incisive compara-
tive analysis as to whether or not our physical features make us comparatively
more attractive or not (given the prevailing ideal of physical attractiveness at
the time). However, the typical person’s self-assessment of her or his physical
attractiveness, and so the person’s self-knowledge in this regard, is not secured
in this fashion. Rather, that assessment and self-knowledge is inextricably tied
to the reality that the reaction of others, generally their nonverbal behavior,
warrants precisely that assessment on the part of the individual.
The previous remarks apply mutatis mutandis to a wealth of self-assessments
that human beings make. Winston Churchill and Martin Luther King, Jr. most
152 on emotions

certainly knew that they were extraordinary public speakers. But this is not
because they did a comparative analysis of their speeches and the speeches of
others. Rather, neither could have wanted for a better indication of their public
speaking abilities than the positive reaction to their speeches, time and time
again, on the part of audiences that heard them. No doubt there were individu-
als who explicitly said to them that they are extraordinary public speakers or
specifically complimented them directly for this or that speech. None of these
comments, though, replaced the thunderous and spontaneous applauses that
they received time and time again after their speeches. Indeed, in the absence
of such reactions it would have been rather odd for either of them to think
that they were particularly gifted public speakers. So it is even if there were
well-informed individuals here and there who offered this assessment to them.
For with rare exception no informed assessment could replace the absence
of a thunderous applause, just as no statement from authorities on physical
attractiveness that a person is drop-dead-gorgeous, as they say, can replace the
admiring glances of others.
Indeed, suppose that either Churchill or King enthusiastically compli-
mented Rachel upon the speech that she gave before a 2,000-person audience.
Obviously, such words would be much appreciated by Rachel. Just so, their
complimenting her would be no substitute for a thunderous applause from the
audience.
As a final example, notice the basis for our self-knowledge regarding a most
mundane form of behavior, namely that we speak a given language fluently.
Once more, the most compelling evidence that we speak a given language flu-
ently is not some comparative analysis based upon comparing our remarks with
the recording of remarks by others who speak the language fluently. Rather,
our basis for the assessment is none other than the reaction of others when
we speak. Nothing does or can underwrite the self-knowledge that we speak a
language fluently like the fact that we can engage in conversation with others
who speak that language fluently. We rightly take it for granted that we speak
our native language fluently. But this self-knowledge is routinely reinforced in
myriad ways time and time again by just the fact that the reactions of other
native speakers of the language to our utterances make it unequivocally clear
that they readily comprehend us. The proof of the preceding observation is that
our self-knowledge in this regard would be readily shaken if throughout any
given day we encountered this person and that person and another person, all
of whom are native speakers of the very same language, who failed to compre-
hend our utterances to them. Most significantly, we never outgrow the need for
that affirmation. Not at the age of 15 or 35 or 55. Never.
The capacity for language is one of the defining aspects of being a human
being. Yet our self-knowledge that we speak a language fluently is inextrica-
bly tied to the affirmation that we receive in this regard from others. Nothing
 Self-K nowledge and the Af fir mation of Love 153

replaces that human affirmation. Needless to say, if this point holds for lan-
guage, then it holds all the more so for human activities generally involving
substantive evaluations.

The Affi rmation of Love: Beginning with Parental Love

Self-evaluation is an ineliminable part of being human. And there is essentially
no limit to what may be the subject of self-evaluation. From abilities to physical
features to the size of body parts to mannerism, human beings can enter into
extraordinary details in the matter of self-evaluation. Indeed, human beings
sometimes make comparisons between themselves and animals. Human beings
get to be “as busy as a beaver” or they “squirrel things away” or they are “as sly
as a fox” or “as strong as an ox.” And so on. Arguably, no other living creature
engages in self-evaluation to such an extent. The chimpanzee that a woman
kept as a pet and with which she took baths and occasionally shared a bed most
certainly did not make the self-evaluation that she was much better off with
him than with a male human being.
Against the backdrop that human beings constantly engage in self-evaluative
behavior across a multitude of dimensions, I should like to suggest that love
plays a most salubrious role in the life of individuals.
This is perhaps immediately obvious with parental love. For parental love
at its best is non-comparative. Parents love their child not because she or he
is better in some respect than other children but simply because the child is
their child and they are absolutely thrilled that that child is their child. Parental
love at its best displays what we may refer to as the everlasting irreplaceable
good factor: it is that child that the parents always wanted to have and to love.
Against that backdrop, shortcomings that the child may perceive herself or
himself to have by comparison with other children invariably fail to have a neg-
ative impact upon the child’s sense of self. Parental love is very much a bridge
in a child’s life over the troubled waters of harsh comparisons that children
invariably make of one another.
Needless to say, parental love at its best constitutes an affirmation of the
child that no child can give to herself or himself. Such love is also the first step
to a child’s grasping, and thus coming to know, that she or he has value simply
in virtue being a human being. Indeed, precisely what we know is that where
parental love is consistently absent in the life of a child that child will have
enormous difficulty grasping the truth that she or he has worth simply in vir-
tue of being a human being.
We know that it is one thing to know a belief intellectually and quite another
for a belief to have an affective significance in our life. This truth is just as
applicable to self-knowledge. A person can have difficulty believing what is true
154 on emotions

about herself or himself, even where the belief pertain to a very positive attrib-
ute on the part of the person. Children who have been victims of child parental
neglect often suffer from precisely this schism. What they know intellectually
does not resonate with them emotionally. Parental love at its best provides a
foundation for self-love that that has no equal among human beings. And self-
love at its best is none other than the conviction that one is worthy of being
loved by another.

The Affi rmation of Love: The Love of Friendship and

Romance
One obvious difference between parental love, on one hand, and the love of
friendship and romance, on the other, is that with parental love the child plays
no role at all in choosing the relationship, whereas with both the love of friend-
ship and the love of romance each individual plays a fundamental and, ideally,
an equal role, in creating the relationship. Furthermore, friendships and roman-
tic loves very much involve making choices based upon comparative differences
with respect to other individuals. Sometimes the set of choices is quite explicit;
sometimes the set of choices is ever so subtle but, when taken together, are clear
enough to all. No doubt the latter is particularly true for friendships, where the
social markers tend to not to be as explicit as they are for romantic ties.
In any event, once a friendship or a romantic tie is clearly in place, com-
parative differences with respect to other people become irrelevant in a fun-
damental way. A newborn is inherently love-worthy. And something has to be
terribly wrong in order for a newborn not to be responsive to parental love in
essentially all the right ways. With friendships and romantic ties, both parties
to the bond choose someone on the grounds that the individual is love-worthy.
And when both parties to the relationship are clear that such is the case of the
other, then the matter is settled. To be sure, it will always be true, at least in
principle, that a tie of love (in terms of either friendship or romance) could be
forged with another person. However, to have forged that tie with a person is
to have achieved a good that never ceases to have value for both parties. Or so
it is when things proceed as they should.
In the parent–child situation, each child can effectively say the following:
You brought me into the world in order to love me. The simplest of gestures on
the part of parents can be none other than a most profound affirmation of this
truth in a child’s heart. Parents merely beaming with pride are a marvelous
case in point. It is an affirmation that no child can give to herself or himself.
In a relationship of either friendship or love each can say to the other: You
have found me to be worthy of your love. Against this backdrop, a rich moment
of laughter can be none other than a profound moment of affirmation with
 Self-K nowledge and the Af fir mation of Love 155

regard to just how much each has meant to the other; for there would have been
little if any laughter if they were mere acquaintances. Each thereby affirms how
much each means to the other. For two people to know that they mean that
much to one another’s life is for each to provide the other with a resounding
measure of affirmation that neither could achieve acting alone.
As Robert Solomon has so eloquently argued, there is between romantic lov-
ers a passion that is entirely without equal. And if we allow for the distinction
between virtuous and vicious love (such as adult sexual interest in a child),
what makes virtuous love ever so affirming are the sublime assessments of the
other that the passion of virtuous romantic love necessarily carries in its wake.
For by way of sexual intercourse lovers make themselves emotionally vulner-
able to one another at a most sublime level, and that level of vulnerability is
affirming rather than disconcerting because it is secured by a most profound
level of trust. Each trusts the other even when each is most enveloped by pas-
sion. Indeed, the mutual trust is precisely what makes that level of passion at
best possible. Needless to say, no amount of ratiocination can do justice to the
reality of the experience of being so trusted by another. Th is form of trust has
no equal in other relationships, including friendship at its best.
It will be noted that the preceding remarks very much accord with Solomon’s
view that the passions are not some liability that human beings must endure.
Quite the contrary, I take it that the life of a human being is that much richer
when she or he can experience the envelopment of the passions in precisely the
right way, which in the case of romance I take to be trust anchored by love.
Before moving on, the following observation is very much in order. No mat-
ter how much a person’s self-examination warrants the view that she or he is
trustworthy, given the person’s ideals and behavior, such a warranted self-as-
sessment is nonetheless no equal whatsoever to the actual experience of being
trusted by morally decent individuals who are a discerning judge of good char-
acter. For to be trusted by discerning individuals is to experience the reality of
such individuals being willing to make themselves vulnerable to one in some
way or the other without any concern whatsoever that one will use their being
vulnerable against them. Trust, of course, admits of degrees, ranging from basic
trust between strangers to ever so personal levels of trust between friends,
family, and romantic partners. Yet there are countless times when even the
most basic trust can be ever so affirming, as when a stranger returns the wallet
to the pedestrian ahead who unknowingly dropped it.

The Gift of the Other: Love

Imagine an inspiring piano player who has a recording of Wolfgang Amedeus
Mozart playing Piano Concerto 5 in D minor. The piano player practiced and
156 on emotions

practiced to the point where the individual was able to play the concerto with
the skill of Mozart himself. Indeed, such a level of perfection was attained that
when the person’s own recording of the concerto was overlapped with Mozart’s
playing of the concerto it was not possible for the piano player, whose hearing
is remarkably good, to discern any difference at all. It was as if Mozart were
playing Piano concerto 5 just a little louder. What better evidence of consider-
able talent could a person want?
Well, I take it to be self-evident that none of this would hardly render
otiose a resounding acknowledgment of such an excellence from an audi-
ence of experts in the field that the piano player has played Mozart’s con-
certo fl awlessly, nay as well as Mozart himself. The explanation for this is
quite simple, namely that not even fl awless evidence that we can provide
for ourselves that we have this or that good quality or exhibit this or that
excellence can take the place of resounding affi rmation to that effect that
is entirely independent of our will. More precisely, our own entirely unbiased
assessment of a good that we have accomplished will never be the equivalent
of an independent autonomous will that makes the very same assessment of
what we have done. For in the latter case, another bears witness to the good
that we have done; hence, the good that we have accomplished is reflected in
the life of another.
Most significantly, although in view of all the recording comparisons that
have been made the piano player has every good reason to believe she or he can
play Mozart’s Concerto 5 at a level of excellence on a par with Mozart himself,
it would in fact be most disconcerting to the piano player, perhaps even psycho-
logically unbearable, if it turned out that all decent and well-meaning experts
failed to make the judgment of the pianist’s playing of Mozart Concerto 5 is
absolutely extraordinarily. More precisely, nothing would underwrite the pia-
nist’s conviction that her or his playing of the concerto is on a par with Mozart’s
playing of the concerto himself like the independent judgment of experts to
that effect. In other words, the pianist’s self-knowledge that her or his playing
of Mozart’s concerto is on a par with Mozart himself is mightily grounded in
the judgments of others, namely, experts, to that effect.
The point just made calls to mind Hume’s observation that “the minds of
human beings are mirrors to one another’s soul.” The kind of life that we live
is reflected in the judgments of those who witness our lives. Accordingly, the
independent judgment of others of an excellence in our lives is an affirmation
that we cannot give ourselves. That is why our self-knowledge is considerably
underwritten by social interaction. These remarks obviously apply equally to
our moral character. In the words of Bernard Williams, it would take a long and
unobvious story for a person to suppose that he lives an upright life if person
after person thought just the opposite (where all these individuals are them-
selves honest people of enormous goodwill).
 Self-K nowledge and the Af fir mation of Love 157

This observation brings us back to love. As already noted, in the absence of

parental love no child would come to have a deep and abiding sense of value
for herself or himself. For a child to grow up with loving parents is for the
child to experience in a multitude of ways and often at a most visceral level,
by way of the emotional displays, reactions, and words of the parents, that she
or she means ever so much to them. Friendships and loves are two kinds of
relationships beyond parental love that make it unequivocally clear to a person
that she or he means so very much to another. Indeed, the love of friendship
and romance do so in a way that parental love cannot. This is because parents
should love their parents. Indeed, it is irresponsible for adults to bring a child
into the world and not love that child. But as I argued in “Reasons for Loving,”
no one is any way morally obligated to become anyone’s friend or romantic
partner. There is no amount of righteous on an individual’s part that requires
another so respond to her or him. Accordingly, when the love of friendship and
romance are at their best, each in its own way, the person who is the object
of either or both forms of love has a most extraordinary affirmation of being
valued by another that is entirely independent of her or his will. As adults, the
knowledge that we are so wondrously valued by another is an affirmation of the
self that is entirely without equal.
Annette Baier is ever so right that there is a need for more than justice.
Ironically, one way of seeing this is that justice is what we owe to others,
whereas the love of friendship and romance is not. And the very truth that
this essay is concerned to bring out is that the most profound affi rmation
that human beings can give to one another, namely love, is tied to the reality
that their love for another is on all accounts freely given. Even parental love
at its best is thought to be freely given, parental obligations to the contrary
notwithstanding. To be sure, there is no gainsaying the truth that an officer
who puts her life on the line in order to save a stranger from being the victim
of an egregious wrongdoing does a considerable good on behalf of that per-
son. What is true, though, is precisely the fact that the officer would do, and
would have done, the very same thing for any other person similarly situated.
And saving that stranger’s would be no more meaningful than saving the life
of a different stranger. In either case, the officer is fulfi lling the call of duty.
But to love a person is not just to sanctify the life of another. It is also to
sanctify the bonds that have forged between oneself and that individual. And
it is possible to be ever so just in one’s treatment of another without there
ever being such bonds with that individual. Indeed, one can hold that a life is
worth saving, and do all that one can to achieve that end, even as one rightly
maintains that there should not be any deep bonds forged between oneself
and that individual.
In understanding ourselves and having a measure of appreciation for who
we are the love of another is without equal. Parental love at its best is that
158 on emotions

most salubrious foundation. The love of friendship and romance is that ever so
extraordinary scaffold that facilitates our negotiating the vicissitudes of life.
In his majestic work, A Theory of Justice, John Rawls maintained that no one
can do everything and that we must often look to others for the excellences
that we cannot realize in our own lives. Alas, there is no excellence like the
excellence of being valued by another—of knowing that we mean ever so much
to another—where none other than love itself is the basis for that valuing. That
excellence, which is so often a reason for pressing forward rather than giving
up, is an affirmation that we cannot give to ourselves. Rather we must look to
others for it. We are among the most fortunate of individuals when another
has bestowed upon us that excellence, which is none other than the gift of love.
And there is perhaps no greater indication of love’s power than that we do not
need all the world to love us. Among the millions—nay, billions—of people
that there are on the face of the earth, it suffices that there are few good people
who love us. Such is the power and affirmation of love.

References
Baier, Annette. 1995. “The Need for More than Justice” in Justice and Care: Essential Readings
in Feminist Ethics, ed. Virginia Held (Oxford: Westview Press).
Deigh, John. 2008. Emotions, Values, and the Law (New York: Oxford University Press).
Helm, Robert Bennett. 2010. Love, Friendship and the Self (New York: Cambridge University
Press).
Harmon, Katherine. 2009. “Why Would a Chimpanzee Attack a Human?” Scientific American,
February 19.
Silverman, Eric J. 2011. “Robert Solomon’s Rejection of Aristotelian Virtue: Is the Passion of
Erotic Love a Virtue that is Independent of Rationality?” Essays in Philosophy 12: 18–31.
Solomon, Robert C. 1981. Love, Emotion, Myth and Metaphor (New York: Doubleday).
Solomon, Robert C. 1991. The Passions (Indianapolis, IN: Hackett Publishing Company).
Solomon, Robert C. and Higgens, Kathleen M. 1991. The Philosophy of Erotic Love (Kansas,
MO: University of Kansas Press, 1991).
Thomas, Laurence. 1991. “Reasons for Loving.” In The Philosophy of Erotic Love, Robert C. Solomon
and Kathleen M. Higgers (eds.), 467–476. University of Kansas Press.
 8

Love and Death

Kathleen Marie Higgins

Introduction
This is in some ways a response to the late Robert C. Solomon’s “Good Grief,”
delivered during one of his last visits to New Zealand. In particular I want to
focus on two of his claims in that talk, first, that grief shows that emotions can
be in some sense obligatory, and, second, that grief is a continuation of love. I
will argue that while he exaggerated the extent to which grief is held to be a
moral obligation, another obligation that he gestured toward is certainly widely
felt within contemporary Western culture—the obligation to get over grief.
This contemporary tendency to consider closure to grief obligatory in fact
runs counter to the insight that grief continues love. It ignores the sense in
which love continues after death, and the extent to which our identities are
formed in relation to those whom we love. In opposition to theories that have
held that relinquishment of attachment is healthy, I will argue that grief has a
narrative structure that importantly involves the continued relationship of the
mourner with the deceased.

The Story
Some readers will no doubt suspect, if they are aware that Bob and I were mar-
ried, that I am drawing on personal experience in taking grief as my topic.
Unavoidably, I am. However, the concerns I want to raise are far from a com-
plaint on my own behalf. I have been fortunate, and I think unusually so, as is
demonstrated by such commemorative contexts as this. My experiences, how-
ever, have given me insight into the concerns of the grieving and those who aim
to support them, and it is these, primarily, on which I will draw.
Given the inevitably personal character of my topic, and the fact that I
aim here to memorialize Bob, I’ll begin with a personal story. A few days

159
160 on emotions

before his demise, Bob met with a publisher to discuss the prospect of his
writing a book about death. (The coincidence is rather eerie, to be sure. One
of Bob’s brothers commented, “A person can take field work too far.”) I asked
Bob what his angle would be in the book. “It’s all about the story,” he said,
rather cryptically.
“It’s all about the story” will be my theme, too, not in connection with death
directly, but in connection with grief. In fact, I think “the story” is a useful
lens through which to approach every major life transition, in part because it
draws attention to their social dimension. In this respect, I find a rather sur-
prising resemblance between the circumstances of getting married and those
of being widowed. One obvious similarity is that both occurrences are marked
by a social event. Many of the same people are gathered on both occasions.
Sometimes, too, a wedding and a funeral take place in the same venue, or recep-
tions that follow them do.
I noticed another similarity, too. In the case of getting married, I found,
somewhat to my surprise, that this change in relation to a particular individual
had social repercussions across the board. Getting married subtly altered all
of my other relationships. As a consequence of getting married, one’s personal
identity is henceforth understood to be connected with another’s. The assump-
tion is that one now lives out one’s story in tandem someone else. The stories
become joined in a way that is an unquestioned social fact.
I might have thought that marriage would inhibit a sense of close relation-
ship to other people, at least some other people (such as former beaus). But in
fact, it made each relationship stand out in relief as the specific relationship
it was. In that sense, it enhanced a feeling of intimacy with people besides my
new spouse, perhaps because each relationship was now more clearly defi ned
or specified (if only by virtue of foreclosing certain options). To my surprise, I
felt closer to everyone who was important to me. I suspect this was also a con-
sequence of the public ritual, the “sharing of my special day.” Whether or not a
particular person had been present for the event of the wedding, all those close
to me seemed to be part of it. Getting married intensified my sense of their
presence in my life and heightened my appreciation of them.
Becoming widowed is the other endpoint of spouses living their stories in
tandem. This transition, too, results in a subtle shift in one’s relationships
across the board. My sense of intimacy with everyone I am close to was once
again in emphatic evidence. Most communicated with me almost immediately,
and many offered or simply provided unsolicited practical help. The support of
all the other individuals who are important in one’s life is, I would argue, cen-
tral to what makes the experiences of getting married and becoming bereaved
what they are.
Another parallel between marriage and the death of a spouse is that one’s
social identity is suddenly transformed, and one’s social circle is typically well
 Love and Death 161

aware of this fact.1 In both cases, other people think in terms of one’s life as
a story. Their responses include a set of narrative concerns. When one loses
a spouse or other person with whom one has shared one’s life, people wonder
about your plans. “Are you going to move?” is a not infrequent question. Such
questions are not well suited to the probable state of mind of a bereaved person,
who is likely to be too stunned over the death to be braced for dealing with
yet another massive change. But the typicality of such inquiries is a sign of
our tendency to think about other people’s lives in narrative terms, as we do
our own. When someone has reached a juncture in which circumstances are
significantly changed, we wonder, “What exactly is the person going to do?”
As will become clear later on, I think our narrative expectations can help us
to navigate through grief, both as grievers and as those who want to support
others who grieve.
The social world is deeply involved in both weddings and funerals. It attends
and heightens the emotional experiences of key participants. The community
of friends and relatives plays a vital role in both types of events, offering emo-
tional support and validating the change of social status. The larger group
shares the joy of a marrying couple in one case and feels condolence with the
bereaved in the other.
And yet, the role of the social world is not entirely benign. Anyone who
has had a wedding knows the pressures caused by the strong views of other
people about how details of the event should be handled. Perhaps in the case
of a wedding, the tensions serve a useful function by testing the betrothed
couple’s ability to maintain commitment and affection in a stressful situation.
Bereavement has the potential for similar stress, but without any resulting
benefit. Although contemporary people may have fewer preconceptions about
ritual details for a funeral than they do for a wedding, most have surprisingly
strong expectations about grief.

Grief and Obligation

Bob summarized his view of grief as obligatory in the last of his books to be
published while he was still alive, True to Our Feelings: “We are not just sur-
prised when a person shows no signs of grief after a very personal loss. We are
morally outraged and think much less of the person.”2 When Bob argued that
grief is obligatory, he had in mind cases such as the one Albert Camus depicts in

1
Colin Murray Parkes claims that stigma is a factor that always figures in bereavement.
See Colin Murray Parkes, Bereavement: Studies of Grief in Adult Life, 3rd ed. (London: Routledge,
1996), 5.
2
Robert C. Solomon, True to Our Feelings (New York: Oxford University Press, 2007), 75.
162 on emotions

The Stranger. Merseult, that work’s narrator, is put on trial for murder. The pros-
ecutor argues that he did not sufficiently grieve the death of his mother; he did
not weep at her funeral, took coffee with the funeral director, and even went
on a pleasant outing with a young woman soon after returning home. Noting
that the next case on the docket was the case of a person accused of murdering
a parent, the prosecutor suggests that Merseult was effectively guilty of that
crime as well. This argument seems to figure in the jury’s decision to convict
him of murder.
A factual case in support of Bob’s position is that of Lindy Chamberlain,
an Australian mother who was accused of the murder of her infant daughter,
Azaria, after the latter disappeared from their campground near Uluru. The
jury did not believe Ms. Chamberlain when she said that a dingo had carried
off her baby, evidently taking her lack of apparent grief on the witness stand to
be evidence of her guilt. She was convicted, but the verdict was overturned on
appeal in light of new evidence that supported her claim that the dingo had
killed the baby. This case illustrates moral condemnation of a person for not
(apparently) grieving someone whom society deems she should grieve.
This attitude toward the grief of other people, according to Bob, indicates
that some emotions are held to be morally obligatory. It is worth pointing
out that Bob had theoretical reasons for emphasizing the idea of an obliga-
tory emotion. One of these is his view that we are in large measure able to
exercise choice and control with respect to our emotions.3 If society reproaches
people for insufficiently grieving the loss of loved ones, folk psychology seems
to acknowledge that we think people have some control over their emotional
responses.
Bob’s skepticism regarding emotional valence is also relevant. Valence
is the idea that emotions can be classified as positive and negative. Bob was
convinced that the positive/negative binary is misleading because it overly
atomizes emotions, treating them as short-term events that have a straight-
forwardly positive or negative value in psychic life. The notion of valence does
not reflect the fact that an emotion that seems undesirable at one point in time,
from a certain perspective, may later, or from another perspective, seem to be
highly adaptive and desirable. If grief is obligatory, it represents a case in which
a seemingly paradigmatic case of a “negative” emotion is valued positively, as
something that ought to take place.4 Hence, once again, it supports Bob’s views
about emotion more broadly.

3
See Robert C. Solomon, “Emotions and Choice,” Review of Metaphysics, 27 (1973): 20–41.
4
Th is is assuming the psychological terminology, in which “positive” emotions involve
appraisal as good while “negative” emotions involve the opposite appraisal. If, by contrast, we
took “negative” to be equivalent to “unpleasant” here, we might still call grief negative; however,
Bob would point out that this underemphasizes the importance of grief in psychological health. It
would be more unpleasant, at least over time, to try to do without it, he might observe.
 Love and Death 163

Bob is, I think, right that most of us suspect a defect of character if some-
one close to a person who dies does not grieve, or does not grieve very much.
But some would not agree. Beverley Raphael comments that a bereaved per-
son behaving as if nothing had happened may be interpreted as “evidence of
strength and coping by many,” a point of view diametrically opposed to what
Bob takes to be the moral consensus.5 George A. Bonnano and colleagues simi-
larly consider absent or only mild symptoms of depression from the time of loss
through eighteen months later as manifesting “a genuine form of resilience.”6
They acknowledge that some people may react negatively to people who dis-
play this pattern, taking it as an indication of indifference, but they claim,
“Alternatively, others may react more favorably to a resilient person because it
is easier for them to be with someone who is less distressed.”7
I submit that more often than seeing it as moral failure or a sign of coping,
we tend to see a bereaved person’s lack of grief as a sign of mental patholo-
gy.8 Raphael contends that usually the lack of any evident distress on the part
of a bereaved individual is a sign of mental disturbance. Absence of grief in a
bereaved person is one species of what clinicians euphemistically term “com-
plicated grief,” in other words, pathological grief, grief that has somehow gone
wrong. Except when motives are present that obstruct the operation of the
principle of charity toward a bereaved individual, we tend to medicalize rather
than moralize absent grief.9
A qualification that should be offered to Bob’s claims about the obligation
to grieve is that while society requires grief from people standing in certain
relationships to the deceased, it proscribes grief in other cases, sometimes
even when relationships are close. Kenneth J. Doka has coined the term

5
Beverly Raphael, The Anatomy of Bereavement (New York: Basic Books, 1983), 205.
6
George Bonnano, Kathrin Boerner, and Camille B. Wortman, “Trajectories of Grieving,” in
Handbook of Bereavement Research and Practice: Advances in Theory and Intervention, ed. Margaret
S. Stroebbe, Robert O. Hansson, Henk Schut, and Wolfgang Stroebe (Washington, DC: American
Psychological Association, 2008), 292.
7
Ibid., 303.
8
Another possibility is that we see the bereaved person’s actual relationship to the deceased
not to have been particularly close, even if the parties are formally related. Probably we would
not draw this conclusion, however, unless we know something about the estranged status of the
apparent relationship.
9
Th is is not to deny that some individuals who are close to the bereaved may well be critical
of the way the bereaved person is handling the situation, particularly when grief is absent. I will,
however, be suggesting that the criticism is less often directed at absent or insufficient grief than
at other aspects of the person’s behavior. Stroebe and colleagues summarize: “Chronic (or pro-
longed) grief has been generally well accepted as a pathological category, whereas delayed, inhib-
ited, and absent grief remain the subject of much debate.” Margaret S. Stroebe, Robert O. Hansson,
Henk Schut, and Wolfgant Stroebe, “Bereavement Research: Contemporary Perspectives,” in
Stroebe et al., Handbook of Bereavement Research and Practice, 8.
164 on emotions

“disenfranchised grief” to refer to situations in which grief is not given social

recognition. He defines disenfranchised grief as “the grief that persons expe-
rience when they incur a loss that is not or cannot be openly acknowledged,
publicly mourned, or socially supported.”10
Doka refers to unrecognized relationships, unrecognized losses, and unrec-
ognized mourners. Grieving can be held to be in violation of social norms when
the person grieving stands in an unsanctioned relationship to the deceased
(for example, an adulterous lover) or one that society does not recognize as
significant (for example, ex-spouses and foster parents). Grieving is also dis-
enfranchised when the deceased has died in a manner that is somehow embar-
rassing (such as suicide or execution) or is not valued by the community. I
discovered this type of situation when a friend of mine, a former teacher, who
was in an extremely stressful situation, snapped. He shot and killed a woman at
point-blank range and was shot and killed himself in the chaos that ensued. The
horrific incidents were reported in the newspaper, and I was initially shocked
to see that my friend was labeled “the gunman.” The story and its aftermath
continued to be reported for several days, and I poured over the pages of the
paper, expecting to see some announcement of where funeral services would be
held for my friend. Eventually it hit me: newspapers do not announce memorial
services for “gunmen.”11
The implication of disenfranchised grief is that not everyone is entitled to
grieve. While grief may be socially mandated for some, others may not have
the right.
The very notion of disenfranchised grief presupposes the psychological impor-
tance of social recognition and public mourning. It is harder to process private
grief than grief that is socially acknowledged. The private mourner experiences
both the loss of the deceased person and the loss of rapport with the rest of the
social world, which neither takes the death as a catastrophe nor recognizes the
grieving person’s loss. Emotional responses are experienced underground. Doka
points out that disenfranchisement tends, consequently, to intensify emotional
responses, particularly those of guilt, anger, and powerlessness.12
The phenomenon of disenfranchised grief also draws attention to the power
of societal expectations regarding what is appropriate and what inappropriate.

10
Kenneth Doka, ed., Disenfranchised Grief: Recognizing Hidden Sorrow (Lexington, MA:
Lexington Books, 1989), 4.
11
Grief can also be disenfranchised by being politically unauthorized or censured, the clas-
sical example being that of Antigone. Judith Butler notes that to grieve those who die as a conse-
quence of one’s own nation’s violence is considered unpatriotic and is morally censured. See Judith
Butler, “Violence, Mourning, Politics,” in Precarious Life: The Powers of Mourning and Violence (New
York: Verso, 2004), 34.
12
Kenneth Doka, “Disenfranchised Grief in Historical and Cultural Perspective,” in Stroebe et
al., Handbook of Bereavement Research and Practice, 234.
 Love and Death 165

Doka notes that societies have “norms that frame grieving,” including “not only
expected behaviors but also feeling, thinking, and spiritual rules.”13 In the case
of disenfranchised grief, “the personal experience of grief is discordant with
the society’s grieving rules.”14
In what follows I will be discussing grieving norms operative in contem-
porary Western society (apart from some subgroups within it). I will not be
making claims about what a healthy course of grieving is in other cultures.
Cultures vary in their grief rules and in the accounts that they give to explain
the sense of these rules. One consequence is that the significance of a specific
grief behavior can be culturally variable even when outwardly similar.15 I will
accordingly restrict my claims about the harm of social demands that grief be
limited to contemporary Western society.

The Obligation to Stop Grieving

Among a society’s grieving rules are rules for how the course of grieving should
go. Bob emphasized the moral obligation for some people to grieve; I want to
consider another kind of social policing, which he noted without much elabora-
tion, the pressure to “get over it.” 16 Treating grief as self-indulgence or an intel-
lectual problem is nothing new. In Shakespeare’s Romeo and Juliet, Lady Capulet
says to Juliet, whom she takes to be grieving for her slain cousin, Tibald:

“Evermore weeping for your cousin’s death?

What, wilt thou wash him from his grave with tears?
An if thou couldst, thou couldst not make him live;
Therefore, have done. Some grief shows much of love,
But much of grief shows still some want of wit.”17

In contemporary times British psychiatrist Colin Murray Parkes points out

that his own society’s attitude toward the newly bereaved is to “find it dif-
ficult to accept their need to mourn,” and he adds “when forced to meet
them we find ourselves at a loss.”18 He cites Geoff rey Gorer, who observes

13
Ibid., 225.
14
Ibid., 226. He adds that not only are these “ folkways, or informally expected behaviors, but
also . . . laws, meaning that these norms may carry sanctions.” See ibid., 224.
15
See Paul C. Rosenblatt, “Grief across Cultures: A Review and Research Agenda,” in Stroebe et
al., Handbook of Bereavement Research and Practice, 211.
16
Bob tended to equate the pressure to get over grief with a general denial of it. I think these
are closely related, but I also think that there are other motives behind this pressure, which I will
be discussing presently.
17
William Shakespeare, Romeo and Juliet, Act 3, Scene 5, lines 69–72.
18
Parkes, Bereavement, 9.
166 on emotions

regarding behavior within contemporary British society: “Mourning is treated

as if it were a weakness, a self-indulgence, a reprehensible bad habit instead of
a psychological necessity.”19
Behind the conviction that those who grieve should get over it in fairly short
order are various motivations. Bob was convinced that this phenomenon was
an amplification of a denial of grief, which he associated in particular with
American culture (aside from certain ethnic groups within it). Another motiva-
tion that he acknowledges is the general distaste for emotion (especially among
men) in Western society. Certainly, these are among the motives, but there are
others as well.
One is fairly self-interested. People are often embarrassed when encounter-
ing a bereaved person. They don’t know what to say. Even more awkwardly, they
might embarrass themselves by forgetting that the bereaved person is grieving
and saying something inappropriate. Such forgetfulness is not unlikely in a con-
text in which external signs of bereavement—such as mourning dress—have
been abandoned.20 A third-party’s wishing for an end to grieving—the circum-
stance that occasions this kind of embarrassment—makes emotional sense.
Moreover, unless one has fairly detailed knowledge of the bereaved indi-
vidual’s beliefs, in a society with nonhomogeneous religious convictions one is
uncertain as to how traditional formulas for comforting the bereaved will sound
to the person. References to God or heaven may be consoling, but they may also
sound alienating. The clash between the beliefs of the would-be comforter and
the beliefs of the bereaved may become evident in the course of any attempt at
offering solace, and this adds to the stress of the situation for both parties.21
Uneasiness is exacerbated by the fact that someone’s bereavement makes
demands on others who were less close to the deceased. In the absence of clear
social etiquette for ongoing behavior toward the bereaved, the demands on third
parties are felt but vague, adding to a sense of anxiety surrounding them.
A related motivation is the desire to avoid feeling powerless. A person who
is genuinely concerned about the bereaved person wants to be able to inter-
vene and fi x the situation. But death precludes this possibility. It is easy for a
concerned third party to feel as though his or her efforts are inadequate to the
problem. This sense of incapacity is certainly unpleasant. Someone who feels it
has a motive for wanting the grief to end.

19
Geoff rey Gorer, Death, Grief and Mourning in Contemporary Britain (London: Cresset, 1965).
20
For a consideration of a society that gave considerable attention to such external markers
of grief, see John Morley, Death, Heaven and the Victorians (Pittsburgh: University of Pittsburgh
Press, 1971).
21
The difficulty here is exacerbated by the fact that individuals tend to be particularly ada-
mant about their beliefs regarding what happens at death. Perhaps they have a particularly strong
needs to construct such an account in order to feel secure in their narrative sense of the life they
are unfolding, since death marks the end (or at least apparent end) of the story.
 Love and Death 167

Another rather self-interested motive is the desire to have a normal rela-

tionship with one’s grieving friend once again. Those who are close to the per-
son, even if they are not suffering from the same loss, will feel the disruption of
everyday interaction with the bereaved person. Given that grieving people are
sometimes emotionally volatile, taking out their distress on those who are near
them, this is not difficult to understand. And in any case a grieving person, par-
ticularly soon after the loss, is often “not all there” in any usual sense.
A further self-interested motivation for wanting a bereaved person to
get past grief may seem superstitious from the standpoint of contemporary
Western society, but I think it remains present. This is the motive of wanting
the world of the living to be separated from the dead by a clear boundary. Mary
Douglas persuasively argues that human beings are deeply concerned with the
danger of contamination by whatever is categorized as impure. Corpses—sites
of decomposition—are paradigm cases of something that would be so classi-
fied in straightforwardly physical terms. Douglas argues that the same catego-
ries of purity and pollution are reflected in our demand for definite conceptual
boundaries between types of things. Things that resist clear classification evoke
repulsion. Horror fi lms trade on the tendency of human beings to find what
resides at the interstices between categories repugnant, with the undead being
horrific because they straddle the boundary between the living and the dead.22
For a person to persist in grief behavior that seems not to acknowledge that a
deceased person is unconditionally on the other side is, I suspect, unnerving. It
defies one of the most basic and emotionally charged distinctions that we use
to make sense of our experience, that between life and death.
Straightforwardly empathetic reactions can also motivate the desire that
someone get past grief. Most obviously, a third party may simply wish that the
mourner were happy, perhaps in the midst of some new chapter of life. Thus,
the third party might encourage a bereaved spouse to start dating again. In this
case grief is not directly the focus of the person’s wishes, but the happiness that
is desired for the bereaved person is possible only if grieving comes to an end.
Parkes points out a second empathetic motivation. He observes that a third
party’s urging a person to stop grieving can be a response to the mourner’s
sense of obligation to the deceased:

In a situation in which well-established norms are absent, the expec-

tations of those around are potent determinants of behaviour. Thus
friends or relatives can indicate, implicitly or explicitly, that grief is
expected and permitted, but they can also indicate that it has gone

22
For a discussion of the relation of horror fi lms to the categories of the pure and the polluted,
see Noël Carroll, “The Nature of Horror,” Journal of Aesthetics and Art Criticism 46, no.1 (1986):
51–59.
168 on emotions

on long enough. To some extent grieving is seen as a duty to the dead

and it may take an outsider to point out that the duty is now done, or
at least that the mourner can be permitted to let up a little. Not that
this is usually a major problem nowadays. It is more likely that the
bereaved person will get little encouragement to mourn at any time—
but when grief has broken through it may take a special circumstance
to get the mourner out of a state of habitual mourning.23

Parkes is convinced that the absence of clear social norms is hard on the
bereaved, leaving them “confused and insecure in their grief.” 24 Certainly, those
who grieve in contemporary Western society no longer experience what Robert
Kastenbaum describes as typical in folk cultures, in which “the pain of personal
loss would meld with role expectations.”25 In such a context, the bereaved per-
son may depend on others for any sense of proportion.
Another empathetic motive is concern that the mourning individual is in
danger of solipsism—or at least of becoming profoundly out of sync with the
rest of the social world—if grief goes on indefinitely, beyond the period that
others recognize as “recently” after the loss. A third party might also think
bringing the grief to the mourner’s attention may hinder the process of healing
that ideally is taking place. The idea seems to be that the work of grief is best
done unconsciously, and any conscious upset threatens to interfere with the
process. This seems to be the rationale behind the common practice of trying
to take a newly bereaved person’s mind “off it.” The reasoning here reminds
me of the point made by Johnny Cash in the film Walk in the Line when one of
the prison authorities complains to him that they wished he wouldn’t remind
inmates that they are in prison: “You think they forgot?”26
So many motives prompt the social tendency to urge the bereaved to get
over grief that it is surely an overdetermined phenomenon. Theories of grief
also influence how grief is experienced and understood, and considerable
theory supports the view that grief should not be too prolonged.27 In other
cultures, metaphysical beliefs sometimes serve as theoretical grounding for
the prescription to emotionally detach, such as the view that the soul of the

23
Parkes, Bereavement, 177.
24
Ibid., 172.
25
Robert Kastenbaum, “Grieving in Contemporary Society,” in Stroebe et al., Handbook of
Bereavement Research and Practice, 73.
26
Walk the Line, directed by James Mangold (2005), Fox 2000 Pictures. Cf. Parkes: “People
entering a household that contains a newly bereaved person are often enjoined not to say any-
thing that will ‘upset’ the bereaved. Since conversation about trivialities is irrelevant at such a
time, this makes communication difficult. Usually such an attitude reflects a mistaken notion
that grief can somehow be avoided” (pp. 174–175).
27
See Tony Walter, “A New Model of Grief: Bereavement and Biography,” Mortality 1 (1996): 20.
 Love and Death 169

deceased is prevented from going on its way by the living person’s holding on.
Social scientific theory from Freud until recent times has also encouraged the
view that one should relinquish one’s ties to the deceased. The remainder of my
discussion will focus on theoretical accounts of grief, beginning with Freud’s
analysis.

Relinquishment Theory
What is the point of bringing grief to an end, according to the relinquish-
ment theorist? Typically the idea is that one should get in touch with reality.
Freud’s theory is that grieving involves the destruction of bonds with the lost
loved one. In Freudian psychology, psychic energy (libido) is literally attached
to an object. In grieving, one’s available psychic energy is largely directed
into reality testing. One sequentially brings up various images of the beloved
and ascertains in each case that it does not correspond to a “real” object any
longer. Th is is a prolonged project, given that one has innumerable images
of the beloved. But as each image is tested and consigned to the category of
“not corresponding to something real,” one frees up some libidinal energy.
In each case, the energy freed is a small quantity, but over time, one gains
considerably more libidinal energy, potentially available for reinvestment in
a new object.
Marcel Proust provides a more poetic account of this process when his narra-
tor describes his state of mind on hearing of the death of his erstwhile lover:

How could she have seemed dead to me when now, in order to think
of her, I had at my disposal only those same images one or the other
of which I used to recall when she was alive, each one being associ-
ated with a particular moment?. . . . A little statuette as we drove to
the island, a large, calm coarsely grained face above the pianola, she
was thus by turned rain-soaked and swift, provoking and diaphanous,
motionless and smiling, an angel of music. So that what would have
to be obliterated in me was not one only, but countless Albertines. . . .
So great a wealth of memories, borrowed from the treasury of her life,
such a profusion of sentiments evoking, implicating her life, seemed
to make it incredible that Albertine should be dead. . . . 28

Anyone who has grieved recognizes the flooding of the mind with images and
memories of the lost loved one. And each does seem to prompt a new wave

28
Marcel Proust, Remembrance of Things Past, in 2 vols. The Sweet Cheat Gone, trans. C. K. Scott
Moncrieff (New York: Random House, 1932), Vol. 2, 725–726.
170 on emotions

of loss, as if one is thinking, “And I’ve lost the person associated with this
memory, too!” One tends to be thrown by the first occurrence of a particular
memory of a newly deceased loved one in a manner that is not the case for sub-
sequent appearances of that particular memory. In the latter case one seems to
be braced for the memory’s coming to mind. One will not be overwhelmed by
that particular recollection again, perhaps, but another will take one by sur-
prise. Freud’s claim that we surmount the assault that each memory produces,
but that surmounting one leaves one vulnerable to the onslaughts of others,
seems to correspond to the phenomenology. To that extent the Freudian theory
is plausible.
Difficulties attend Freud’s account, however. In the fi rst place, his account
of libidinal cathexis involves what Bob called the hydraulic model of the
psyche, a model that is dubiously mechanistic. The hydraulic model treats
emotions as something like a fluid fi lling a container. Th is is a metaphor,
obviously, but one that Freud virtually literalizes when he treats psychic
energy as the fluid content of the psychic apparatus and as susceptible to
direction and redirection, as though one were diverting it through alterna-
tive pipes.
Theoretical scruples about this model aside, a problem with Freud’s account
is that it does not explain why the recurrence of memories of the deceased,
though jarring soon after a loss, quite often come to be treasured. What is the
basis for this treasuring if one’s earlier review of the memory is a matter of
exploding the libidinal connection? Treasured memories of dead loved ones
suggest that a sense of connection hasn’t been eliminated after all. Attachment
seems to continue.
Although Freud theorizes that the relinquishment of bonds is the optimal
outcome of grief, Freud does not consider the stages of grief prior to relinquish-
ment to be pathological. In “Mourning and Melancholia,” he remarks:

It is . . . well worth notice that, although grief involves grave departures

from the normal attitude to life, it never occurs to us to regard it as a
morbid condition and hand the mourner over to medical treatment.
We rest assured that after a lapse of time it will be overcome, and we
look upon any interference with it as inadvisable or even harmful.29

Freud’s view in this connection, however, is no longer the consensus.

Some psychologists argue for including a category in future editions of the
Diagnostic and Statistical Manual of Mental Disorders (DSM ) for “prolonged grief

29
Sigmund Freud, “Mourning and Melancholia” (1917), in General Psychological Theory: Papers
on Metapsychology (New York: Macmillan, 1963), 164–165.
 Love and Death 171

disorder.”30 Although proponents of this inclusion are quick to point out that
length of grief is only one feature of the condition they want to categorize, the
fact that the proposed name for the disorder makes reference to the persis-
tence of grief indicates their expectation that grief should pass rather quickly.
(In fact, the proponents would take a year after loss to be the point at which
they could make a determinate diagnosis.) This proposal would reinforce the
current, problematic social tendency to grant only temporary space for grief.
Relinquishment theory is inadequate. Although it is overdetermined that
third parties will want grief to come to an end, the love that grief presupposes
encourages exactly the opposite situation. Assuming that rapid recovery is par-
ticularly unlikely where strong bonds with the deceased exist, the very people
whom one would expect to grieve the most are the most vulnerable to censure
for extended grieving. The social pressure to stop grieving amounts to a disen-
franchisement of all grief that extends beyond a limited span of time.

Continuing Bonds and Personal Identity

Fortunately, a growing perspective within the social sciences is that mental
health depends on attachment, including attachments to deceased loved ones.
The “continuing bonds perspective” holds that maintaining emotional ties to the
departed is a healthy adaptation.31 Klass, Silverman, and Nickman, who coined
this expression, challenge the relinquishment view, arguing that continuing
attachment to the deceased person can enrich life after loss.32 Psychiatrist John

30
The fi fth edition of the DSM is, as of this writing, is slated to include grief after the death
of loved one under the more general category of “Major Depressive Disorder” and would allow
the diagnosis just beyond two weeks after the loss. (The previous edition had had a “bereave-
ment exemption” for the diagnosis in recognition that the symptoms of grief are often similar to
those of major depression.) Th is planned revision of the DSM has generated considerable negative
public response, including a petition posted by the Society for Humanistic Psychology which was
signed by over 7000. Lancet and the New York Times have editorialized against this categorization.
See Lancet, 379, no. 9816 (February 18, 2012): 589. (doi:10.1016/S0140–6736(12)60248–7); and
Benedict Carey, “When Does a Broken Heart Become a Diagnosis?” New York Times, January 24,
2012. See also Allen Frances, M.D., “DSM 5 to the Barricades of Grief,” Psychology Today, February
18, 2012. Available at: http://www.psychologytoday.com/blog/dsm5-in-distress/201202/dsm-5-
the-barricades-grief. For a recent scholarly discussion of the issue, see Jerome C. Wakefield,
“Should Prolonged Grief Be Reclassified as a Mental Disorder in DSM-5?: Reconsidering the
Empirical and Conceptual Arguments for Complicated Grief Disorder,” Journal of Nervous &
Mental Disease 200, no. 6 (June 2012): 499–511.
31
See Nigel P. Field, “Whether to Relinquish or Maintain a Bond with the Deceased,” in Stroebe
et al., Handbook of Bereavement Research and Practice, 113.
32
Dennis Klass, Phyllis R. Silverman, and Steven L. Nickman, Continuing Bonds: New
Understandings Of Grief, Series in Death Education, Aging, and Health Care (Washington, DC:
Taylor & Francis, 1996).
172 on emotions

Bowlby, the most prominent figure among attachment theorists, contends

that even though forming new attachments with the living is a desirable out-
come, grief is aimed at reorganizing one’s attachments, and that continuing
symbolic attachments to dead loved ones is part of integrating the loss into
one’s life in the new situation.33
I am convinced that this conception of grief is vastly superior to one that
urges the dissolution of bonds to the deceased. If Bob is right that grief is a con-
tinuation of love, relinquishment theory misconstrues what grief is all about.
I will defend two other reasons for rejecting relinquishment theory in favor of
recognizing the importance of continuing bonds. First, the notion of “getting
over” grief does not cohere with the reality of a bereaved person’s own con-
ception of identity in the context of the new situation. To actually eliminate
the relationship with a beloved dead person in one’s psychic life would entail
eliminating much of one’s sense of self as well. Second, those who grieve make
use of stories, which seem to assist the efforts to reorganize their lives, and
this suggests that narrative structure, with its temporally unfurled character,
is valuable to processing grief in a way that the “on–off ” notion of “getting over
it” is not. I will discuss these points in turn.
With respect to the bereaved person’s sense of identity, how would it fare if
the person really could get over the loss of a deceased loved one? Loving some-
one has an impact on one’s identity so that one’s sense of oneself is linked to
each of the other people one loves. Bob’s account of romantic love was that
it involves reconstituting one’s identity with and through another person.
Romantic love involves a dyadic, reciprocal relationship, and the reconstruc-
tion of identity that is involved is given social acknowledgment, at least when
it culminates in marriage.
But love of all sorts involves reconstituting one’s identity, for love under-
mines the usual boundaries separating the self from another person. The
concerns of a person you love are importantly your concerns. Of course, the
concerns of the people one loves can conflict with one’s personal concerns and
the concerns of others one loves. This is part of what makes love complicated.
The working out of these complications contributes to the formulation of the
stories that constitutes our personal identities. A sense of one’s ongoing exist-
ence as a person involves interaction with significant other people, some of
whom will probably die before oneself.34
One does not cease to love someone because that person dies. Indeed, one’s
sense of one’s love for the person is, at least for a time, intensified by loss.

33
See John Bowlby, Attachment and Loss ( New York: Basic Books, 1980). For a summary of the
attachment perspective, see Mario Mikulincer and Phillip R. Shaver, “An Attachment Perspective
on Bereavement,” in Stroebe et al., Handbook of Bereavement Research and Practice, 87–112.
34
See Jacques Derrida, Politics of Friendship, trans. George Collins (New York: Verso, 1997).
 Love and Death 173

One’s realistic expectations regarding interaction with another person are

irreparably altered by that person’s death; but one’s sense of identity continues
to be constructed in part on the basis of one’s relationship to that person. I do
not cease to think of myself as being my mother’s daughter just because she has
died. Nor do other people who know both of us. In this respect, my mother still
plays a role in my social identity.
Dead loved ones also remain active characters in our psyches. For many
people at least, they continue to play roles in dreams, and not necessarily as
dead people. One also continues to use some of these people (deceased parents,
for example) as guides from whom one takes one’s bearings. Psychologist Nigel
Field describes this in terms of the legacy of the departed, and he points to var-
ious roles that a deceased loved one can play in survivors’ lives. These include
providing “an internal image of his or her comforting presence when under
duress,” or “serving as an important reference point when making important
autonomy-promoting decisions.”35 To eliminate the connection with the other
person not only seems disrespectful of the relationship that one has had; it
undermines what remains of one’s sense of identity at a juncture when it is
already threatened because of the disruption of one’s social status that loss has
produced.

The Story in Grief

My second reason for rejecting the relinquishment model is that it ignores the
role of the story in grief. I would propose that grief is all about the story. Both
as mourners and as those who would comfort them, we need a narrative model
of grief, akin to the one Bob claimed we need toward our own deaths.36 Alasdair
MacIntyre emphasizes the importance of narrative to both personal identity
and our ability to share social understandings of what is going on.37 If we need
stories to make sense of our experience, they seem particularly important at
times when our experience is as inchoate and insistent as it is in the context
of grief.
Indeed, those who are grieving commonly do make use of stories in pro-
cessing grief. One of the problems caused by disenfranchised grief is that the
lack of social support also makes it difficult for the bereaved to find occasions

35
Nigel P. Field, “Whether to Relinquish or Maintain a Bond with the Deceased,” in Stroebe et
al., Handbook of Bereavement Research and Practice, 118, 120–121.
36
Some psychologists have argued in favor of this approach. See Arnar Á rnarson, “Biography,
Bereavement, Story,” Mortality 5 (2000): 189–204. See also Walter, “A New Model of Grief,” 7–25.
I will consider some of their views directly.
37
See Alasdair MacIntyre, After Virtue: A Study in Moral Theory (Notre Dame, IN: University of
Notre Dame Press, 1981), 201.
174 on emotions

for talking, constructing narratives, about the loss.38 Shakespeare recognizes

the therapeutic value of expressing grief. In Macbeth Malcolm says to Macduff,
“Give sorrow words. The grief that does not speak, knits up the o’erwrought
heart and bids it break.”39 Rosenblatt observes that telling stories while griev-
ing is a widespread phenomenon:

An important part of the grieving of quite a few adults in many societ-

ies is to develop a narrative about the person who died, how the death
came about, what the death means, what the bereaved person’s rela-
tionship is with the deceased, and what has happened as a result of
the death.40

Psychologist James W. Pennebaker has experimentally demonstrated that tell-

ing stories about traumatic experiences and one’s feelings surrounding them
has positive impact on physical as well as mental health.41 Presumably, being in
a state of grief would be a circumstance in which narrating stories would have
these beneficial effects.
A recent development in the psychology literature is discussion of narrative
models of grief. Tony Walter argues that talking about the deceased, and par-
ticularly constructing a “durable biography,” is key to finding an appropriate
place for the person in the grieving person’s on-going life.42 He argues that talk-
ing with others who knew the person facilitates this process by enabling the
mourner to subject his sense of who the person was to a kind of reality-testing
(albeit a very different kind than that suggested by Freud). By discussing the
deceased person, the mourner compares notes, as it were, with others who
mourn the loss. By settling on a relatively stable story, a grieving person is bet-
ter able to find a secure place for the deceased in his or her life. One of the vir-
tues of Walter’s account is that it acknowledges the social dimension of grief.
I think Walter is right that we want to fi nd a secure place for deceased loved
ones in our lives and that we use stories to do this. Particularly in the disori-
enting days following a loss, when memories of all sorts may arise almost at
random, gaining a clear impression of the deceased person certainly amounts
to progress in coping. I have two reservations about Walter’s model, however.
One is that his emphasis on reality testing and the durable character of the

38
See Rosenblatt, “Grief across Cultures,” 215.
39
William Shakespeare, Macbeth, Act IV, Scene iii.
40
Rosenblatt, “Grief across Cultures,” 210.
41
J. W. Pennebaker and J. D. Seagal, “Forming a Story: The Health Benefits of Narrative,”
Journal of Clinical Psychology 55, no. 10 (1999): 1243–1254; and James W. Pennebaker and Robin C.
O’Heeron, “Confiding In Others and Illness Rate among Spouses of Suicide and Accidental-Death
Victims,” Journal of Abnormal Psychology 93, no. 4 (1984): 473–476.
42
Walter, “A New Model of Grief.”
 Love and Death 175

biography to be constructed may short-change the extent to which the con-

tinuing relationship with the deceased is nurtured through ongoing reflec-
tion and reconstruction of the story. Constructing a narrative account of the
deceased serves the dual functions of (1) providing a clear enough picture of
the deceased to restore a sense of the person and where he or she fits within the
survivor’s own life, and (2) symbolically reanimating the dead.
I was gratified when, after a talk about Bob that I was asked to present at a
conference, a couple of people from the audience told me, “I never met him, but
now I almost feel like I know him.” Bringing a sense of who the person was to
people who are present revivifies the deceased loved one in imagination, and it
is satisfying for this reason. The dead person lives in the stories told about him
or her. Part of the value of jointly working out a story about the deceased is that
it enriches each person’s conception of the lost person. It reawakens and even
enhances everyone’s sense of what the person was really like.
Narratives also symbolically reanimate of the dead because they allow fresh
insights, recalling something of the continual potential for surprise in an ongo-
ing relationship. A story invites interpretation, and its meaning remains in flux
as the interpreter reconsiders various features of it.43
My second reservation about Walter’s model is that it does not sufficiently
acknowledge the role of narratives about bereavement and grief itself. Those
who are grieving feel virtually compelled to narrate stories about how their
loved ones’ deaths came about from their own points of view. Indeed, this is
often what others who knew the deceased are interested in, although trau-
matic repetition of such sagas can surely be exhausting to listen to. Situating
the story of the death and the mourner’s behavior afterward within the larger
story of the relationship helps to make sense of what happened and gives the
grieving person some sense of actually living, of having experiences, when one’s
normal life has broken down. Walter sees sharing stories with others who have
experienced loss as second best after joint narration about the deceased, but
such sharing legitimates the mourners’ experiences as being stuff of narrative.
Relating one’s own stories to those of others can help one attain perspective.
Most importantly, it can provide the bereaved person with a sense of still hav-
ing a life.
Arnar Á rnason criticizes Walter’s biographical account on grounds some-
what similar to my own. He emphases the creative character of telling stores
about the deceased. He also observes that the mourner’s account of the

43
One could call this reinterpretation of a story, but the bereaved person carries the narrative
of a deceased loved one within the psyche, and reinterpreting it amounts to revising the story.
The continued evolution of the narrative on further reflection allows the bereaved person to be
able to recapture something of the “live” character of the relationship, in which the other party in
the relationship remains somewhat unpredictable, however well one knows that person.
176 on emotions

deceased loved one merges with an account of the mourner’s own life. I think
this is important, and it reflects the role the deceased person still plays in the
bereaved person’s sense of identity. Á rnason also draws attention to the fact
that mourners’ stories make use of conventional genres and plot lines to make
sense of life experiences. He describes an informant who presents himself and
his late wife as “hard-working, decent, ordinary working-class people, whose
marriage was based on both love and devotion.”44
Á rnason calls the construction of narrative on the basis of standard models
emplottment. Emplottment helps the bereaved to make a kind of sense of their
experience through models offered by the society. Not only do such accounts
draw on social narrative types to make sense of one’s life before and after
bereavement. According to Á rnason, his informant’s story presents the latter’s
actions as “social acts because they are attempts at establishing an argument,
a justification that explains his relationship with Doreen and his experiences
after her death.”45
Á rnason’s elaboration on the role of narratives makes apparent several
advantages of taking a narrative approach to grief as well as to the life of the
dead person. One is that it acknowledges the interconnection between the
story of the grieving person and the story of the deceased.46 By doing so, it
reveals the necessity of a continued role for the deceased within the life of
the survivor, as essentially involved in the story of his or her own life, with a
plot that continues to evolve. Second, the narrative of the relationship before
and after the death can help the bereaved justify himself or herself against
the onslaught of feelings of guilt and incapacity. The story helps to make the
survivor’s behavior appear legitimate by presenting it as coherently motivated
and integral within the story. Third, the narrative approach reflects the emo-
tional complexity of grief itself. Grief involves an extended temporal trajectory
with emotional complications. Any model that calls for quick recovery does not
acknowledge this.
Fourth, a narrative model of grief can do justice to the fact that the trajec-
tory of grief is “irreducibly singular” (in Sartre’s expression). Each person’s
grief story is unique, and thus healthy grieving cannot be defi ned in terms of
statistical averages. As Kastenbaum puts it, “The closer we come to the experi-
ential core of grief, the less we can rely on orthodox objectivistic measures.”47
Fifth, the narrative model is consistent with a feature of grief observed by
Janet McCracken, the desire it involves to commemorate the deceased. She

44
Á rnason, “Biography, Bereavement, Story,” 201.
45
Ibid.
46
By contrast, Walter proposes his account in opposition to bereavement counseling that
focuses on the feelings of the bereaved, and he emphasizes the importance of narratives being
about the deceased, not the mourner.
47
Kastenbaum, “Grieving in Contemporary Society,” 71.
 Love and Death 177

terms this the dedicatory character of grief.48 Grief impels us to want to do

something to honor the deceased person and to fulfi ll that person’s wishes.
We want to continue the story of the lost loved one by carrying out that
person’s wishes, acting on that person’s behalf to execute his or her will. The
pun here is intended. The legal activity of executing the will of the deceased
reflects a more basic desire of those who love the person to restore his or
her willful activity within the life world. Trying to do what the deceased
would have wanted is another symbolic effort to reanimate the dead. It is
also a means of continuing the conjunction of the deceased person’s story
with one’s own. The things one might want to do to honor the late loved one
can encompass a wide scope. Almost any effort on the part of the survivor
might, in principle, be undertaken in commemoration of the deceased per-
son, although some such commemorations would be odd. One could even
consider the effort to conduct one’s continuing life admirably to be dedica-
tory in honoring the dead.
Finally, the narrative approach acknowledges the potential for growth
through grief. Christopher G. Davis points out that in dozens of studies of
bereaved persons, “people frequently report that they have greater appre-
ciation for life, that they no longer take things for granted, that they are
closer to loved ones, and that they are stronger and more self-confident.”49
Even though this list sounds unbelievably upbeat from the perspective of
the newly bereaved, the possibility of personal development and spiritual
growth as a consequence of grief should be acknowledged, and any account
of how such growth can occur will necessarily be structured in narrative
terms.

Conclusion
Although we tend to think of our lives and other people’s lives as stories, we sel-
dom think of grief this way. I suspect that this is because grief seems to mark
the end of a story. Bob’s suggestion that grief is a continuation of love amounts
to an assertion that grief does not end a story. Instead, it is a chapter within
the relationship and within the survivor’s life narrative. Th inking of it in this
way can help the bereaved regain a psychological foothold in the on-going flow
of life. Considering grief as part of a narrative acknowledges the continued
connection with the deceased and with the many other people who play roles

48
Janet McCracken, “Falsely, Sanely, Shallowly: Reflections on the Special Character of Grief,”
International Journal of Applied Philosophy 19, no. 1 (2005): 139–156.
49
Christopher G. Davis, “Redefi ning Goals and Redefi ning Self: A Closer Look at Posttraumatic
Growth Following Loss,” in Stroebe et al., Handbook of Bereavement Research and Practice, 311.
178 on emotions

within the mourner’s life. Rather than urging the bereaved to get over grief, we
would do well to encourage them to go on spinning the tales of their lives, tales
in which deceased loved ones remain integral to the fabric, as they were when
they were alive.

Acknowledgments
I wish to thank Colleen Cook, Agneta Fischer, Betty Sue Flowers, Henry Frankel,
Ali Gallagher, and Clancy Martin for their helpful suggestions regarding this
paper.
 9

Guilt in War
Nancy Sherman

What makes guilt particularly interesting to Freud and the psy-

choanalysts as well as to novelists like Dostoevsky and Kafka,
is the fact that one can feel guilty despite the fact that one has
done nothing wrong. . . . One might think that this feeling of
guilt without actual guilt is very much like . . . irrational fear,
where one is afraid of something that one knows not to be dan-
gerous. But I think the cases are quite different, first because
no one has proposed anything remotely like a brain mechanism
for guilt, as has been so elegantly demonstrated in the case of
fear reactions, second because there is a complexity to guilt
that far outstrips anything that we saw in primitive fear.
Robert Solomon, “Self-Reproach in Guilt,
Shame, and Pride” in True to Our Feelings,
New York: Oxford University Press, 2007, 92

My topic is that which Bob Solomon points to in this passage, namely, feeling
guilty “despite the fact that one has done nothing wrong.”1As Solomon sug-
gests, subjective guilt is not a simple kind of irrationality, on the model of being
afraid when one knows there is nothing there to fear. Nor is it easily dismissed,

1
For other important discussions, see Herbert Morris, “Nonmoral Guilt,” in Responsibility,
Character, and the Emotions: New Essays in Moral Psychology, Cambridge: Cambridge University
Press, 1988, 220–241; John Deigh, “All Kinds of Guilt,” in Emotions, Values, and the Law, Oxford:
Oxford University Press, 2008, 159–170 (originally presented as part of a symposium on Herbert
Morris and published in Law and Philosophy 18 (1999): 313–325). Also, in the background of
my remarks is P. F. Strawson’s influential article on the place of reactive emotions in recording
and constituting a sense of holding oneself and others responsible: “Freedom and Resentment,”
Proceedings of the British Academy 47: 1–25, reprinted in Free Will, ed. G. Watson, 59–80, New
York: Oxford University Press, 1982. For a helpful discussion of issues surrounding the notion
of holding oneself and others responsible, see R. Jay Wallace, Responsibility and the Moral

179
180 on emotions

he muses, as a mere artifact of the Catholic Church or overbearing Jewish

mothers. It is a morally significant and complex phenomenon and, on my view,
not well understood until we look at particular concrete cases in which it is felt
and, all too often, not easily relieved. While subjective guilt may not be fully
rational, it is not as irrational as it is often thought to be.
My recent work has been with soldiers, soldiers of the current wars.2 I have
interviewed some forty soldiers at length, sometimes alone, others in multiple
interviews, collecting narratives about the arc of combat—what it is like to
put on the uniform, deploy, and return home. I have listened with a philo-
sophical and psychoanalytic ear, 3 as one interested in the moral emotions
and the confl icts they reveal. With less than 1% of the population fighting
the wars in Iraq and Afghanistan that have now raged for over a decade, most
of us do not intimately understand the moral burdens of soldiering in these
wars. We overlook the fact that psychological anguish and trauma is often
moral in nature and that for many a soldier war’s shadow lingers long in the
emotions of guilt, shame, and the frustrations of limited agency.4 These mix
with intense love and solidarity and a thirst for payback when the enemy kills
one’s own. The payback can turn inward when one kills one’s own by accident
or somehow fails to protect those one loves or innocents who are helpless to
protect themselves.
Indeed, what gripped me the most as I listened to soldiers was the sense of
guilt. Guilt was often the elephant in the room. It was a hard feeling for soldiers
to articulate, but it fi lled their thoughts. The guilt was primarily subjective
guilt—guilt that did not track, in any straightforward way, moral culpability.

Sentiments, Cambridge, MA: Harvard University Press, 1996. For a critique of Wallace, see Coleen
MacNamara, “Holding Others Responsible,” Philosophical Studies (2011) 15: 81–102. For a dis-
tinction between two kinds of appropriate emotions, see Justin D’Armes and Daniel Jacobson,
“The Moralistic Fallacy: On the ‘Appropriateness’ of Emotions,” Philosophy and Phenomenological
Research, (July 2000) 61, no. 1: 65–90.
2
See The Untold War: Inside the Hearts, Minds, and Souls of our Soldiers, New York: W.W. Norton,
2010. Th is essay is a reworking of Ch. 4 of that book. See also Stoic Warriors: The Ancient Philosophy
behind the Military Mind, New York: Oxford University Press, 2005. Despite service rivalries and
terminological differences, I use “soldiers” here inclusively to include service men and women
in all the U.S. Armed Forces, that is, the Army, Marines, Navy, Air Force, National Guard, and
Reserves.
3
While I have research training in psychoanalysis, I neither am a therapist nor viewed myself
in a therapeutic role in my interviews with soldiers.
4
Th is is especially obscured in the posttraumatic stress literature, which stresses three classes
of symptoms: persistent hyperarousal or hypervigilance; recurrent intrusive recollection or fl ash-
backs; and persistent numbing and social avoidance. See, for example, the restatement at the U.S.
Department of Veterans Affairs website, and its National Center for PTSD (posttraumatic stress
disorder; http://www.ptsd.va.gov/professional/pages/dsm-iv-tr-ptsd.asp). For some, the term
“disorder” is highly stigmatizing and exacerbates the differences between visible limb injuries,
which are not thought of as disorders, and invisible mental injuries, which are.
 G u i l t i n Wa r 181

Over time, I came to think of that guilt as taking three principal forms: acci-
dent guilt, luck guilt, and for want of a better term, collateral damage guilt.

Accident Guilt
Some soldiers blamed themselves for mishaps in equipment that took the lives
of their buddies or the lives of innocents, though there was no negligence or
culpable ignorance for which they could be held morally or legally responsible.
In one wrenching case, the gun on a Bradley fighting vehicle misfired, blow-
ing off most of the face of a private who was standing guard near the vehicle. It
was an accident due to a faulty replacement battery. Authorization was given
to use a Marine battery in place of the original Army one. This Marine battery
had the same voltage as the Army one, but different amperage. That made all
the difference. The aftermath of this accident is that the marine battery is now
banned from the use in these Army vehicles.
The army officer in charge reconstructed the scene for me, narrating every
detail, the way a person who has relived the scene over and over does:

It was as if an ice cream scoop just scooped out his face. . . . He survived
the initial blast, if you can believe it. We were in the medic tent with
him. It was one of the most traumatic things I have ever seen in my
entire life. To literally see someone’s face completely scooped out, to
see just the very bottom part of his jaw working. . . . He couldn’t see,
couldn’t hear, couldn’t scream. . . . I mean, he had no eyes, obviously.
No face. I can only imagine the terror, the fear, the pain he was in. He
obviously couldn’t breathe because he had no nose or mouth to take in
air. . . . It was one of the few times in my life I’ve really cried—tears just
streaming down my face because I’m watching 10 people work over
this kid. . . . It was an unbelievable thing to see. . . . It is one of those
images that will be in your head until you die.5

He then turned to his feelings of responsibility: “I’m the one who placed the vehi-
cles; I’m the one who set the security. [As with] most accidents, I’m not in jail right
now. . . . I wasn’t egregiously responsible. . . . Any one of a dozen decisions made
over the course of a two-month period and none of them really occurs to you at
the time. Any one of those made differently may have saved his life. So I dealt
with and still deal with the guilt of having cost him his life essentially. . . . There’s
probably not a day that goes by that I don’t think about it, at least fleetingly.6”
Bernard Williams and Thomas Nagel have put us philosophers in the habit
of thinking of this feeling as a kind of regret, not about mere happenstance,

5
The following is based on an interview with John Prior on September 27, 2006.
6
Ibid.
182 on emotions

such as the weather, but of the weighty consequences in the real world of
our agency. Hence, “agent-regret:” one may be “causally responsible,” though
not morally responsible “in virtue of something one intentionally did.” 7 But
regret doesn’t begin to capture what Maj. John Prior feels. It doesn’t capture
the moral phenomenology—the despair or depth of the feeling—the awful
weight of self-indictment, the empathy with the victim and survivors, and the
need to make moral repair in order to be allowed back into a community in
which he feels he has somehow jeopardized his standing. When Maj. Prior says
he doesn’t hold himself as “egregiously responsible,” he means that he knows
he didn’t commit a careless blunder. But he still doesn’t think of himself as
fully, morally cleared. In his case he sought moral repair through an empathic,
painful connection with Mayek’s mom. After the fratricide, Maj. Prior and his
first sergeant dutifully wrote a letter to Mayek’s mother. For some time after,
she would send care packages to the company with letters. “Oh it was terrible,”
said Maj. Prior. “It was horrible.” The letters weren’t “just very matter of fact—
here’s what we did today; it was more like a mother writing to a son.” For this
mother, Maj. Prior and his company became the son who was no longer. “It
was her way of dealing with the grief. . . . And so I had a responsibility to try
to give back.”
It is important to be clear here that in some ways Maj. Prior’s accident guilt
is strictly speaking irrational. There is no wrongdoing in respect to which the
guilt feeling is fitting. It distorts and aggrandizes moral accountability. In that
sense it seems inappropriate. Yet it is all too appropriate morally. It is reason-
able. We understand it and admire him for it. We would think less of him as a
commander, ready to lead his troops and have them follow, if he viewed what
happened as just an accident, only remotely connected with his own choices.
The guilt records something morally significant, something that marks his deep
connection to his troops and his moral accountability to them and to himself. It
is not just a response to the fact that he was causally implicated. He feels mor-
ally implicated, even if the bad consequences were ones he couldn’t reasonably
expect or prevent. How he deals with the repercussions of his actions, however
nonvoluntary, becomes morally defining for him, in a personal way and as a
leader.8 Though military men and women often think of themselves as stoic,

7
Williams, “Moral Luck,” in Moral Luck, Cambridge, UK: Cambridge University Press, 1981,
28. Also, see Thomas Nagel’s important account of luck in “Moral Luck,” in Mortal Questions,
Cambridge, UK: Cambridge University Press, 1979. For an important discussion of these pieces,
see Susan Wolf, “The Moral of Moral Luck,” in Cheshire Calhoun, ed., Setting the Moral Compass:
Essays by Women Philosophers, New York: Oxford University Press, 2004. For the notion that moral
luck highlights the idea of our personal agency as “impure,” in the sense of requiring resources
of character to meet the combination of choice and fortune in life, see Margaret Urban Walker,
“Moral Luck and The Virtues of Impure Agency,” Metaphilosophy (1991) 22: 14–27.
8
For insightful remarks here, see Wolf, “Moral of Moral Luck.”
 G u i l t i n Wa r 183

and here in the tradition of the ancient Stoics—restricting one’s happiness to

what one can control, in the case of a nonnegligent accident like this, morally
worthy responsibility, here, typically goes beyond what one can control to how
one responds to and accepts contingency.9
Of course, guilt that can be admirable can easily turn pathological. Nietzsche
is the modern philosopher who understands this so well: “Bad conscience”10
can become “torture without end,”11 undoing any prospects for happiness. This
subjective guilt, he says, doesn’t grow in the soil where you would most expect
it, such as in prisons where there are actually “guilty” parties who should feel
remorse for wrongdoing. Rather it is often a “question of someone who . . . caused
harm,” that is, someone who causes a misfortune for which she is not strictly,
morally culpable.12
He appeals to Spinoza for support, who wrote: the “bite of conscience” has
to do with an “offense” where “something has gone unexpectedly wrong”; it is
not really a case of “I ought not to have done that.”13 Freud famously elaborates
upon the theme: enduring the recriminations of a harsh superego (“a bad con-
science”) is the cost of civilization. Essentially these moderns document what
ancients, like Seneca in De Ira, warned long ago about outward-facing rage. The
fury of wrath can leave its possessor more tormented than satisfied. So too, the
moderns argue, the fury of self-directed anger.
We might plausibly look to Kant for lessons in understanding guilt, for he,
after all, formalizes the idea that morality is primarily a first-person preoccupa-
tion requiring vigilant assessment of one’s actions and omissions and efforts
at promoting obligatory ends. Guilt just is a self-regarding or first-person emo-
tion of moral appraisal. But despite the focus on self-judgment, Kant is neither
particularly interested in what from ancient times forward has been described
as the “bite”14 of bad feelings for doing wrongs nor the fear of punishment
that can prod us to do what is right. The absence is deliberate. Kant’s prevail-
ing concern is with the mature conscience. The attitude of that conscience is
not pain that gnaws or fear that goads and bullies, but sublime respect or rev-
erence in the sense of attention and submission (Achtung) to our own author-
ity as moral legislators. Indeed when the subject of the anguished emotion of
repentance comes up, Kant warns against a morality that is “cheerless, morose,

9
For a fuller discussion of the stoic ethos of the military, see Sherman, Stoic Warriors.
10
Friedrich Nietzsche, On the Genealogy of Morality, edited by Keith Ansell-Pearson. Translated
by C. Diethe. Cambridge: Cambridge University Press, 1994, 60–61.
11
Ibid., 68.
12
Ibid., 59; see 60–66.
13
Ibid., 60. He refers to Spinoza, Ethics, Proposition 18, Scholium 1 and 2.
14
For a discussion of the Stoics on the feelings that bite (for example, morsus, in Cicero), see
Margaret Graver, Cicero on the Emotions: Tusculan Disputations 3 and 4, Chicago: Chicago University
Press, 2002, 36, 45, 84–85, 125–127, 140.
184 on emotions

and surly”—a self-punishment that becomes “self-torture.” Beware of “hypo-

critical self-loathing” and the moral melancholy that detracts from the real
work of morality, Kant warns.15 True virtue ought to be cheerful and soar with
the sublimity of respect for its law. But in all this, Kant doesn’t appreciate that
stern self-judgment is not necessarily self-righteousness. It can be inseparable
from empathy with those we harm and a sense of responsibility and duty of
reparation even when the harm is not intentional, or intentional and warranted
but no less loathed, as in just killing in war. The self-judgment, the guilt can be
redemptive.
Here is precisely where we think Freud should be able to do better than Kant
in recognizing the place of guilt felt for actions that are not, strictly speaking,
culpable. And he does in some ways, but not entirely. Freud famously tells a
“just so story” of how morality comes to be internalized as part of our person-
alities: our conscience is the product of a childhood battle of sorts. As children
we both want parental love and fear its loss and repudiation in punishment
when we do wrong and “are found out.” With moral growth and independence,
we move that tension inside ourselves, into an internal representation of
authority and regulation that stands over and above our ego or self; thus, the
super-ego is born. Through that superego we become our own judges, vigi-
lant and fearful of self-indictment. So Freud declares, “the superego” is the
Kantian “moral law within us.”16 Guilt, on Freud’s view, is a kind of leftover
tension from childhood;17 its predominant feeling is fear of punishment and
retaliation.
But despite the Kantian legacy, Freud does not really offer an account that
internalizes the moral law. Indeed, he does not offer an account of morality.
Moving parental or conventional authority from outside to in is not enough
to ensure that that internal authority and judgment is legitimate or really
moral. We can bully ourselves and be harsh tormentors without justification,
just as parents can be cruel or military leaders overbearingly harsh. What
the move from outside to in does do, as Freud rightly insists, is set up “a gar-
rison” whose watch is inescapable.18 Th at is, if childhood morality is about
“being found out,” then once an internal garrison is set up, “nothing can be
hidden” from self, “not even thoughts.” 19 Of course, Freud famously leaves
room for the work of repression and self-deception. But we need to veil our-
selves from ourselves precisely because we have become our own watchdogs.

15
Kant, The Doctrine of Virtue, e. M. J. Gregor, Philadelphia: University of Pennsylvania Press,
1964/1797, 485.
16
See Freud, New Introductory Lectures, in Standard Edition (SE) of the Complete Psychological
Works of Sigmund Freud, trans. J. Strachey, London: Hogarth Press, 1955, 22: 61, 163.
17
For example, in Civilization and Its Discontents, SE 21:123.
18
Ibid. 21: 125.
19
Ibid., 21: 125.
 G u i l t i n Wa r 185

Still, without some clever philosophical patching up of his theory,20 Freud

cannot explain why some self-indictments from that watchdog have at their
base guilt that is more rooted in objective culpability than in the subjective
bullying of a harsh superego. Nor, to bring us back to our current worries,
does he really seem to appreciate how very thin the wedge can be between
the two and in what sense the subjective guilt can itself be morally signifi-
cant. Th is is at the heart of understanding “the personal guilt” that Maj.
John Prior carries with him. And it is also at the heart of understanding
other kinds of subjective guilt that soldiers experience.

Luck Guilt
Many I spoke to experienced “luck guilt,” a generalized form of “survivor guilt.”
“Survivor guilt” is itself a relatively new term, but the phenomenon is not. The
term, when introduced into the psychiatric literature in 1961, referenced the
severe guilt felt by survivors of Hitler’s genocide of European Jewry. Those who
survived the genocide felt they were the “living dead;” they carried the uncon-
scious belief that “merely remaining alive was a betrayal of the dead.” 21 Strictly
speaking, survivor guilt is not rational guilt, for surviving the Holocaust, or
surviving battle, our theme, is not typically because a person has deliberately
let another take his place in harm. To be sure, it may be like that. Achilles’
guilt over the death of his beloved war comrade, Patroclus, is, perhaps, one
of the earliest records we have of a soldier’s survivor guilt. And in this case,
Patroclus was killed, to reconstruct the moment, because Achilles refused to
fight and Patroclus took up Achilles’ shield and armor and did battle for him.
What Achilles wants more than anything is to undo the original switch, to
trade places not just in fantasy, but in fact, to suffer the harm and endure the
death himself: “My spirit rebels—I’ve lost the will to live.” “The man I loved
beyond all other comrades, loved as my own life—I’ve lost him.” 22 He can’t
undo the switch and he feels profound guilt.
But the guilt of surviving war buddies, or of not being exposed to exactly
the same lethal dangers, is not always or typically like this. It is often a mat-
ter of dumb luck that a soldier makes it out alive, and tragic luck that others

20
David Velleman tries to do just this in melding the discussion of the superego with Freud’s
notion of an ego ideal that represents enlightened moral ideals and the inculcation of those ego
ideals through parental love of children that fosters independent moral judgment. Velleman,
“A Rational Superego,” Philosophical Review (1999) 108: 529–558.
21
See Lawrence Friedman’s insightful discussion of William Niederland in Friedman,
“Towards a Reconceptualization of Guilt,” Contemporary Psychoanalysis (1985) 21: 501–547.
W. G. Niederland, “Clinical Observations on the “Survivor Syndrome,” International Journal of
Psychoanalysis 49: 313–15; also Niederland, “Further Observations and Dimensions,” Journal of
the American Psychoanalytic Association 29(2): 413–25.
22
Iliad 18.105, 94–96, using Fagles translation. New York: Penguin, 1999.
186 on emotions

do not. Yet many who have that good luck feel that it is a betrayal of those
who don’t—a betrayal of solidarity. It is not unlike what is sometimes felt
in a family, when one sibling does well and another does not. The child who
succeeds may feel guilty, experienced as if the distribution of goods is based
on a zero sum game: to do well means she has deprived, in some noninnocent
way, her sibling of limited goods. Psychoanalyst Arnold Modell describes aptly
this kind of guilt. There is “in mental life something that might be termed
an unconscious bookkeeping system, i.e., a system that takes account of the
distribution of the available ‘good’ within a nuclear family so that the current
fate of other family members will determine how much ‘good’ one possesses.
If fate has dealt harshly with other members of the family the survivor may
experience guilt.” 23 Soldiers, as a result of tight unit cohesion and solidarity,
may have similar unconscious thoughts—that the cost of their good luck is
another soldier’s bad luck. To be lucky is to deprive a buddy of comparable good
fortune. Though they do no wrong, they blame themselves as a way of sharing
the ill fate. Sharing the evil is a way to negate the awful sense of betrayal.
Marines I interviewed in Annapolis, shortly after their return from Baghdad
after its fall, anguished about their undeserved luck at being in the scenic sail-
boat setting of the Naval Academy, far away from their brothers and sisters
still at war. Soldiers I interviewed at Walter Reed Hospital, themselves severely
wounded, felt guilty for not suffering more, or as visibly, with limb loss or facial
disfigurement. Their good luck was a betrayal of those who lacked it. What
they felt, at bottom, was deep empathic distress of the sort Melanie Klein, the
British object-relations psychoanalyst writes about—guilt felt for injury and
harm inseparable from empathic love. As Klein understands it, we preserve as
residue from early childhood, the anxiety that we have hurt and persecuted—
physically damaged—the very objects that we depend upon for sustenance and
love. On her view, guilt is not, as Freud thought, rooted in anxiety about doing
wrong and fear of being punished for that—“castration anxiety,” archetypi-
cally. Rather guilt is rooted in a different residue from early childhood—from
the experience of harming or betraying what one loves and depends upon.24 In
their own eyes, these soldiers felt they had failed to take care of their buddies;
they had broken a bond of solidarity, and even worse, failed, they worried, to
honor the duty of fidelity that enabled them to fight in the first place.
Maj. Michael Mooney, an intelligence officer, was part of a lead battalion
that on March 23, 2003, took catastrophic losses in an attack in Nazariah,

23
Arnold Modell, “The Origin of Certain Forms of Pre-Oedipal Guilt and the Implications for
a Psychoanalytic Theory of Affects,” International Journal of Psychoanalysis (1971) 52: 337–346, as
quoted in Friedman, “Towards a Reconceptualization of Guilt.” 521–22.
24
Melanie Klein, Love, Guilt and Reparation and Other Works 1921–1945, New York: Delta,
1977.
 G u i l t i n Wa r 187

Iraq.25 Eighty were killed and nearly another eighty were wounded in one
day. The attack came just after a rescue operation in which the same battal-
ion pulled out the dead and wounded from the freshly ambushed convoy of
the 507 Maintenance Company (made famous by Jessica Lynch’s story). Two
months later, in mid-May, the battalion was ordered back to Nazariah for
peace and stability operations. The troops were jittery about returning to the
site of so much carnage. “Jittery” is a word some used, a somatic word for the
anxiety, like “braced” or “fl inching,” as you will hear in a moment. “I remem-
ber driving into this city from the north this time, almost fl inching, bracing
as we were going down Ambush Alley, because I remember the last time I
was in it. . . . [we were] still seeing the charred marks on the road where we
lost amtraks [amphibious tractors]—catastrophic kills where every Marine
in that amtrak was killed in action.” Soon after a battalion-wide memorial
service on the very fields where Marines had fallen, Maj. Mooney got orders
to report immediately to the Naval Academy. He left behind his regiment
who would sail back slowly to the States. He was put on a commercial, not
military, fl ight, sitting next to civilians who simply occupied a different world
from his.
The rupture with his unit is palpable, but so too the expectation that, as a
Marine, he should be able to just carry on: “I didn’t have the same experience
or opportunity as the other Marines leaving the combat zone. I had orders to
report to the Naval Academy. I had to get here very quickly or I wasn’t going
to be able to execute those orders. It was around May 15. . . . So as my unit was
backloading, moving down to Kuwait, I had caught a plane, got shuttled down
to Kuwait City, and was thrown on a British Airways flight. Seven hours after
being in Iraq, I was in Frankfurt, Germany, sitting in a first class lounge. It was
very surreal. Trying to actually look at the porcelain toilets—‘cause the last
time I had taken a shower was I think March the 7th—and then 10 hours after
that I was meeting my family and my wife-to-be . . . in Reagan National, which
was also very surreal. There wasn’t a decompression time . . . Basically, I just had
to start at the Academy. . . . It was a very interesting dynamic being transitioned
so quickly . . . boom, boom, boom.”
When I ask him if he was able to manage with the transition, he answers,
without pause, “Yes, ma’am.”
But seconds later he reveals the true weight he bears: “Just for me, person-
ally, I don’t think there’s a day goes by that I don’t think about those times, just
because of the accomplishments we achieved and the incredible loss of poten-
tial for the Marines, even though I didn’t know them personally. Just seeing
and being present when the Marines were being pulled out of the amtraks. . . . ”

25
I interviewed Maj. Mooney as part of a group in Annapolis on October 21, 2005.
188 on emotions

He says he “reflects back” often on the “lost potential,” his restrained words for
the grief he feels for those he saw die far too young in Nazariah.
Mooney talks about the horrific with steel-like comportment. He carries
his feelings tightly inside. But there is unspoken guilt that he is in tranquil
Annapolis while others continue to fight. He is ready to go back and he is pre-
paring his new wife for that reality: “I say you’ve got to prepare yourself for
this because after sitting here in Annapolis for three years, after wonderful
air conditioning in Annapolis, while my brothers and sisters have been out on
their second and third tours . . . you need to come to come to grips with that I’m
going to be away for awhile.” (As I write, this marine, now a Lt. Colonel, has
deployed to Afghanistan.)
As I listened then to this Marine officer speak about getting back to his
brothers and sisters in arms, my thoughts wandered to another young officer,
Siegfried Sassoon, the British World War I officer and poet. Against the wishes
of his doctor, the eminent Freudian-inspired physiologist and anthropologist,
Capt. W. H. R. Rivers, Sassoon returned to the trenches out of profound mix
of love and guilt and a sense of futility at his war protest (a protest, in part,
inspired by meetings with Bertram Russell). What he now felt, separated from
his troops, was nothing less than abandonment and banishment—I am ban-
ished from the patient men who fight.
“They smote my heart to pity, built my pride.”
To return to the front line might well be a kind of “death,” he writes in his
memoirs, but it is also “my only chance of peace.” “Love drives me back to grope
with them through hell,” he pens in “Banishment.” Another poem, “Sick Leave,”
concludes with the Battalion whispering softly to Sassoon, “When are you going
out to them again? Are they not still your brothers through our blood?”26
In a recent interview I conducted, the theme of luck guilt emerged again,
wrenchingly. The case has to do with the bad luck of being on leave and in posses-
sion of information that could have prevented a buddy’s death. The details of the
case are important for understanding the specific nature of the guilt feelings and
their relation to taking conscientious responsibility for one’s duties of office.

26
I draw here from Paul Fussell’s edition of Sassoon’s Sherston Memoirs: Siegfried Sassoon’s Long
Journey: Selections from the Sherston Memoirs, New York: K.S. Giniger Company, 1983. Also, see
Pat Barker’s Regeneration trilogy: Regeneration, New York: Penguin, 1992; The Eye in the Door, New
York: Plume, 1993; The Ghost Road, New York: Plume, 1995. Sassoon’s attachment to his troops
is, no doubt, complicated by his homosexuality, recently brought to light by a biography by Max
Egremont, Siegfried Sassoon: A Life, New York: Farrar, Strauss, Giroux, 2005. For a discussion,
see Jonathan Thompson, “New Diaries Reveal the ‘Dark Secrets’ of Siegfried Sassoon’s Swooning
Affair,” Independent, October 23, 2005, http://www.independent.co.uk/news/uk/this-britain/ne
w-diaries-reveal-the-dark-secrets-of-siegfried-sassoons-swooning-affair-512165. Also, “Sassoon
Unplugged,” by Andrew Motion, Guardian, December 24, 2005, http://www.guardian.co.uk/
books/2005/dec/24/featuresreviews.guardianreview8.
 G u i l t i n Wa r 189

Tom Fiebrandt served in Iraq between July 2001 and December 2005.27 At
twenty-one he was a young sergeant and a team leader of a group of intelli-
gence analysts attached to an Army cavalry squadron of 410 men in Tal Afar,
a desert town not far from Mosul, about forty miles from the Syrian border.
Like traditional cavalry units, his unit served as the “eyes and ears” of the bat-
talion, collecting and sorting intelligence critical for a dynamic picture of the
current battlefield. The unit was a “bridge” between those inside and outside
the wire, with Fiebrandt himself spending much of his time outside the wire,
talking to troops and locals, drawing and redrawing a fairly visual, firsthand
picture of the vicinity and its dangers. He knew how tall buildings were on dif-
ferent streets, where snipers could lurk, where you did and didn’t want to be.
He became the point guy that noncommissioned officers (NCOs) and officers
alike turned to get their information. As he put it, with modesty but candor, his
superiors “had confidence in his competence.”
About three months before his deployment was up, he was ordered to take
a few days of “R and R” (rest and relaxation) in Qatar before returning to the
States for a longer two-week leave. Fiebrandt was reluctant to abandon the unit
so close to the end of their deployment, but an order was an order, and leave
time was mandatory anyway. He was stressed of late, “bouncing inside and
outside the wire,” as he put it, and at some level, he knew that a break was prob-
ably a good idea.
En route to Qatar he learned that his unit was unit was about to run a cordon
and search operation in the southeast corner of town. Tal Afar had become a
major smuggling hub, with weapons pouring in from unsecured borders with
Syria. It was now time to flush out the weapon caches and insurgents with a
strong show of troop forces and a door-to-door raid. What Fiebrandt didn’t
know was that as part of the preparation, one of the platoons, headed by
Lt. Edens, a close friend, was ordered to scout out a potential egress route out
of the backside of the city, where a wall of troops could be mounted to block
insurgents fleeing the raid into the desert. It was during this preparatory drive-
through that an improvised explosive device struck the lieutenant’s vehicle,
killing him and two others. Fiebrandt learned about the incident a few days
after he arrived at Qatar. It hit him hard:

What bothered me was that it was in an area that I knew very well. It
was in a part of the city that you really had to see in order to visualize.
And I had this lurking suspicion that my soldiers, who had never actu-
ally, personally been there didn’t really have a grasp on all the informa-
tion that I felt I did. In some way, I almost felt responsible for not being

27
I interviewed Fiebrandt in fall 2010. I came to know him through a class I taught at
Georgetown that fall on the ethics of war.
190 on emotions

there to provide them with the information that may have potentially
resulted in a different outcome. So it is rough. It is a difficult thing
for me to process. . . . So here I was sitting by a pool, and I hear this. It
was—I don’t even know how to describe it. It was—devastating.

Had Fiebrandt been there, he is sure he would have recommended against

Edens taking that road. He knew that back area of the city was especially dan-
gerous and that no unit vehicles had traveled down that road for a good reason.
He would have urged more reconnaissance on the routes and potential alterna-
tives. “Whether or not I would have been successful in getting that to become
the battle plan, I don’t know.” But given that he was relied on for this kind of
information, he had a good chance of making the case. In his mind, he let down
his command as well as friend. What happened, as he puts it, “reflected poorly”
on him. He “faults” himself for not being there, and though he is “frustrated”
that his unit members “didn’t have the same clout” as he did and couldn’t “pick
up the slack” in his absence, he doesn’t fault them for failing to make the call.
But significantly, it is just this sense of feeling that he is the only guy who
can do the job and that it is a job that requires constant vigilance, without gaps
and breaks that both hounds him and ultimately opens the way for self-excul-
pation. The fact that he didn’t choose to take the leave—that he was acting on
an order—gets him only so far. The real exculpation comes some three to four
months after the incident, when his deployment is over and he reflects on the
incident in connection with whether he should reenlist. He now sees, somehow,
that the demand he put on himself to be quasi-omniscient, to keep constant
vigil of the changing battlefield, as he puts it several times, without “gaps in his
knowledge,” is unsustainable. He reconstructs the thinking:

Well, God, I thought to myself, if am not here on a two-week period

of time, and things go to hell in a handbasket . . . what is the situation
going to be like when I get back, having been away longer. I am going
to be less equipped to handle any further situations, because now I
have a real gap in my knowledge. So all of this was coalescing at the
same time and it took me a while to sort of realize that I couldn’t be
the person that was there all the time. I could only be in one spot at a
time. I could reenlist and I could stay in the job. But ultimately I am
never going to cover the whole country. I was never going to be the
one-stop intel analyst for the whole Army. Maybe my role was actually
very small.

Looking on from the outside, we might say, “Well, of course.” However well
Fiebrandt served in his role, and however critical he was to the safety of his
unit, he wasn’t there that day, wasn’t at fault for not being there that day, and
 G u i l t i n Wa r 191

wasn’t at fault for not briefi ng in advance his unit about a mission that he
didn’t even know was going to take place. It was bad luck to not be there, trag-
ically bad luck, but he was “off -duty” legitimately, and his fi rsthand, visual
sense of the city, with all its nooks and crannies, wasn’t something that he
could have easily downloaded in some fi le. If asked, he could produce the rel-
evant details, but about this specific route and potential sniper hideouts on
it, there wasn’t the occasion. Perhaps he should have fi led the information
with his unit. But it was more implicit background knowledge that he had,
than an explicit data point for entry. At best, he was minimally negligent, if
even that.
Put this way, holding himself responsible seems grandiose, and Fiebrandt
would likely agree. It requires too idealized a sense of his role-responsibilities
and duties, too idealized a set of expectations and injunctions about how was
supposed to function. Yet the unreasonableness of the demands to which he
held himself dawned on him, with time, only when he realized their absurd
implications—that he was expecting of himself something close to full
omniscience and omnipresence, a constant vigil that could produce an accu-
rate, automatically refreshed picture, without gaps, breaks, and breaches,
of the vulnerabilities of the battlefield. He chuckles as he thinks about the
absurdity of it all and of the reductio that it took to get him to realize it. But it
is a tentative laugh. He still knows the pull of those expectations, and what
it is like to be in their grip. He may no longer endorse the evaluations that
grounded the feelings, but when he says, “I kind of fault myself,” “I almost
felt responsible for not being there,” he still can put himself in the mindset
of what it was like to endorse those evaluations. He is now at a point where
he has moved on. But he got there only through an honest moral struggle
about what it means to be vigilant as an intel guy. There were epistemic fi ni-
tudes and frailties that he had to accept, however compromised they made
his agency.
The Stoics address a related issue regarding intellectual virtues. They warn
that we have epistemic duties to be “nonprecipitous” and “noncareless” in
our judgments about the world, in part, so that we are not derailed by seduc-
tive impressions that are at the heart of paralyzing emotions.28 But in this
case, with these emotions and the appraisals that are at their core, Fiebrandt
wasn’t precipitous or rash. The construal of himself, at fault or somehow der-
elict in his duties, was compelling, or at least compelling enough to block the
counterevidence.

28
See Diogenes Laertius, VII, 46–48 in Lives of Eminent Philosophers, Cambridge, MA: Loeb
Classical Library, 1972. For a discussion of the intellectual virtues involved, see Nancy Sherman
with Heath White, “Intellectual Virtue, Luck, and the Ancients,” in Intellectual Virtue, ed. M.
DePaul and L. Zagzebski, New York: Oxford University Press, 2003, 34–54.
192 on emotions

Like many soldiers I have spoken to, Fiebrandt doesn’t readily use the word
“guilt.” He speaks instead of “fault” and “responsibility.” No doubt, too, there
is shame—that in this perceived breach he saw himself as less than what he
hoped he is and aspires to be. Shame is, in part, that sense of falling down in
one’s own estimate and bearing the pain of that deflation.29 It is not implau-
sible to think that holding himself responsible is expressed both in guilt at
indirectly harming a friend but also shame at being someone who could fall
short or betray in that way.30
Part of the recovery of full humanity after war involves reconstructing
moral agency. Given the contributions of my actions and omissions to some
awful event, am I to blame or not? What are the reasonable boundaries of my
epistemic responsibilities and autonomy in a hierarchy with clear command
structure? In Fiebrandt’s case, he doesn’t know the answers until he carefully
revisits what happened. He doesn’t know if he was duly conscientious and
nonnegligent until he figures out the standards he is holding himself to and
whether those standards are reasonable.
In some ways this is a matter of taking seriously what Kant would call duties of
moral self-perfection, some of which are specified (or “schematized”) in terms of
circumstance and role.31 But it is worth remembering that Kant always insists on
the fairness of oneself as moral judge; that is the aspiration of universal reason,
whether in guarding against temptations to exempt oneself in the case of perfect
duties, or in figuring out the requirements of imperfect, but obligatory, duties of
virtue or end, and what counts as enough or too much where there is discretion-
ary latitude. Kant is famous for warning against moral fanaticism and supermer-
itorious conceptions of duty that ignore one’s own true needs and standing as an
equal among others.32 But to be fair, Fiebrandt isn’t thinking of himself as heroic
or self-sacrificing when he thinks back about the incident. He is thinking simply
in terms of doing what a good soldier is supposed to do. Fiebrandt may or may not
be a fair judge of this, and that is, in part, what he also must assess. But he may
not be in a position to be a fair judge until time has elapsed and the pain of the
loss, and the self-persecution that seems so redemptive in the face of loss, abates.
Only then is the process of self-forgiveness in order.

29
For some accounts of shame, see John Rawls, A Theory of Justice, Cambridge, MA: Harvard
University Press, 1971; Bernard Williams, Shame and Necessity, Berkeley: University of California
Press, 1993; John Deigh, “Shame and Self-Esteem: A Critique,” Ethics (1983) 93, no. 2: 225–245;
David Velleman, “The Genesis of Shame,” Philosophy and Public Affairs (2001) 30, no. 1: 27–52.
30
See Jay Wallace’s helpful remarks on disambiguating the two stances, 239–241. See Rawls,
A Theory of Justice, Cambridge, MA: Harvard University Press, 1971, 482. Also, for general back-
ground, see Gabrielle Taylor, Pride, Shame, and Guilt, Oxford: Oxford University Press, 1985.
31
Doctrine of Virtue, 468.
32
For further discussion of Kant on supererogation, see Nancy Sherman, Making a Necessity of
Virtue, New York: Cambridge University Press, 1995, ch. 8.
 G u i l t i n Wa r 193

Collateral Damage Guilt

The most troubling kind of guilt I heard about had to do with accidental or
unintended but foreseen killings of innocents—what I call for lack of a better
term, and with appreciation of how troubling this euphemism is, “collateral
damage guilt.” One Marine colonel, Col. Bob Durkin, commanded a battalion
just south of Baghdad during Operation Iraqi Freedom II. 33 He told me how
emotionally devastated his marines became when Iraqi children were injured
or killed after cars ran the trigger lines at vehicle checkpoints. If the injuries
or deaths were of adult men whom they suspected were suicide bombers or
women who might be concealing explosives under their burkas, his marines
would “generally fluff it off and justify it to themselves, rightly or wrongly.”
They would reason counterfactually: “Even if I couldn’t find out, it could been
this or coulda been that. But when children were involved, there was a dramatic
psychological difference.” In the case of a badly hurt child, “they would go out
of their way to try calling in medevac aircraft to get the kid out to the hospital,”
It was just different,” he said, “when kids were involved.” He continued, “Many
times, at the trigger lines, they waited and waited before they fired, putting
themselves at risk with cars skidding to a halt right in the middle of their posi-
tion. And I would yell at them for it, because had there been a bomb in there,
they would have been dead. But having said that, when a child was injured or a
child was killed, they would be emotionally upset by it.” They were more vulner-
able. They couldn’t shake what they had done.
We might speculate that these marines feel rationally appropriate guilt for
having incurred unjustified collateral harm. Perhaps so, though in many check-
points at this time, soldiers took upon themselves considerable risk, perhaps
even the preponderance of risk,34 as they allowed cars to wind their way through
warning signs, concertina wire, and cement barriers before inspections.35 They
were not unexposed to concealed bombs going off. In this particular case, the

33
I interviewed Bob Durkin as part of a group with other senior officers in Annapolis on
October 21, 2005.
34
For the notion of morally legitimate collateral damage requiring not just foreseen and unin-
tended harm but also a positive intention to minimize harm, see Michael Walzer, Just and Unjust
Wars, New York: Basic Books, 2000, 151–159. For a more recent discussion of troops taking on the
preponderance of harm, and hence transferring the risk from civilians to themselves, see Avishai
Margalit and Michael Walzer, “Israel: Civilians and Combatants,” New York Review of Books, May 14,
2009, http://www.nybooks.com/articles/archives/2009/may/14/israel-civilians-combatants/.
35
From conversations with Col. Tony Pfaff, who served in Iraq during this period. For a report-
er’s perspective, see http://www.washingtonpost.com/wp-dyn/articles/A28924–2005Mar12.
html. For a gripping description of a soldier’s view of the risk at a checkpoint, see David Grossman’s
To the End of the Land, New York: Knopf, 2010, 442. A young soldier tells his father about his role
at a checkpoint and the consequence of being unable to detect new kinds of concealed explosives:
“But, Dad, that’s my job! I stand there precisely so they’ll blow themselves up on me and not in
Tel Aviv.”
194 on emotions

troops wanted to take on more risk than the rules of engagement, or at least
how they were interpreted and enforced, seemed to allow.
No doubt, it is easier, psychologically, for troops to accept the transfer of
risk when children are the civilians. The Marines are vulnerable precisely
because the children are vulnerable. They can easily psychologically regress
to become those children, for after all, many are themselves boy-warriors,
not much older than those they killed or injured. But there are also moral rea-
sons at work, here, for accepting the transfer of risk. In the face of children
at checkpoints, the marines probably rightly see themselves as required to
be more police than warfighter. They need to be the good cop, who restrains
deadly force, develops a situation without risking innocent lives, and acts
as overall social healer and protector in a scene of violence.36 To fail to do
so in the eyes of a child desperate for moral order and a way out of war’s
chaos is morally devastating. It may be children who reinforce the meaning
of an “innocent” in a war zone and make psychologically palatable the moral
requirement of restraint. But the moral requirement is there in other, less
palatable cases, as well.
In this regard, consider a more recent case in Afghanistan where troops were
under more restrictive rules of engagement than those in place early in the war
in Iraq (when counterinsurgency had not been widely operationalized). It is a
case where troops minimized the risk to civilians, transferring the preponder-
ance of risk to themselves.37 But these troops took on not only mortal risks.
They took on moral risks not usually noticed and the moral injuries that come
with those risks.
In late December 2009 an incident in the Helmand Province made vivid the
point.38 In a command center, Marine Lt. James Wendy watched live video feed
of two men digging a hole for what seemed unmistakably a roadside bomb.
Other intelligence pointed in the same direction, including a radio intercept
from the suspected insurgents who talked about planting a bomb, and a sight-
ing of a swimmer crossing a canal carrying the sort of wires routinely used to
detonate a device. Near the hole, men were spotted with a yellow jug, an icon
in these parts for a roadside bomb that contains a lethal mix of fertilizer and
diesel fuel. Wendy’s unit called for air support, but the guns never opened up
because the insurgents were too close to a compound of houses. The bomb went

36
For insight here, see John Kleining, The Ethics of Policing, Cambridge, UK: Cambridge
University Press, 2005. For a discussion of restraint in the use of lethal force in military policing,
see Tony Pfaff, “Military Ethics in Complex Contingencies,” in The Future of the Army Profession,
ed. L. Matthews and D. Snider, Boston: McGraw Hill, 2005.
37
Again, echoing Walzer’s language.
38
For an account of the incident, see “Rules of Engagement are a Dilemma for U.S. Troops,”
Tom Bowman and Renee Montagne, National Public Radio, December 11, 2009, http://www.npr.
org/templates/story/story.php?storyId=121330893.
 G u i l t i n Wa r 195

off and the unit took significant losses. Some of the Marines stormed out of the
command center in rage.
After the incident, Wendy was philosophical about his own rage: “If I had
known how frustrating it’d be and was able to better prepare myself for that
mentally, I think that maybe I would’ve been better off.”39 Stoic teachings
come to mind here: Anticipate the hardship and the evils will have less force.40
Soldiers should not just be desensitized to face bloody, physical gore in live tis-
sue exercises; they should practice loss.
But, of course, Wendy is talking about a special kind of loss.
Counterinsurgency requires the loss that comes with restraint, but for many
troops that restraint is experienced as nothing short of the betrayal of battle
buddies. The Soldier’s Creed, drilled in at all levels of the Army (and implicit
also, it would seem, in the Marine ethos of Semper Fidelis), spells out the gen-
eral expectations: Soldiers “watch each other’s back.” A soldier “will never
leave a fallen comrade” behind. Soldiers do not leave each other exposed. They
deploy “to destroy and engage an enemy.”41 But in wars in which civilians are
the battlefield—wars fought to win civilian trust and loyalty and to woo locals
away from insurgents, the idea that war is primarily about closing in and fight-
ing a war of attrition is out of step. But so, too is the expectation that you can
always cover with guns or air support a buddy who has fallen in harm’s way.
Soldiers who aren’t properly trained for that loss, for that shift in protection
and danger, are themselves at risk.
Rethinking the warrior ethos requires coming to grips with the moral and
emotional expectations that are part of a new warfare that is population-cen-
tric (where the battlefield just is the people)42 and requires added transfer of
risk from civilians to soldiers. But that, in turn, requires a new kind of armor,
a new training, whereby soldiers do not view themselves as betraying each
other when they put themselves at risk in order to protect civilians. I don’t
know just how that resilience is built and if it is different from learning how
to accept loss to enemy fire when civilians are not such a clear part of the
equation. Troops these days often grumble that fighting with restrictive rules
of engagement amounts to “fighting with one hand tied behind one’s back.”
The phrase suggests that the fighting is unfair, that soldiers can no longer
fight “by big boy rules,” as I have heard it put. But I suspect what is unfair are
not the restraints, but the lack of training in internalizing them as part of a

39
Ibid.
40
On prerehearsal, see Cicero’s redaction, Tusculan Disputations, 3.58 in Graver’s translation
and my discussion, Stoic Warriors, 145. For similar views from Seneca, see Stoic Warriors, 117.
41
From the “Army Leadership” manual, defi ning the core concepts of leadership doctrine, FM
6–22 (FM 22–100) http://usacac.army.mil/CAC2/repository/materials/fm6–22.pdf
42
For the notion of “war amongst the people,” see British Gen. Rupert Smith, The Utility of
Force, New York: Vintage, 2005.
196 on emotions

revised warrior code reflective of the requirements of a revised just conduct

when civilians are the battlefield.

Staying Whole
I have been arguing that guilt in war is redemptive; it is a way of reconnecting
with one’s humanity, a way of trying to say whole when war is a supreme exer-
cise in numbing.
Michel de Montaigne famously warns about the high price of trying to stay
whole. His own advice is often of moral insulation and compartmentalization
of roles: “The Mayor and Montaigne have always been separate,” he boasts. “I
know some who transubstantiate and metamorphose themselves into as many
new beings and forms as the dignities which they assume: they are prelates
down to their guts and livers and uphold their offices on their lavatory-seat. I
cannot make them see the difference between hats doffed to them and those
doffed to their commissions, their retinue or their mule.”43
The message reinforces the idea of the externality of the uniform. It is
always the uniform one salutes, and not the person in it. Erving Goff man, the
great American sociologist of roles, wrote powerfully about the “language and
mask” characteristic of ritualized social interaction.44 Soldiers, more than oth-
ers, inhabit a world of public ritual and decorum. To walk on a base, stateside,
is to take note of that world—the spit-polish shoes, the creased uniforms, the
stolid demeanor and gaze, the stiff salute, the parades, the “yes, ma’am-ing”
and “yes, sir-ing.” Even in death, there are the official photos that become the
public faces of the fallen.
But soldiering, and especially wartime soldiering, is never about just a uni-
form or an outward face. It is about the deep internal changes that go with
putting on the uniform and deploying.
Yet what moved me as I listened to soldiers, especially those recently
returned, is how desperately they wanted to feel their old civilian selves. Or at
least, they wanted to feel more porous boundaries between being a soldier and
a civilian—whether as a parent, a child, or a colleague in the workforce; keeping
their different selves fully separate was just too numbing. “We’re taught to suck
it up and truck on,” a former Army major told me. “I’m tired of being stoic.”
In this regard, even the awful guilt and shame soldiers felt for doing well
what war requires was a way of being alive to a fuller self. It was therapeutic,
excruciatingly painful to be sure, but also a way of not being callous to war’s
devastation and one’s participation in it. In this sense, guilt is not just a feeling

43
Michel de Montaigne, The Complete Essays, Screech ed., London: Penguin, 1993), 1144.
44
See, for example, Erving Goff man, The Presentation of Self in Everyday Life, New York: Anchor
Books, Doubleday, 1959.
 G u i l t i n Wa r 197

to be removed in a clinician’s office, excised because it represents a pathological

or grandiose sense of agency in the face of luck or accident or conflict. It needs
to be relieved and lessened but also understood for what it is.
Here, I turn to a young Army interrogator. He offered a striking analogy for
what it felt like to be the interrogator he once was: Entering the interrogation
cell was a bit like going into a mass with Gregorian chants sung in Latin: It
takes place, he said, “in a different universe.” “War, too, takes place in a differ-
ent time and space.” In essence, he was describing dissociation, compartmen-
talization, stoic numbing. Yet for this young interrogator dissociation was not
ultimately a viable solution: “I know I am the same person who was doing those
things. And that’s what tears at your soul.” To not feel that bite, that guilt for
his actions and omissions in war was to risk losing his humanity as a civilian
and as a soldier.
 10

Emotion and Personal Identity

Charles Nussbaum

Introduction
In his Preface to On the Genealogy of Morals (1887/1967), Nietzsche informs
us that it was his reading of a “clear, tidy, shrewd, and precocious little book”
on the origins of morality by his friend (and fellow suitor for the hand of the
endlessly fascinating Lou Andreas-Salomé) Dr. Paul Rée that first motivated
him to publish a piece on the same subject. Nietzsche’s response to the book
was antipodal: every proposition, every inference elicited a resolute, if not
an ill-humored or impatient, “No.” The problem was that poor Rée had been
infected by the “English” disease of moral theorizing based on altruism, by
which Nietzsche presumably means some noxious mixture of utilitarianism
and sentimentalism. While sensitive to the dangers such a comparison poses,
I own that I had a similarly antipodal, but far less patronizing, reaction to
another precocious book, a book also very English (though hardly ein Büchlein),
namely, Derek Parfit’s Reasons and Persons.
Having made this comparison, I would like to qualify it further. First, I have
no sympathies at all for Nietzsche’s philosophical Anglophobia, happily admit-
ting a considerable philosophical Anglophilia. Second, and philosophically
more important, it is Parfit’s account of personal identity, not his moral theory,
that struck me as so very “English” in Nietzsche’s pejorative sense, and that
provoked the antipodal reaction. Convinced that “English” associationist psy-
chology degrades human mental activity to “something purely passive, auto-
matic, reflexive, molecular, and thoroughly stupid” (1887/1967, 24), Nietzsche
judges it unworthy of respect and every bit as pernicious and faithless to the
vocation of humanity as altruistic morality.
Parfit’s metaphysics of the person might be characterized as Humean asso-
ciationism on steroids. Although he judges Hume’s view of personal identity
ultimately “inadequate,” Parfit immediately promises to “defend a view that, in
the relevant respects, follows Hume” (1984, 139). Part of Hume’s inadequacy

198
 E m o t i o n a n d Pe r s o n a l I d e n t i t y 199

seems to consist in the fact that radical as his defl ationary account of personal
identity is, it is not radical enough. A skepticism that plunged the young Hume
into black depths of alienation and despair turns out to be too weak a potion
to bring about the bracing effects Parfit derives from his own, far more revi-
sionary, account of personhood. Parfit’s moral theory, on the other hand, while
definitely consequentialist, is no simple utilitarian or sentimentalist doctrine
(as indeed Hume’s moral theory is not), but one that might qualify as eudai-
monistic by the standards of ancient Greece. If Parfit’s deflationary account
of personal identity, but not his ethics, suffers from the “English” disease of
associationistic psychology, what, then, is the cure? The prescription now, as in
the eighteenth century, I shall argue, calls for a strong dose of Teutonic tran-
scendentalism, but fi ltered through American pragmatism, contemporary neu-
roscience, and evolutionary theory.

Persons, Selves, and the Subneocortical Brain

“The ‘I think’ which Kant said must be able to accompany all my objects
[sic., Vorstellungen],” William James asserts in his Essays in Radical Empiricism
(1971b, 164), “is the ‘I breathe’ which actually does accompany them.” Even a
reader familiar with James’s theory of the emotions1 might be puzzled by this
until she realizes that breathing is a semivoluntary, partly autonomic motor
function that enlists two of the most primitive areas of the brain, the medulla
and the pons, by way of the vagus nerve. Certain nuclei in the brainstem, includ-
ing the cuneiform (or cuneate) nucleus in the medulla, the pontis oralis in the
the pons, and other aminergic and acetylcholine-sensitive brainstem nuclei,
along with the hypothalamus and certain old mammalian cortical structures
(the insular cortex and the medial parietal cortices) support what Damasio
(1999), a neo-Jamesian in emotion theory, has termed the “proto-self.” These
structures nonconsciously monitor and modulate the states of the body and
are centrally implicated in the generation of emotions that prime the human
organism to respond appropriately to threats to homeostasis. The preservation

1
It is not entirely clear what James’s theory of the emotions is. At times, James seems to
identify emotions with the feelings caused by bodily changes, while at other times he identifies
them with the bodily changes themselves. For an example of the fi rst tendency, see Principles of
Psychology (1890, Vol. 2, 449, emphasis in original): “My theory is that the bodily changes follow
directly the perception of the exciting fact, and that our feeling of the same changes as they occur is
the emotion.” For an example of the second, see ibid., 452 (emphasis mine): “The more closely I
scrutinize my states, the more persuaded I become that whatever moods, affections, and passions
I have are in very truth constituted by, and made up of, those bodily changes which we ordinarily
call their expression or consequence. . . . ” What is clear is that bodily changes are, for James, not
mere effects of emotion feelings.
200 on emotions

of homeostasis and bodily integrity, the maintenance of a fundamental

self–not-self, inside–outside distinction, in turn, is fundamental to all life. The
phylogenetically more sophisticated, but only episodically conscious, “core self”
enlists, in addition to proto-self structures, the cingular and somatosensory
cortices; and the temporally extended conscious and self-conscious “autobio-
graphical self” recruits neocortical structures, including the association corti-
ces. Both the core self and the autobiographical self, however, are functionally
dependent on the proto-self. The person can (or should) be said to comprise all
three “selves.”2
Thought experimentation remains an indispensable tool in philosophy, but
thought experimentation uninformed by empirical findings can yield skewed
results. I shall argue that much of the thought experimentation on which Parfit
relies to lend plausibility to his conclusions concerning personal identity is
done without the benefit of relevant neuroscience. As a result, he misinter-
prets the significance of data, especially data concerning so-called split-brain
phenomena, and asks us to accept as possible, scenarios that are “deeply” or
biologically impossible3 rather than being, as he pretends, only “technically”
impossible. Like Hume, Parfit relies heavily on memory and its causal prov-
enance in his account of personal identity.4 Memory, however, at least as it is
understood by Parfit, is declarative and a function of the autobiographical or
extended self.5
A thought experiment conceived by Bernard Williams and discussed at some
length by Parfit provides the most direct entry into our central issues. Imagine,
says Williams (1973, 52), that someone:

2
Damasio (1999, 22) comes close to saying this himself: “The deep roots for the self, including
the elaborate self which encompasses identity and personhood, are to be found in the ensemble of
brain devices which continuously and nonconsciously maintain the body state within the narrow
range and relative stability for survival.” Zahavi’s cognate view (2003, 59) approximates mine:
“Thus the experiential self could be characterized as the core self.” (2003, 62): “the notion of the
self introduced by the narrative model is not only far more complex than, but also logically and
ontologically dependent upon, the experiential self. . . . When we are dealing with the experien-
tial self, we should stick to the term ‘self’, since we are exactly dealing with a primitive form of
self-givenness or self-referentiality. But when we are dealing with the narrative model, it would
be better to speak not of the self, but of the person as the narrative construction. After all, what
is being addressed by this model is exactly the nature of my personality or personal character: a
personality that evolves through time, and which is shaped by the values I endorse, and by my
moral and intellectual convictions and decisions.”
3
I say “biologically impossible” rather than “nomologically impossible” because the identity
and status of unrestricted biological laws are controversial.
4
Cf. Hume (1739/1978, 262): “memory does not so much produce as discover personal identity
by showing us the relation of cause and effect among our different perceptions.”
5
Parfit (1984, 517 n. 26): “there would be certain kinds of non-distinctive [psychological] con-
tinuity, such as how to walk and run. The Psychological Criterion [for personal identity] should
not appeal to these kinds of psychological continuity.”
 E m o t i o n a n d Pe r s o n a l I d e n t i t y 201

In whose power I am tells me that I am going to be tortured tomorrow.

I am frightened, and look forward to tomorrow in great apprehension.
He adds that when the time comes, I shall not remember being told
that this was going to happen to me. . . . He then adds that . . . when the
moment of torture comes, I shall not remember any of the things I
am now in a position to remember. . . . He now further adds that at the
moment of torture I shall not only not remember the things that I am
now in a position to remember, but will have a different set of impres-
sions of my past, quite different from the memories I now have.

Understandably, none of these assurances does much to cheer Williams up. Nor
does he see why he:

Should be put into any better frame of mind by the person in charge
adding lastly that the impressions of my past with which I shall be
equipped on the eve of torture will exactly fit the past of another
person now living, and that I shall acquire these impressions by (for
instance) information now in his brain being copied into mine. Fear,
surely, would still be the proper reaction: not because one did not know
what was going to happen, but because in one vital aspect at least one
did know what was going to happen—torture, which one can indeed
expect to happen to oneself, and to be preceded by certain mental
derangements as well.

Williams’s intuitions regarding this thought experiment lead him (tentatively)

to conclude that “one’s fears can extend to future pain whatever psychological
changes precede it,” thereby implying that personal identity depends not on
psychological continuity but on physical continuity (1973, 63; Parfit 1984, 230).
Despite the radical changes in his autobiographical memories, Williams fears
that it is he, not the person with the matching memories, who will experience
the future pain, will hate it, and will want it to stop. The thought experiment
does allow for what Parfit terms a “psychological spectrum” of alterations, with
minimal alteration at the near end and total change of memory and personal-
ity at the far end, at which point the subject comes to believe something totally
deranged, say that he is Greta Garbo. Even in this case, thinks Williams, the
subject would have reason to be fearful. Here, I believe, Williams is entirely cor-
rect. But his attempt to base personal identity on physical continuity remains
vulnerable.
Invoking the complementary notion of a physical spectrum, Parfit has little
trouble calling into doubt the adequacy of physical continuity to secure per-
sonal identity. At the near end of the physical spectrum, your body and brain
remain intact. At the far end, you undergo a complete replication process.
202 on emotions

Your present body is destroyed and your memories stored. Your body is then
replicated molecule for molecule and your stored memories are transferred to
the newly constituted brain. Located in between are various stages of physical
replacement. No one doubts, says Parfit, that the recipient of a heart or liver
transplant remains the same person. Under conditions of severe infection, a
patient may undergo a complete blood “exchange.” When the procedure is com-
pleted, the patient remains the same person. So at what point in this physi-
cal spectrum does your identity change? Just how much of your body must
be replaced before you become another person? There is, Parfit contends, no
principled answer. At “central” (or intermediate) locations on the physical spec-
trum, there is no bright line that marks the point at which one identity gives
way to another. Intuition generally favors preservation of personal identity
through a complete replacement of physical components if the replacement is
gradual. Even in the extreme case of instantaneous replication and simultane-
ous destruction of your old body, where intuition might favor change in per-
sonal identity, your duplicate will think he or she is you and will continue your
life and projects. Perhaps most importantly, there will be no way for anyone,
including you, to determine that personal identity has not been maintained.
Parfit endorses no crude verificationism: he does not brand as senseless the
claim that there is a persisting nonphysical entity like a Cartesian ego that
is the seat of personal identity (1984, 224). But he does insist that there is no
good reason to think that there is any such thing. Consequently, we are free
across a wide range of cases to say either that personal identity is maintained
or that it is not maintained.6 What matters, says Parfit, is not personal iden-
tity but something he calls relation R: strong psychological connectedness and
continuity caused in the right way. Strong psychological connectedness con-
sists in a critical (but unspecified) number of robust short-term memories, that
number being “at least half of those short-term memories that hold, over every
day, in the lives of every actual person” (1984, 206), and continuity consists
in the overlapping chains of strong connectedness that constitute long-term
memory. Note that “caused in the right way” does not necessarily mean “caused
in the normal way,” that is, by the physical continuity of the body. In the case of
replication, the cause of memory is not normal, but it is reliable. Yet R does not
even require reliability: any cause will do. Since it may be denied that mental
states abnormally or unreliably caused are memories at all, they may have to be
called “quasi-memories.” But if their status as quasi-memories remains unde-
tectable, that will not matter either, to you or to anyone else.

6
Cf. Hume (1739/1978, 262): “all the nice and subtile questions concerning personal identity
can never possibly be decided, and are to be regarded as grammatical rather than philosophical
difficulties.” Parfit (1984, 277): “To illustrate these claims, I have repeated Hume’s comparison.
Persons are like nations, clubs, or political parties.”
 E m o t i o n a n d Pe r s o n a l I d e n t i t y 203

In the case of a “branching line,” where, after replication, your body is not
immediately destroyed and you are able to carry on a brief conversation with
your newly constituted replica, Parfit admits that we are not free to deny that
the original, and not the duplicate, is you: “Since I can talk to my Replica, it
seems clear that he is not me. Though he is exactly like me, he is one person,
and I am another. When I pinch myself, he feels nothing. And when I am dead
he will live for another forty years.” Yet, rather remarkably, Parfit denies the
“natural” assumption: “If we believe that my Replica is not me, it is natural
to assume that my prospect, on the Branch Line, is almost as bad as ordinary
death. I shall deny this assumption. As I shall argue later, being destroyed and
Replicated is almost as good as ordinary survival” (201). I, on the contrary,
shall argue that it is much worse than ordinary survival. It will not be Parfit,
but his replica, who is enjoying or suffering this future.
Parfit counters any attempt to formulate a “hybrid” view of personal iden-
tity that depends jointly on physical and psychological continuity by a radical
combination of the physical and psychological spectra. Imagine, he asks, that
his body changes gradually so as to become a replica of Greta Garbo at thirty
and his memory set is slowly replaced by the set she had at that age (1984, 237).
Here, he also denies that personal identity is always determinate: at the far end
point of the combined spectrum personal identity, he allows, is not preserved.
But, unless we countenance a Cartesian ego or some “deep further fact” beyond
physical and psychological continuity, we are also at a complete loss to identify
any sharp dividing line that marks the transition from being Parfit to being
Garbo’s replica. Any choice among intermediate points on the combined spec-
trum will be arbitrary. What matters, once again, is relation R: as psychological
continuity is increasingly attenuated, Parfit fades from existence. I disagree.
I submit that if physical replacement is gradual, without total destruction and
subsequent replication, and if subneocortical brain function is maintained, the
subjective experience had by Garbo’s replica will be Parfit’s future experience.
And if torture lies in the replica’s future, Parfit should be worried.
In an attempt to seal his case, Parfit takes up the phenomenon of “divided
minds.” It is important to note that he is not always careful to distinguish
between divided hemispheres and divided brains.7 While the former phenom-
enon has some basis in fact, the latter is sheer fantasy, indeed, as I shall argue,
“deeply impossible” fantasy. Parfit describes the bifurcated conscious experi-
ences had by certain patients who have undergone commissurotomy:

One of these people is shown a wide screen, whose left half is red and
right half is blue. On each half in a darker shade are the words, “How
many colours can you see?” With both hands the person writes, “Only

7
Robinson (1988, 324) makes a similar charge against Parfit.
204 on emotions

one”. The words are now changed to read, “Which is the only colour
that you can see?” With one of his hands the person writes “Red”, with
the other he writes “Blue”. (1984, 245)

These phenomena motivate a curious thought experiment. While the capaci-

ties of the hemispheres in most individuals are specialized to some degree for
different cognitive tasks, it is apparently the case, or so Parfit claims, that in
a minority of people, there is “no difference between the abilities of the two
hemispheres” (1984, 246). Now imagine a student with functionally identical
hemispheres taking a physics exam. This student possesses the unusual abil-
ity to block and then unblock at will communication across the corpus callo-
sum. With only fifteen minutes remaining to complete the exam, she realizes
that there are two ways of tackling the final problem. She divides her conscious
mind, putting one half to work using one strategy, the second half using the
other. During the process, the student can take the standpoint supported by
her left hemisphere while she consciously moves her right hand and merely
observes her left hand scribbling away, or she can take the standpoint sup-
ported by her right hemisphere while she consciously moves her left hand and
merely observes her right hand scribbling away. At the end of the exam, she
promptly reunites her mind and remembers having worked both strategies.
What this thought experiment shows, Parfit maintains, is that during the
last fifteen minutes of the exam there are two streams of consciousness just as
there apparently are two streams of consciousness in the case of patients with
divided hemispheres. But do we want to say that there are two persons tak-
ing the exam? Parfit thinks not, and I agree, though not for his reasons. If we
ascribe these different streams of consciousness to separate subjects of experi-
ence, he argues, we must then grant that there are subjects that are not per-
sons. Yet after the exam, the subject of experience is the person. To admit this
is to admit that in the lives of persons, there could be two kinds of subjects of
experiences, those that are and those that are not persons. This, in Parfit’s
view, robs the notion of a subject of experience of any plausibility and leads us
to doubt that there are subjects of experience (1984, 250). There are just bodies
and continuous streams of consciousness. If it were possible to divide a stream
of consciousness permanently, say by some manner of split brain transplant
into different bodies such that both of two successor streams bore relation R
to the earlier single stream, relation R would be preserved, but personal iden-
tity would not be, since identity is a one–one relation. From here, it is a short
step to the “Reductionist” view that the language of physical and psychological
continuity exhaustively describes what facts there are and that what matters
(or what ought to matter) is not personal identity, but relation R. When R is a
one–one relation, personal identity reduces to the holding of R; when it is not,
personal identity is not preserved. There is nothing more to be said.
 E m o t i o n a n d Pe r s o n a l I d e n t i t y 205

Since Damasio enables us to distinguish in a principled way between selves

and persons as functional entities, there is something else we can and perhaps
should say. Parfit holds that if we reject reductionism, “we must claim that in
these [divided mind] cases, though there is only a single person, there are two
subjects of experiences. We must therefore claim that there can be, in a person’s
life, subjects of experience that are not persons. It is hard to believe there are
such things” (1984, 276). I counter that it is not hard to believe at all. We may
say that during the physics exam, one person briefly shares (alternately) two
autobiographical selves, two streams of consciousness that remain functionally
dependent on one underlying proto-self and core self. An autobiographical self
is not a person. This makes sense of the real-life divided hemisphere phenom-
ena and perhaps even the ultimate unification of experience at the conclusion
of the fantastic exam scenario.
That Parfit fails to distinguish carefully enough between the cerebral
hemispheres and the sub-neocortical structures that support the core and
proto-selves is clear from the subsequent course of his argument. In one place
(254) he says, “there are many people who have survived when a stroke or injury
puts out of action one of their hemispheres.” True enough. People have even
survived without life support (other than feeding) with both hemispheres “out
of action.” But no one survives (unassisted by life-support machinery) when a
stroke or injury puts one half of the entire brainstem out of action. Although
nuclei in the brainstem are bilateral, important ascending and descending
neural pathways (the medial lemniscus, the fasciculus cuneatus, and the cor-
ticospinal or pyramidal tracts) decussate or cross over in the medulla. Indeed,
all descending cortical projections and all ascending somatic afferent projec-
tions traverse the brainstem at some point (Kandel et al. 1991, 691). Sagittal
division of the brainstem, therefore, would severely disrupt important neural
signaling. Moving up from the brainstem to the diencephalon, it is doubtful
that a sagittally divided thalamus or hypothalamus could function normally,
despite bilateralism of the nuclei in these structures. Moreover, division of the
diencephalon would require division of the pituitary gland, which is directly
controlled by the hypothalamus and is not bilateral. The pituitary is an impor-
tant component of the endocrine system that secretes various hormones and
regulates water concentrations in body fluids. Nevertheless, on the same page
Parfit goes on to claim that “I could survive with half my brain. . . . I would sur-
vive if half my brain was successfully transplanted into my [brain-dead] twin’s
body and the other half destroyed.” Notice the shift here from “hemisphere” to
“brain.” I make no accusation of bait and switch. Parfit grants that he is “assum-
ing that we could divide, not just the upper hemispheres, but also the lower
brain.” He also admits that “it seems likely that it would never be possible to
divide the lower brain, in a way that did not impair its functioning” (1984, 255).
But this impossibility, he says, is merely “technical,” not “deep” or biological.
206 on emotions

I believe this to be quite wrong. First, the neurological evidence strongly

suggests that lower brain function cannot be bifurcated in the way Parfit
alleges hemispheric function can be and that this is a principled impossibility,
not a merely technical one.8 Second, if lower brain function is maintained at all,
it can support only one proto-self and one core self in one body, and it is these
functions, I contend, that lend credence to Williams’s intuitions concerning
his future torture. Compare, in light of Parfit’s divided minds, Damasio’s dis-
cussion of the different, but not unrelated, phenomenon of so-called multiple
personality disorder (1999, 355 n. 14):

In such patients, to a certain extent, it is reasonable to talk about more

than one “autobiographical memory,” and more than one construction
of identity and response manner, connected to different life histories
and anticipated futures. It is apparent, however, that in spite of being
able to display more than one autobiographical self, such patients con-
tinue to have only one mechanism of core consciousness and only one
core self. Each of the autobiographical selves must use the same central
resource . . . generation of the core self is closely related to the proto-self
which, in turn, is based on the representations of one singular body in its
singular brain. Given a single set of representations for one body state,
it would require a major pathological distortion to generate more than
one proto-self and more than one core self. Presumably the distortion
would not be compatible with life. On the other hand, the generation of
the autobiographical self occurs at a higher anatomical and functional
level. . . . (emphasis mine)

When using the notion of the physical spectrum to attack Williams’s proposal
that personal identity depends on physical, not psychological, continuity, Parfit
adroitly raises the following point:

Perhaps there is some critical part of my brain which is such that, if

the surgeons do not replace this part, the resulting person will be me,
but if they do, it will be someone else. (1984, 235)

As might be expected, Parfit then runs his sorites argument against this pro-
posal, asking what percentage of this critical part of the brain must be replaced
before personal identity changes. Damasio’s model provides a principled answer
to this question. Of the three options Parfit offers, Cartesianism, the “deep
further fact” option, and reductionism, I claim that the second is best. We

8
Robinson (1988, 325–327) presents compelling empirical evidence that hemispheric func-
tion cannot be bifurcated in the way Parfit describes.
 E m o t i o n a n d Pe r s o n a l I d e n t i t y 207

must, however, specify that this deep further fact is a functional fact. Physical
components may be replaced as required. But if proto-self and core-self brain
functions are permanently shut down, if the brain structures supporting
them die and deteriorate beyond resuscitation, the person who will suffer in
Williams’s thought experiment ceases to exist: just then, not before and not
afterward.9 Contrary to Parfit, this is a sharp line that marks the boundary of
the relevant change in identity. The preservation of core-self function, a “deep
further fact” beyond physical or psychological continuity as Parfit defines them,
allows us to vindicate Williams’s intuition, while both refuting Parfit’s sorites
arguments and resisting reductionism. Recall the particulars of the case: All of
the psychological alterations promised by the torturer concern autobiographi-
cal memory alone. Nothing is said about those aspects of mentality realized
by lower brain functions. In fact, these aspects seem to be ruled out by Parfit’s
apparent stipulation that memory implicated in psychological connectedness
and continuity be declarative and not procedural. It is eminently reasonable to
surmise that even if all his autobiographical memories are altered, it will still
be Williams who suffers the pain and emotional distress of the future torture.
What counts in the Williams scenario is not relation R or physical continuity
but the functional continuity of the subneocortical brain.10 Even if Williams’s
cerebral cortex were destroyed and replaced with a functionally equivalent
structure provided with an entirely new set of autobiographical memories,
he should still worry about his future torture, for there is no reason to think
that his cortical replacement will serve to buffer his suffering. If, on the other
hand, the living half of Parfit’s cerebral cortex were, as he proposes, to be trans-
planted into his twin, but a twin who was not brain dead and who possessed a
functioning subneocortex, and that twin were to be promised torture after the
operation, then his twin should expect to undergo suffering. The importation

9
Continuity of proto- and core-self functions are, then, necessary, if not sufficient, for per-
sonal identity.
10
Suppose the molecules constituting Williams’s brain were to be replaced one by one, with
the extracted molecules preserved and then reconstituted into a new living brain. Supervenience
would seem to require that if these molecules are recombined in just the right way, then Williams’s
core-self functions would be restored and personal identity preserved. Yet I have maintained that
Williams’s brain is the one whose components have undergone gradual replacement. If so, his
brain cannot be the reconstituted one. Notice that this scenario would amount to a return from
the dead, the survival by the core self of the destruction and subsequent reconstitution of its sup-
porting neural structures. Th is is counterintuitive, but it does not amount to a reductio. Robinson
(1988, 319–320) proposes a similar thought experiment and concludes that personal identity
would be preserved after reconstitution. I am not convinced. It may be that functional continuity
of the subneocortical brain imposes causal-historical conditions on core-self identity, including
procedural psychological continuity, which would be violated, much as cladist interpretations of
species identity impose causal-historical identity conditions that are independent of superve-
nience on microstructure.
208 on emotions

of Parfit’s autobiographical memory also will not buffer that suffering: after
transplantation, Parfit’s autobiographical memory will feel like it belongs to his
twin, who will believe himself to be Parfit.
Parfit’s skepticism concerning Cartesianism is salutary, but only to a point.
There was a time, around the middle of the last century, when it was fashion-
able to endorse Georg Christoph Lichtenberg’s criticism of the Cartesian cogito.
Descartes, Lichtenberg maintained, had no justification for asserting “I think,”
but only “It thinks” or “There is thinking.”11 Although Parfit, predictably
enough, endorses Lichtenberg, I have never been convinced. Lichtenberg was
no doubt correct that Descartes had no justification to assume the ontology or
the identity conditions for a substantial subject of experience. Still, I always
have believed that Descartes had the phenomenology of thinking right, if not
rightly construed: barring severe neurological disorder, our thoughts do come
to us as our own. Parfit’s physics exam thought experiment is designed to serve
as a counterexample to the Kantian transcendental principle that the unity of
experience is a coherence condition on any discursive understanding, that the
“‘I think’ must be able to accompany all my representations.” Whatever the
merits of its case against Kant’s transcendental argument for the unity of expe-
rience, the physics exam fails as a counterexample to an empirical, naturalistic
account of the biological bases for our evolved, terrestrial version of experi-
ential unity and thought ownership, for that unity involves functions that
lie much deeper than the unity of the autobiographical self. Indeed, these are
functions on which the unity of the autobiographical self depends. The outlines
of such a naturalistic account are just what James and Damasio give us. The
proto-self “is based on the representations of one singular body in its singular
brain,” Damasio has declared. I take this to mean that the proto-self is based
on representations, in a particular brain, of the particular body in which that
brain is ensconced, for the body and brain are material particulars, numeri-
cally identical with themselves. But they are functional particulars that allow
replacement of parts, so long as function is preserved: the proto-self is this mate-
rial functional entity and no other.
The role of the proto-self is to monitor and regulate the state of the body,
we are told. Though the proto-self is not conscious, the body is its intentional
object, or the object of the neural representations on which it is “based.” Not
so the conscious core self. Its intentional objects are out in the world and there
is good reason to believe that they include and must include objects of emo-
tion. The core self, recall, does not require the specialized sensory cortices,
not even the early sensory areas, only the somatosensory cortex, the cortical

11
Lichtenberg (1971, 412): “We are acquainted only with the existence of our sensations, rep-
resentations and thoughts. There is thinking, one should say, as one says, there is lightning [es
blitzt]. To say cogito is already too much, as soon as one translates it by I think.”
 E m o t i o n a n d Pe r s o n a l I d e n t i t y 209

area that registers states of the body. Emotions, on the neo-Jamesian view,
are mental states that represent the environment by way of valent or evalua-
tive responses of the body. Although the “nominal content” of these emotions
concerns the body, their “real content” concerns what is out in the world (Prinz
2004, 68–69). “Emotions and core consciousness,” says Damasio (1999, 100),
“tend to go together, in the literal sense, by being present together or absent
together. . . . Both emotions and core consciousness require, in part, the same
neural substrates.”
Now it could be said that emotion and core consciousness are independent
functions and that their copresence and absence are artifacts of this partial
sharing of neural substrates.12 I think Damasio’s account supports a stronger
line. For this would be an accidental functional correlation, not be the “close
functional relationship” (1999, 100) between emotional processing and core
consciousness that he highlights. Since the conscious core self is built on the
unconscious proto-self and augments its bodily homeostasis-preserving func-
tions with emotional appraisals of worldly circumstances, the core self’s repre-
sentational mode is inherently emotional. The core self, that is to say, represents
the world in terms of “core relational themes” (Lazarus 1991, 121–123): How
does this exigent situation bear on this particular organism; how does it bear
on me? “Some degree of continuous emoting is virtually inseparable from the
conscious state,” we are told (Damasio 1999, 100). Yet “there is no such close
functional relationship between emotional processing and extended conscious-
ness,” the mode of consciousness that belongs to the autobiographical self
(100–101). From this it follows that “the continuous emoting” that is “virtually
inseparable from the conscious state” requires, indeed perhaps even derives
from, core consciousness. As a result of the core self’s emotional appraisals of
how things bear on me, the “I think” is implicit in core consciousness, but it is
already there. And something similar can be said about the extended conscious-
ness of the autobiographical self, where the “I think” has, in humans, become
explicit: because of its functional dependence on the proto- and core selves, its
thoughts are tinged with a certain “‘warmth’ of bodily existence,” as James says
(1971a, 103), a distinctive emotional glow that marks them as mine. “This cen-
tral part of the self is felt,” he asserts. “It may be all the Transcendentalists say
it is, and all the Empiricists say it is in the bargain, but it is at any rate no mere
ens rationis, cognized only in an intellectual way, and no mere summation of
memories. . . . But when it is found, it is felt; just as the body is felt. . . . ” (1971a,
89, emphasis in original). Personal identity involves just the sort of “deep fur-
ther fact” Parfit dismisses. But it is a functional, biologically deep further fact,
the functional unity of the subneocortical brain and the body it regulates.

12
Th anks to John Deigh for this objection.
210 on emotions

Curing the Schizophrenic Cogito

Lichtenberg’s version of the cogito is curiously suggestive of the “thought inser-
tion” experienced by schizophrenics,13 the sense that thoughts of which one
is conscious are not one’s own. The early neurologist Manfred Bleuler hypoth-
esized that schizophrenics suffered from a split between the cognitive and the
affective or emotional side of the personality (Kandel et al. 1991, 855). Wilder
Penfield reported (1974, 242) that when he caused a conscious patient to move
his hand by applying an electrical stimulus directly to the primary motor cortex
of the contralateral hemisphere, the patient’s response was, “I didn’t do that,
you did.” When, by similar means, Penfield caused “the record of his stream of
consciousness to run again and so presented to him the replay of his past expe-
rience,” the patient was able to recall significant detail, but remained strangely
aloof. He seemed somehow to pass judgment on his thoughts, saying “things
seem familiar,” but not “I have been through this before.”
These considerations motivated Irwin Feinberg (1978) to hypothesize an eti-
ology for schizophrenia based on the notion of corollary discharge. The theory
of corollary discharge was first proposed by Helmholtz but was empirically dem-
onstrated in 1950 independently by Sperry and by von Holst and Mittelstaedt
by experimentation on fish and insects. When an animal engages in voluntary
movement, it is aware (or it nonconsciously registers) that it is moving, not
its environment, despite the sweep of stimulations across its receptors. Such
registration is a primitive, embodied mode of self-awareness and is explained
as follows. A corollary discharge or efference copy of the central motor com-
mand is sent to a comparator, which, in humans, is thought to be located in
the cerebellum. The comparator integrates the efference copy with a reaffer-
ence signal returning from peripheral receptors (Kandel et al. 1991, 627).
Homologues of the human cerebellum are found not only in nonhuman mam-
mals but also in all but the most primitive proto-chordates (Sarnat and Netsky
1981, 214). When efference copy matches reafferance signal, this information
is transmitted to the appropriate sensory areas and apparent environmental
motion is eliminated. In the case of vision, a voluntary turn to the right would
be expected to produce an apparent leftward motion of an animal’s visual field.
This does not happen, because efference copy and reafference signal cancel each
other like a photograph and its matching negative. As a result, the surround is
experienced as stationary and the motion is felt as self-generated. One might
say it acquires James’s “warmth” of bodily existence. An attempt to move eyes
that suffer from extra-ocular muscle palsy results in an apparent displacement

13
Feinberg and Guazzelli (1999, 197) make passing reference to Lichtenberg, a somewhat
obscure eighteenth-century German thinker. Th is is a rather startling occurrence in a contempo-
rary scientific research paper on schizophrenia.
 E m o t i o n a n d Pe r s o n a l I d e n t i t y 211

of the visual field. This is explained by the receipt of an efference copy of the
motor command without a reafference signal, since the attempt brings about
no ocular movement and no sequence of receptor stimulations (Feinberg and
Guazzelli 1999, 198).
Although evidence for the corollary discharge theory of schizophrenia is
not conclusive, the theory does have empirical support and has attracted advo-
cates in the psychiatric community and elsewhere (Frith 1992; Andreasen et
al. 1998; Campbell 1999; Feinberg and Guazzelli 1999; Cahill and Frith 2000;
Currie and Ravenscroft 2002; Blakemore and Frith 2003; van Hoof 2003; Ford
and Mathalon 2004, 2005). Following Penfield (1974), who, in turn, was influ-
enced by speculations of John Hughlings Jackson (1932 I, 323; II 399), Feinberg
proposed the radical hypothesis that thinking itself is a sensorimotor func-
tion, indeed, the “highest” sensorimotor function (see also Campbell 1999). As
such, the effort of thinking might be expected to generate a variety of subtle
muscle commands and corollary discharges. According to proponents of the
corollary discharge theory of schizophrenia, the voices and inserted thoughts
that plague sufferers from the disease are the result of failure of efference cop-
ies of cues originating in the prefrontal association areas of the frontal lobe
that initiate internal speech to reach the comparator. The muscle commands
are sent and the reafference signals received. However, because these reaffer-
ence signals are not canceled by efference copies, self-initiated thoughts are
not recognized as such.14 The circumstances here seem to be the opposite of
those that obtained in the case of the paralyzed extraocular muscle. There,
the efference copy was received, but, because of the paralysis, there was no
reafference signal to cancel. Here, there is reafference but no efference copy
received.
The corollary discharge theory of schizophrenia is motivated, in part, by
considerations of explanatory parsimony, the tendency of natural design to
perform exaptations on already existing structures and functions. In this case,
cognitive function is taken to be an exaptation of motor function, the brain
having originated phylogenetically as an organ of motor control. More than
thirty years ago, Wilfrid Sellars proposed that an occurrent thought was a
suppressed proximate short-term propensity to think out loud (Sellars 1975).
Any such suppressed propensity would be expected to generate motor signal
traces that do not culminate in overt behaviors, along with efference copies of
these traces. But how, then, are reafference signals generated without periph-
eral activations? In humans, the spino-cerebellum (one of three main func-
tional regions of the cerebellum) contains complete somatosensory maps of

14
“‘Corollary discharge’ essentially corresponds to a cue for the ‘intention’ component of the
speech generation act; the absence of this cue resulting in the misattribution of source” (Cahill
and Frith 2000, 285).
212 on emotions

the body, and the cerebellum receives input from all levels of the central and
peripheral nervous systems (Kandel et al. 1991, 627, 634). Reafference, there-
fore, need not depend solely on the activation of peripheral receptors but can
derive from the activation of central motor centers, including ganglia in the
spinal cord, thereby generating internal feedback that occurs prior to periph-
eral activation and response. The cerebellum is thus thought to participate
in a feed-forward system that enables the human organism to modulate and
inhibit behaviors that are initiated but not yet consummated (Evarts 1971,
98). This is a very handy ability to possess. Because these feedback signals will
be emotionally valenced, but offl ine, any such feed-forward system will rely
heavily on anticipations and their emotional appraisals, as opposed to responses
to actual impingements on peripheral receptors. By then, tissue damage may
well already have occurred.
Although the cerebellum is part of the brainstem and although the
vestibulo-cerebellum (the phylogenetically oldest functional area of the
cerebellum) communicates directly with the vestibular nuclei in the pons,
Damasio explicitly denies cerebellar function any role in support of the pro-
to-self (1999, 158). However, if the cerebellum is the comparator of efference
copy and reafference signal, or at least if it is a comparator component15 in
a cortico-ponto-cerebellar loop (Ramnani 2006, 512) or corticopontocer-
ebellar system (Kandel et al. 1991, 627), perhaps he shouldn’t. According to
Feinberg and Guazzelli (1999, 199), “Corollary discharge, by permitting the
distinction of self-generated from environmental conscious experiences, con-
tributes to the distinction between self and other.” Notice that inputs to the
cerebro-cerebellum, the functional area of the cerebellum that signals the pre-
frontal cortex and motor cortices by way of the thalamus, originate exclusively
in the pontine nuclei (Kandel et al.1991, 633), which Damasio includes among
proto-self structures.
Jaak Panksepp (2003, 202) expresses agreement with Damasio’s emphasis
on the homeostasis-maintaining functions of the brainstem for support of core
consciousness, but objects that Damasio’s account gives insufficient attention
to motor coordination, the principal function of the cerebellum. The mainte-
nance of homeostasis and bodily integrity that devolves upon Damasio’s pro-
to-self is, in part, a motor process.16 Rather than, or at least in addition to,

15
Comparator function in humans has also been attributed to the subiculum, part of the hip-
pocampal formation, which is a component of the emotion-supporting limbic system (Jeannerod
1997, 171; Gray 1995, 1169).
16
Cf. Jeannerod (1997, 169): “The corollary discharge was thus part of a mechanism for com-
paring outflow signals generated by the command apparatus and reafferent [sic.] signals gen-
erated by the effects of the outflow on the environment. Th is defi nition, which recalls that of
homeostatic regulation, can be extended to any mechanism which implies a comparison between
two causally related signals.”
 E m o t i o n a n d Pe r s o n a l I d e n t i t y 213

asserting “I think and perceive, therefore I am,” says Panksepp (203), the cogito
should assert, “I act and feel, therefore I am.” James would have smiled.17

References
Andreasen, N., S. Paradiso, and D. O’Leary. 1998. “‘Cognitive Dysmetria’ as an Integrative
Theory of Schizophrenia: A Dysfunction in Cortical-Subcortical-Cerebellar Circuitry?”
Schizophrenia Bulletin, 24, no. 2, 203–218.
Blakemore, S. and C. Frith. 2003. “Disorders of Self-Monitoring and the Symptoms of
Schizophrenia.” In The Self in Neuroscience and Psychiatry, T. Kircher and A. David (eds.),
407–424. Cambridge, UK : Cambridge University Press.
Cahill, C. and C. Frith. 2000. “False Perceptions or False Beliefs: Hallucinations and Delusions
in Schizophrenia.” In Method in Madness: Case Studies in Cognitive Neuropsychiatry,
P. Halligan and J. C. Marshall (eds.), 267–291. Hove: Psychiatry Press.
Campbell, J. 1999. “Schizophrenia, the Space of Reasons, and Th inking as a Motor Process.”
Monist, 82, no. 4, 609–625.
Currie, G. and I. Ravenscroft. 2002. Recreative Minds. Oxford: Oxford University Press.
Damasio, A. 1999. The Feeling of What Happens: Body and Emotion in the Making of
Consciousness. San Diego: Harcourt, Inc.
Evarts, E. 1971. “Feedback and Corollary Discharge: A Merging of the Concepts.”
Neurosciences Res. Prog. Bulletin, 9, no. 1, 86–112.
Feinberg , I. 1978. “Efference Copy and Corollary Discharge: Implications for Th inking and Its
Disorders.” Schizophrenia Bulletin, 4, 636–640.
Feinberg , I. and M. Guazzelli. 1999. “Schizophrenia: A Disorder of the Corollary Discharge
Systems that Integrate the Motor Systems of Thought with the Sensory Systems of
Consciousness.” British Journal of Psychiatry, 174, 196–204.
Ford, J. and D. Mathalon. 2004. “Electrophysiological Evidence of Corollary Discharge
Dysfunction in Schizophrenia during Talking and Th inking.” Journal of Psychiatric
Research, 38, no. 1, 37–46.
Ford, M. and D. Mathalon. 2005. “Corollary Discharge Dysfunction in Schizophrenia: Can It
Explain Auditory Hallucinations?” International Journal of Psychophysiology, 85, nos. 2–3,
179–189.
Frith, C. 1992. The Cognitive Neuropsychology of Schizophrenia. Hillsdale, NJ: Lawrence
Erlbaum Associates.
Gray, J. 1995. “A Model of the Limbic System and Basal Ganglia: Applications to Anxiety
and Schizophrenia.” In The Cognitive Neurosciences, M. Gazzaniga (ed.), 1165–1176.
Cambridge, MA : Bradford/MIT Press.
Hume, D. 1739/1978. A Treatise of Human Nature, P. H. Nidditch (ed.), Second Edition. Oxford:
Oxford University Press.
Jackson, J. H. 1932. Selected Writings of John Hughlings Jackson, Vol. 1 and II, J. Taylor (ed.).
London: Hodder and Straughton Limited.
James, W. 1890. The Principles of Psychology, Vol. I and II. New York: Henry Holt and Company.
James, W. 1971a. The Principles of Psychology, excerpted in William James: The Essential
Writings, B. Wilshire (ed.), 44–161. New York: Harper & Row.

17
In fairness, it should be said that Lichtenberg goes on to add (as few seem to know he did)
that “to assume, to postulate the I is [a] practical requirement [ praktisches Bedürfnis]” (1971, 412;
noted by Feinberg and Guazzelli 1999, 197). It is not clear exactly what Lichtenberg meant by this
comment. But it can be interpreted to suggest that the I is represented procedurally rather than
declaratively, and that would be consistent with a modification of the cogito like the one proposed
by Panksepp.
214 on emotions

James, W. 1971b. Essays in Radical Empiricism, excerpted in William James: The Essential
Writings, 162–221.
Jeannerod, M. 1997. The Cognitive Neuroscience of Action. Oxford: Blackwell.
Kandel. E., J. Schwartz, and T. Jessell. 1991. Principles of Neuroscience (Th ird Edition).
Norwalk, CT: Appleton and Lange.
Lazarus, R. 1991. Emotion and Adaptation. Oxford: Oxford University Press.
Lichtenberg , G. C. 1971. Schriften und Briefe, Zweiter Band. Darmstadt: Wissenschaftliche
Buchgesellschaft.
Nietzsche, F. 1887/1967. On the Genealogy of Morals, trans. W. Kaufmann. New York: Random
House.
Panksepp, J. 2003. “The Neural Nature of the SELF: Implications for Understanding
Schizophrenia.” In The Self in Neuroscience and Psychiatry, T. Kircher and A. David (eds.),
197–213. Cambridge, UK : Cambridge University Press.
Parfit, D. 1984. Reasons and Persons. Oxford: Oxford University Press.
Penfield, W. 1974. “The Mind and the Highest Brain-Mechanism.” American Scholar, 43,
237–246.
Prinz, J. 2004. Gut Reactions: A Perceptual Theory of the Emotions. New York: Oxford University
Press.
Ramnani, N. 2006. “The Primate Cortico-Cerebellar System: Anatomy and Function.” Nature
Reviews Neuroscience, 7, 511–522.
Robinson, J. 1988. “Personal Identity and Survival.” Journal of Philosophy, 85, no. 6, 319–328.
Sarnat, H. and M. Netsky. 1981. Evolution of the Nervous System (Second Edition). New York:
Oxford University Press.
Sellars, W. 1975. “Epistemic Principles.” Reprinted in Epistemology: An Anthology (Second
Edition), E. Sosa, J. Kim, J. Fantl, and M. McGrath (eds.), 99–108. Oxford: Blackwell.
van Hoof, J. J. M. 2003. “A Motor Hypothesis of the Origin of Schizophrenia.” Schizophrenia
Research, 62, nos. 1–2, 183–185.
Williams, B. 1973. “The Self and the Future.” In Problems of the Self. Cambridge: Cambridge
University Press, 46–63.
Zahavi, D. 2003. “Phenomenology of Self.” In The Self in Neuroscience and Psychiatry, T. Kircher
and A. David (eds.), 56–72. Cambridge, UK : Cambridge University Press.
 11

The Emergence of Emotion as an Object

of Scientific Study
John Deigh

Nothing excites the academic sleuth more than the discovery of an overlooked
passage in some great book that throws new and better light on the book’s core
argument. Nothing, that is, unless perhaps it is the discovery of a work by some
obscure thinker from another era that clears up a long-standing mystery in a
field of study. The importance of the discovery in either case is shown in its
effect on scholarship in its field. And the thought of having a major impact on
such scholarship, along with the sheer pleasure of solving a scholarly puzzle, is
undoubtedly an incentive to look into the unread or dark parts of an academic
treatise or corpus. Unfortunately, though, it also makes a scholar liable to over-
estimate or misjudge the importance of what he finds there. Misguided claims
as to the significance of the passage or thinker are a common failing of such
investigations.
A case in point is Thomas Dixon’s 2003 monograph From Passions to Emotions:
The Creation of a Secular Psychological Category.1 Dixon’s thesis in this work is that
the modern scientific study of emotions emerged in the nineteenth century
with the shift by writers of psychological treatises to the term ‘emotion’ as the
general name for phenomena that writers of such treatises in earlier centuries
had variously referred to as passions, affections, and sentiments. Dixon iden-
tifies one Scottish writer in particular, Thomas Brown, as the initiator of this
shift in nomenclature. Brown, according to Dixon, was the first to use the term
in this all-purpose way. He did so in his multivolume Lectures on the Philosophy
of the Human Mind,2 which appeared posthumously in 1820. It was in its day a

1
Thomas Dixon, From Passions to Emotions: The Creation of a Secular Psychological Category
(Cambridge: Cambridge University Press, 2003).
2
Thomas Brown, Lectures on the Philosophy of the Human Mind, 3 vols. (Andover: Mark Newman,
1822).

215
216 on emotions

popular work. It went through nineteen editions.3 By the middle of the twenti-
eth century, however, if not before, Brown had become a forgotten figure, one
of many nineteenth-century associationists whose programs became lost to
history with the rise of experimental psychology and behaviorism. Still, Dixon
thinks Brown’s novel use of ‘emotion’ as a general name for phenomena that
philosophers since Plato had systematically studied places him at the forefront
of thinkers who transformed the study of the emotions into a science. Indeed,
he credits Brown with being the pivotal figure in this transformation.4 Brown,
according to Dixon, was “the inventor of the emotions.”5
Curiously, though, Dixon does not attribute to Brown any original ideas or
significant findings that would explain why his novel use of ‘emotion’ merits
such distinction. Apparently, Dixon thinks it is sufficient that Brown made
this revision to the nomenclature of nineteenth-century writing on the work-
ings of the mind. For this reason we should be skeptical of Dixon’s estimation
of Brown’s importance to the transformation of the study of emotions into a
science. That someone is the first to use a term in a new way within a field of
study is not by itself a reason to think that the person has significantly altered
how the subject of that field is understood or significantly advanced the under-
standing of the phenomena studied in that field, much less transformed that
understanding. The new use must also reflect a new way of conceiving of those
phenomena.
Consider the term ‘planet.’ In ancient astronomy, whose understanding of
the universe was geocentric, ‘planet’ was the general name for the seven celes-
tial bodies that move in the sky relative to the stars that to the naked eye appear
fi xed.6 Accordingly, on this understanding, not only are Mercury, Venus, Mars,
Jupiter, and Saturn planets, but the sun and the moon are planets too. Earth,
however, is not. Th is use of the term continued throughout the middle ages
and into the Renaissance. The shift to the modern use began with Copernicus’s
publication in De Revolutionibus Orbium Coelestrium of his heliocentric theory
of the motions of these celestial bodies. Arguing for understanding the earth
as also moving relative to the stars, Copernicus suggested that it too be “reck-
oned as a Planet.”7 And Copernicus then proceeded to argue for understanding
both the earth’s motion relative to the stars and that of Mercury, Venus, Mars,
Jupiter, and Saturn as being a consequence of their circling the sun. His new
use of the term ‘planet’ thus reflected his reconceiving the relation of the earth
to these other celestial bodies. And we celebrate him as the founder of modern

3
“Brown, Thomas,” Encyclopedia Britannica, 11th edition.
4
Dixon, p. 110.
5
Ibid., p. 109.
6
Thomas Kuhn, The Copernican Revolution: Planetary Astronomy in the Development of Western
Thought (Cambridge, MA: Harvard University Press, 1957).
7
From book 1, chapter 9 of De Revolutionibus, as quoted in Kuhn, p. 153.
 The Emergence of Emotion as an Object of Scientific Study 217

astronomy because of this conceptual innovation and not because of the lin-
guistic change it involved.
The fallacy of Dixon’s approach is due to his confusing words with concepts.
A word’s meaning, in one sense of ‘meaning,’ is the concept it expresses, and
the two, the word and the concept, are importantly distinct from each other.
If they were not, two different words could not have the same meaning, and a
single word could not have different meanings. The English word ‘blue’ and the
German word ‘blau,’ for instance, are different words that have the same mean-
ing. That is, they both express the same concept. Similarly, ‘cougar’ and ‘puma’
are exact synonyms. They express the same concept. On the other hand, the
word ‘bark’ can express either of two distinct concepts. Sometimes one uses
it to express the concept of the exterior covering of a tree, and sometimes one
uses it to express the concept of a loud, sharp vocal sound characteristic of dogs.
While significant conceptual change is central to the history of science, the cor-
responding linguistic change is merely a side effect. What is more, linguistic
change may occur in a science without any corresponding conceptual change,
in which case it is no more than a curiosity. Brown’s novel use of ‘emotion’ is
an example, for his work contains no significant conceptual change within psy-
chology that it reflects. Indeed, his new use of ‘emotion’ simply consists in his
renaming a general type of mental state that David Hume had defined in A
Treatise of Human Nature as a secondary impression.8
Specifically, Hume divided all states of mind into two types, impressions
and ideas. The latter are fainter copies of the former. That is, ideas, according
to Hume, chiefly differ from impressions in how they originate and in being
less vivid. The division derives from Locke’s threefold division of ideas. Its
main departure from Locke’s division is the removal of the category of abstract
ideas. Hume, following Berkeley, rejected this type as incoherent. Hume divided
impressions into original and secondary or, as he sometimes says, impressions
of sensation and impressions of reflection. The former arise in the mind inde-
pendently of any prior impression or idea. The latter arise from prior impres-
sions or ideas or both. They are, Hume wrote, “the passions and other emotions
resembling them.”9 Brown too proposes a threefold division of mental states
corresponding to Hume’s division of such states into original impressions,
ideas, and secondary impressions. Like Hume’s, Brown’s taxonomy consists of
a principal division and a subdivision within this principal division. Brown’s
principal division, however, is between original impressions, which he labels
“external affections” and a category that combines ideas and secondary impres-
sions. He labels this second category “internal affections.” He then divides the

8
David Hume, A Treatise of Human Nature, 2nd ed., L. A. Selby-Bigge, ed. (Oxford: Oxford
University Press, 1978), p. 275.
9
Ibid.
218 on emotions

internal affections into ideas and secondary impressions. He labels the former
“intellectual states” and the latter “emotions.”10 Apart from this reshuffl ing
of where one draws the principal division and where one draws a subdivision
within the principal division, his taxonomy is the same as Hume’s. His use
of the term ‘emotions’ for the states Hume referred to as “passions and other
emotions” is merely a linguistic novelty. Both express the same concept, which
Hume defined as impressions that arise from prior impressions or ideas.
While Dixon’s identification of Brown as the pivotal figure in the trans-
formation of the study of emotions into a science is misguided, the question
interest in which led Dixon to make this identification certainly merits inves-
tigation. How did the study of phenomena that are variously named passions,
sentiments, and affections that goes back to Plato turn into the modern science
of emotions? Did the convergence by nineteenth-century writers of psychologi-
cal treatises on using the term ‘emotion’ as the general name for these phenom-
ena merely represent a change in nomenclature? That is, did it merely consist
in the gradual replacement of the several terms that earlier writers had used to
denote them with this single term? Or did it also result in a significant change
in how the object of the study was conceived? The remainder of this essay will
deal with these questions. My thesis is that the change in nomenclature does
represent a significant change in how the object of the study was conceived.
Specifically, the shift to which Dixon refers in the title of his book, the shift
from using ‘passion’ as the chief term for the object of this study to using ‘emo-
tion,’ represents, I will argue, a significant conceptual change whose inception
is traceable to Hume but whose completion, in the work of William James, took
more than a century to reach.

II
Hume, according to Dixon, was the first among writers of psychological trea-
tises to use the term ‘emotion’ liberally.11 Earlier writers made little or no use
of it, or they gave it a special use.12 At the same time, Dixon observes, Hume did
not use the term systematically. In the history of how ‘emotion’ came to replace
all other terms as the principal name for affective mental states, Dixon tells us,
Hume and other writers after him, like Joseph Priestly and Adam Smith, pre-
pared the way for Brown’s innovation by including ‘emotion’ among the terms
that they used to describe affective phenomena. At the same time, they did not

10
Brown, I, 249–253, 254–255.
11
Dixon, p. 104.
12
Descartes, for example, reserved “emocion” for certainly purely intellectual states of
affection.
 The Emergence of Emotion as an Object of Scientific Study 219

define it or otherwise give it a distinctive role in their accounts of these phe-

nomena. Consequently, Dixon treats Hume as only a bit player in the history
he recounts.
Such treatment, however, underestimates Hume’s importance in this his-
tory. While Hume did not define ‘emotion’ or use it systematically, one can
nonetheless discern in certain passages in which he used it a meaning that is
distinct from what he meant by ‘passion,’ and his use of the term with this
meaning is consequential to the later emergence of the scientific study of emo-
tions. Of course, whether Hume’s use of the term with this meaning contrib-
uted to how emotions came to be conceived as the science emerged and grew or
whether it merely anticipated it is hard to establish. But given Hume’s influence
on later thinkers, it is reasonable to treat him as having planted the seed from
which the later concept germinated.
‘Passion’ was the main term Hume used for the secondary or reflective
impressions that are the subject of book II of the Treatise. Hume signaled this
immediately by giving book II the title “Of the Passions.”13 In introducing the
subject, Hume drew two distinctions, that between calm and violent passions
and that between direct and indirect passions. Let us consider his distinction
between calm and violent passions first. In explaining it, Hume wrote:

The reflective impressions may be divided into two kinds, viz. the calm
and the violent. Of the first kind is the sense of beauty and deformity
in action, composition, and external objects. Of the second are the pas-
sions of love and hatred, grief and joy, pride and humility. This division
is far from being exact. The raptures of poetry and music frequently
rise to the greatest height; while those other impressions, properly
call’d passions, may decay into so soft an emotion as to become, in a
manner, imperceptible. But as in general the passions are more violent
than the emotions arising from beauty and deformity, these impres-
sions have been commonly distinguished from each other.14

Much later in book II, when Hume returned to the topic of calm passions, he
observed that they are sometimes mistaken for reason:

Now ’tis certain, there are certain calm desires and tendencies, which,
tho’ they be real passions, produce little emotion in the mind, and are
more known by their effects than by the immediate feeling or sen-
sation. These desires are of two kinds; either certain instincts origi-
nally implanted in our natures, such as benevolence and resentment,

13
Treatise, 275.
14
Ibid.
220 on emotions

the love of life, and kindness to children; or the general appetite to

good and aversion to evil consider’d merely as such. When any of these
passions are calm, and cause no disorder in the soul, they are very
readily taken for the determinations of reason, and are suppos’d to
proceed from the same faculty, with that, which judges of truth and
falsehood.15

Yet passions of either kind, Hume immediately went on to note, are not neces-
sarily calm:

Besides these calm passions, which often determine the will, there are
certain violent emotions of the same kind, which have likewise a great
influence on that faculty. When I receive any injury from another,
I often feel a violent passion of resentment, which makes me desire
his evil and punishment, independent of all considerations of pleasure
and advantage to myself. When I am immediately threatened with
grievous ill, my fears, apprehensions, and aversions rise to a great
height and produce a sensible emotion.16

Having thus pointed out that a passion’s influence on the will is not a func-
tion of how violent the passion is, Hume then distinguished between these two
aspects of a passion:

’Tis evident passions influence not the will in proportion to their vio-
lence, or the disorder they occasion in the temper; but on the contrary,
that when a passion has once become a settled principle of action, and
is the predominant inclination of the soul, it commonly produces no
longer any sensible agitation. As repeated custom and its own force
have made every thing yield to it, it directs the actions and conduct
without that opposition and emotion, which so naturally attend every
gust of passion. We must, therefore, distinguish betwixt a calm and a
weak passion; betwixt a violent and a strong one.17

On the evidence of these passages, we may conclude that Hume understood

the difference between a calm passion and a violent one to consist in the amount
of emotion with which a passion occurs. The more emotion the more violent the
passion. Calm passions are passions that occur with so little emotion as to be
“in a manner, imperceptible”; violent ones are passions that occur with enough

15
Ibid., p. 417.
16
Ibid., pp. 417–418.
17
Ibid., p. 418–419.
 The Emergence of Emotion as an Object of Scientific Study 221

emotion to create considerable agitation in the soul. At the same time, we can
infer from Hume’s observation that calm passions no less than violent ones
influence the will, that the amount of emotion with which a passion occurs is
independent of the passion’s motivational force. Accordingly, Hume used ‘emo-
tion’ as a term for that feature of a passion that characterizes how violent the
passion is and understood this feature to be distinct from the passion’s being a
motive or spring of action. He understood it, that is, as a phenomenal property
of passions, which as such implies nothing about their motivational strength
or power to produce action.
Hume, however, did not consistently use ‘emotion’ in this way. He also used
it as a common name for secondary impressions, and indeed for secondary
impressions for which he also used the term ‘passion.’ Note, for instance, his
referring in the third of the four previous passages to “violent emotions” which
are of the same kind as certain calm passions. This usage complicates interpreta-
tion of Hume because it is hard to square with the first way in which he used the
term. Indeed, on a superficial reading, the two ways appear to clash with each
other, for traditionally passions are understood to be motives. Consequently, it
appears that Hume, by using ‘emotion’ in this second way, implied, contrary to
what the first way in which he used it implies, that emotion has motivational
force. The appearance, though, is misleading. Hume’s concept of a passion is not
the traditional one. Passions, as Hume conceived of them, are not necessarily
motives. Indeed, some passions, according to Hume, are not motives. So his
two ways of using ‘emotion’ do not clash with each other.
Nonetheless, they are independent of each other. Neither, that is, derives
from the other.18 Nor did he try to avoid confusion by restricting his use of
‘emotion’ as a common name for secondary impressions to secondary impres-
sions that were not motives. To the contrary, he readily used it to denote pas-
sions that were motives. A good example occurs in the course of his well-known
argument for the inertness of reason. “’Tis obvious,” he wrote, “that when we
have the prospect of pain or pleasure from any object, we feel a consequent
emotion of aversion or propensity, and are carry’d to avoid or embrace what
will give us this uneasiness or satisfaction.”19 In the face of passages like this
one, then, we must conclude that Hume used ‘emotion’ with two different
meanings without ever explicitly distinguishing them. The potential for con-
fusion was, if anything, exacerbated by Hume’s departure from the traditional
concept of a passion.
Hume’s departure from this concept is a consequence of the second of the two
distinctions among the passions that he drew in the opening section of book II.

18
Nor is there any evidence that Hume meant to be using ‘emotion’ metonymically when he
used it to as a name for secondary impressions.
19
Ibid., p. 414.
222 on emotions

This is his distinction between direct and indirect passions. Direct passions are
motives of action. They arise immediately from pleasure and pain or from the
prospect of pleasure and pain and, in either case, produce a volition to embrace
what gives pleasure or avoid what gives pain.20 In addition, Hume identified
certain direct passions that “arise from a natural impulse or instinct, which is
perfectly unaccountable.”21 These include hunger and other bodily appetites, the
desire for revenge or punishment of one’s enemies, and the desire for the happi-
ness of those whom one loves. Like direct passions that arise immediately from
pleasure and pain, these too produce volitions to secure their objects. Indirect
passions, by contrast, do not. For this reason, they are not in themselves motives.
At the same time, some give rise to direct passions and are therefore sources of
motives. These are, chiefly, love and hatred, which give rise to a desire for the
happiness of one’s beloved, which Hume identified with benevolence, and to a
desire for the misery of one’s enemies, which Hume identified with anger. Other
indirect passions, however, do not give rise to direct passions. These are, chiefly,
pride and humility. They are neither motives nor sources of motives. Hume set
out this difference between the two sets of indirect passions in the section of
the Treatise on benevolence and anger:

The passions of love and hatred are always followed by, or rather,
conjoin’d with benevolence and anger. ’Tis this conjunction, which
chiefly distinguishes these affections from pride and humility. For
pride and humility are pure emotions in the soul, unattended with
any desire and not immediately exciting to action. But love and hatred
are not compleated within themselves, nor rest in that emotion, which
they produce, but carry the mind to something farther.22

The traditional concept of a passion, in modern philosophy, is well illus-

trated in Descartes’ Les Passions de l’ame (The Passions of the Soul ). Descartes
held that all passions give rise to volitions, or as he put it, in keeping with his

20
Hume fi rst characterizes direct passions as “aris[ing] immediately from good or evil, pain
or pleasure.” (Ibid., p. 276; see also p. 399). Sometimes, though, he speaks of them as arising from
the prospect of pleasure or pain; see p. 414. He also describes volitions as impressions that arise
immediately from pleasure and pain while at the same time excluding them from being passions:
“Of all the immediate effects of pain and pleasure, there is none more remarkable than the will.”
(Ibid., p. 399; see also, p. 574). Later, however, he implies that the will is determined by passions:
“Beside these calm passions, which often determine the will, . . . ” (p. 417). I believe the latter the-
sis best fits his argument and have interpreted him accordingly, that is, by presenting as his view
that volitions are produced by direct passions. Note that in either case Hume excludes volitions
from the category of passions: “tho’, properly speaking, [the will] be not comprehended among the
passions, yet . . . we shall here make it the subject of our enquiry” (p. 399).
21
Ibid., p. 439.
22
Ibid., p. 367.
 The Emergence of Emotion as an Object of Scientific Study 223

view that the will is free, they all dispose the will to act as they bid. Thus, in
article 40 of Les Passions de l’ame, Descartes wrote, “Le principal effet de toutes
les passions dans les homes est qu’elles incitent et disposent leur âme à vouloir
les choses auxquelles elles préparent leur corps: en sorte que le sentiment de
la peur l’incite à vouloir fuir, celui de la hardiesse à vouloir combattre: et ainsi
des autres.” (The principal effect of all the passions in men is that they incite
and dispose their soul to will the things for which they prepare their body,
so that the sensation of fear incites it to flee, that of boldness to will to do
battle, and all the rest.)23 What Descartes described as the principal effect of
all passions, a volition, is for Hume the principal effect of direct passions only.
By distinguishing indirect passions from direct passions by virtue of their not
giving rise to volitions, Hume therefore not only departed from the traditional
concept of a passion but rejected as well the understanding of a passion as an
inclination to act that, through an exercise of will, a volition, one either fol-
lows or resists. Because Hume regarded volitions as secondary impressions,24
he rejected Descartes’ model of the will as acting on a passion and endorsed
instead a mechanical model in which a so-called act of will, a volition, is a sec-
ondary impression produced by a direct passion.
Even more instructive, perhaps, for seeing the significance of Hume’s depar-
ture from the traditional concept is Thomas Reid’s criticism, in his Essays on
the Active Powers of the Human Mind, of Hume’s notion of a calm passion.25
Reid objected in particular to Hume’s thesis that calm passions are “readily
taken for the determinations of reason.” Hume had advanced this thesis to
explain how the great majority of philosophers, both ancient and modern, had
come to the false view of reason and passion as combatants that oppose each
other in struggles to determine the will. Reason, Hume had argued, is the
slave of passion. Being powerless to oppose passion, it is necessarily under
its rule. He then, to supplement this argument, gave his explanation of how
philosophers had come to the false view of the combat between reason and
passion. Both the argument and the explanation, Reid rejoined, were based on
an abuse of language. Passions, Reid observed, are, in the most general sense
of the term, “principles of action” that are also commonly referred to as appe-
tites and affections.26 They are the propensities to action that humans share

23
René Descartes, Les Passions de l’âme (Gallimard, 1988), Art. 40. For translation see
Descartes, The Passions of the Soul, S. Voss trans. (Indianapolis, IN: Hackett Publishing Co., 1989),
Article 40.
24
“The impressions, which arise from good and evil most naturally, and with the least prep-
aration are the direct passions of desire and aversion, grief and joy, hope and fear, along with
volition.” Ibid., p. 438.
25
Thomas Reid, Essays on the Active Powers of the Human Mind (Cambridge, MA: MIT Press,
1969).
26
Reid, p. 70.
224 on emotions

with brutes and must therefore be understood to spring from the irrational
part of the human frame. Yet ‘passion,’ Reid maintained, is more commonly
used with a more specific sense that distinguishes passions from other natural
desires and affections.27 In this more specific sense, its ordinary sense, pas-
sions are principles of action whose strength agitates the body and clouds the
understanding. Indeed, their strength as motives of action corresponds to the
violence with which they occur, for the stronger the passion the more the body
is uncontrollable and the mind is clouded. As Reid saw things, Hume’s talk of
calm passions, passions that prompt action without any sensible agitation, is
oxymoronic.
A defender of Hume could, of course, grant Reid’s point about common
usage without conceding any significant error in Hume’s position. Hume’s
position depends on there being propensities to action that influence the
will and produce action without any sensible agitation. Since Reid allowed
that there were such propensities, since he allowed that natural desires and
affections could be calm in Hume’s sense, his criticism of Hume for his sham
use of ‘passion’ appears to be no threat to Hume’s thesis that these calm
affections and desires are commonly mistaken for determinations of rea-
son. But Reid’s criticism goes deeper into Hume’s thought than its linguistic
surface. It strikes rather at Hume’s distinction between the motivational
strength of a passion and its violence. The distinction underpins Hume’s
thesis, and Reid, in assuming a correspondence between the strength of a
passion and its violence, thus rejected the distinction. In Reid’s view, one
could not, pace Hume, mistake in a man who resolutely resists temptation a
passion for the determinations of his reason, for if the man’s will to resist
temptation derived from a passion, the violence of that passion would be
greater than that of the passion it defeated. Its presence and operation
would, therefore, be unmistakable. And if one were to suppose instead that
the man resisted the passion in order to secure a good that he desired and
his desire for that good were calm, then its being calm, Reid held, would
mean that it neither weakened the man’s control over his body nor clouded
his judgment. Consequently, his resistance would have to be due to a deter-
mination of reason, for the desire, being calm, would have been too weak to
check the passion he resisted. Hume’s idea that such resistance could be the
work of a calm passion was, in other words, given a correspondence between
the violence of a passion or other affection and its motivational strength,
incoherent.
Reid, as a result of his rejecting Hume’s distinction between the strength
of a passion and its violence, denied in effect that the amount of emotion with
which a passion occurred was a feature of a passion distinct from its power to

27
Ibid., pp. 177–178.
 The Emergence of Emotion as an Object of Scientific Study 225

produce action. Reid, then, because he held to the traditional concept of pas-
sion, did not allow the possibility of a passion’s not being a motive. It was for
him inconceivable. For Hume, by contrast, it was certainly conceivable. The dis-
tinction between the amount of emotion in a passion and a passion’s power to
produce action guaranteed its possibility. After all, one need only think of a
turbulent state of mind that had no tendency to produce action to conceive of
such a passion. And Hume in distinguishing indirect passions from direct ones
did just that. His account of pride and humility as passions that were “pure
emotions” fit the bill exactly.28 Reid, as we have seen, treated motivational
strength as inseparable from turbulence. The term ‘emotion’ appears nowhere
in his chapter on the passions.29
Hume’s characterization of pride and humility as pure emotions thus
reflects his distinction between the violence of a passion and its strength.
It reflects, in other words, his use of ‘emotion’ to denote that feature of a
passion that characterizes how violent it is together with his understand-
ing of this feature as independent of a passion’s power to produce action. By
virtue of this understanding, Hume’s concept of a passion departed from the
traditional one. Later writers who took up Hume’s other use of ‘emotion,’ his
use of it as a general name for secondary impressions, did so without also
departing, as Hume did, from the traditional concept of a passion. Brown,
for instance, while he took ‘emotion’ as the general name for mental states
that writers before him had variously called “passions” “affections,” “appe-
tites,” and “desires,” took it to express the same traditional concept of those
states as they (save for Hume) expressed by these other names. His saying,
in keeping with Reid’s understanding of these states as principles of action,
that by “emotion” he understood “all, or nearly all the mental states, which
have been classed, by others, under the head of active powers,” implies as
much. 30 Generally, then, while Hume’s use of ‘emotion’ as a name for sec-
ondary impressions, including passions, emerged as the prevailing one in
the nineteenth century, his use of the term as a name for a feature of a pas-
sion that is distinct from that of the passion’s motivational strength did not
acquire a similar following. The notion it expressed disappeared for much of
the century. Not until William James published his great work in psychology
did it—or rather something like it—reappear and come fi rmly and distinctly
into view. 31

28
Hume, p. 367.
29
Reid, pp. 175–186.
30
Brown, II, 251.
31
C. G. Lang, independently of James’s publication, published a similar account of emotion
around the same time. See “The Emotions,” trans. I. A. Haupt, in William James and Carl Lang,
The Emotions, ed. Knight Dunlap (Baltimore: Williams and Wilkins, Co., 1922), pp. 33–90.
226 on emotions

III
To understand the lasting significance of James’s work, however, one must
attend to more than his writings on emotion.32 While his article “What Is an
Emotion?” is often cited as a landmark in the field, much of its significance
depends on ideas and arguments James published elsewhere and brought
together in the famous ninth chapter of The Principles of Psychology.33 This chap-
ter contains James’s attack on the conceptual scheme for mental states char-
acteristic of classical British empiricism, the scheme of Locke’s psychology and
Hume’s along with that of each of a host of minor philosophers, like Brown,
who belong to this tradition. The scheme consists, first, in treating all mental
states as either simple or complex, and if complex, then analyzable into simple
states, and, second, in treating these states as occurring in the mind singly,
discretely, and serially. Hume’s wonderful description of the mind as “a kind of
theatre, where several perceptions successively make their appearance; pass,
re-pass, glide away, and mingle in an infinite variety of postures and situations”
nicely captures this empiricist conception of mental states.34 James rejected it
as both false to our experience, when carefully considered, and inconsistent
with any understanding of consciousness as a product neurophysiological pro-
cesses. Conscious experience, which fi lls our waking lives, does not break down
into sequences of individual, recurring impressions and ideas, “mental atoms
or molecules” as James liked to call them.35 In a seeming rejoinder to Hume,
James declared, “A permanently existing ‘idea’ . . . which makes its appearance
before the footlights of consciousness at periodical intervals is as mythological
an entity as the Jack of Spades.”36
James then pressed this attack on the empiricist program for studying
emotions. “The trouble with the emotions in psychology,” he wrote, “is that
they are regarded too much as absolutely individual things. So long as they are
set down as so many eternal and sacred psychic entities, like the old immu-
table ‘species’ in natural history, so long all that can be done with them is

32
See my “Emotions: The Legacy of James and Freud,” International Journal of Psychoanalysis
82 (2001), 1247–1256; reprinted in John Deigh, Emotions, Values, and the Law (New York: Oxford
University Press, 2008).
33
William James, The Principles of Psychology (New York: Dover Press, 1950), vols. 1 and 2.
Much of chapter ix appeared in article form before the publication of “What Is an Emotion?” See
James, “On Some Omissions of Introspective Psychology,” Mind 9 (1884): 1–26.
34
Hume, p. 253.
35
James, Principles of Psychology, I, 230.
36
Ibid., I, 236. Th at James has Hume in mind is then made explicit on the next page. “A neces-
sary consequence of the belief in permanent self-identical psychic facts that absent themselves
and recur periodically is the Humian [sic.] doctrine that our thought is composed of separate inde-
pendent parts and is not a sensibly continuous stream. . . . Th is doctrine entirely misrepresents
the natural appearances” (I, 237).
 The Emergence of Emotion as an Object of Scientific Study 227

reverently to catalogue their separate characters, points, and effects. But if we

regard them as products of more general causes (as ‘species’ are now regarded
as products of heredity and variation), the mere distinguishing and catalogu-
ing becomes of subsidiary importance.”37 James, as this passage makes clear,
aimed at reorienting the study of emotions. Progress in the scientific study
of emotions, he believed, required investigating their general causes, and to
investigate them psychologists must study the different external conditions
and events to which emotions are reactions and the bodily processes that
mediate those reactions.
To this end, James defined an emotion as the feelings of bodily changes, as
they occur, which changes directly follow upon perception of an exciting fact.38
Plainly, he did not put forward this definition to establish a category of things
whose “separate characters, points, and effects” can be catalogued. He put it
forward so as to turn our attention in the study of emotions to the general
causes of the bodily changes that occur in episodes of emotion. While the feel-
ings of those changes, as they occur, are what ‘emotion,’ as he defined the term,
means, they are not, he held, isolable, recurring units of consciousness whose
nature and composition is the subject of scientific study. They are rather, like
rapids and eddies in a river, to be understood as disturbances and agitations
in an unbroken stream of thought, which one studies by examining the forces
and conditions that produce such changes in the flow. The study of emotions, in
other words, on James’s view, is the study of the causes of bodily changes that
are made manifest in the mind through turbulent or stirring feelings.
Because these bodily changes include the overt, voluntary behavior of which
passions, on the traditional concept of them, are motives, James’s definition
of an emotion departs from this concept. James, to be sure, was aware of his
definition’s unorthodoxy. Indeed, he regarded its being contrary to common
sense on this point as its hallmark. Immediately, after giving his definition,
he wrote:

Common sense says, we lose our fortune, are sorry and weep; we meet
a bear, are frightened and run; we are insulted by a rival, are angry
and strike. The hypothesis here to be defended says that this order of
sequence is incorrect, that the one mental state is not immediately
induced by the other, that the bodily manifestations must first be
interposed between, and that the more rational statement is that we
feel sorry because we cry, angry because we strike, afraid because we
tremble, and not that we cry, strike, or tremble, because we are sorry,
angry, or fearful, as the case may be.39

37
Ibid., II, 449.
38
Ibid.
39
Ibid., II, 449–450.
228 on emotions

Two things are noteworthy about this passage. First, it shows that James, by
defining an emotion as the feelings of bodily changes, not only took an emotion
to be the sensible agitation in the soul that Hume had identified as the determi-
nant of a passion’s violence but also understood that, as such, an emotion was
not a motive of action. To this extent, his definition recovers the notion Hume
had in mind in using ‘emotion’ to denote the feature of passions that character-
ized how violent they are. Second, it spells out the chief implication of James’s
definition, its incompatibility with the classical empiricist taxonomy of mental
states. For by pointing out that, contrary to common sense, an emotion is not
the direct consequence of a perception or thought but rather arises only after
the interposition of those bodily changes that a perception or thought excites,
James jettisoned one of the principal categories of mental states on the clas-
sical empiricist taxonomy. On that taxonomy, emotions, passions, affections,
and the like all belong to the category of states of mind that arise from ante-
cedent sensory or intellectual states. For Locke these were internal sensations;
for Hume, they were secondary or reflective impressions. James’s definition in
effect removes them from the study of psychology. Consequently, his definition
differs from the notion Hume had in mind in that it does not represent an emo-
tion as a feature of some mental state that may also have, independently of that
feature, the power to produce action. On James’s account of human psychology,
there are no such mental states.
The question, then, given that James had removed passions, as traditionally
conceived, from human psychology, is how he accounted for human motives.
The answer is found in the chapter, in the Principles, on instincts, chapter
xxiii, which immediately precedes the one on emotions. This chapter and the
one immediately preceding it, chapter xxii, “The Production of Movement,”
concern the bodily actions whose feelings James identified as emotions. Many
of these bodily actions are autonomic. Shortness of breath, a palpitating heart,
and shivering are examples. James discussed such actions in chapter xxii. He
then dealt with voluntary actions in chapter xxiii. Each instinct, according to
James, is an innate tendency to act, a tendency built into the nervous system
of human beings and other animals as a result of natural selection.40 Each
such tendency, each instinct, operates through an impulse to action.41 While
instincts originally operate blindly, which is to say, reflexively, in response
to certain sensory stimuli, they become in animals capable of foresight the
source of voluntary actions through repeated experiences of their operation.
An animal’s familiarity with its circumstances enables it to anticipate the

40
James, in the chapter on instinct, does not explain instincts as the result of natural selec-
tion but treats them as innate and the result of adaptation. In the last chapter of Principles, how-
ever, in the section “The Origin of Instincts,” he argues for the superiority of Darwin’s theory of
natural selection to Lemarck’s theory of the inheritance of acquired traits. See II, 678–688.
41
Ibid., II, 385.
 The Emergence of Emotion as an Object of Scientific Study 229

impulses they excite and to encourage or resist them by thinking of objects

that strengthen or counter them. In this way, the animal gains some measure
of control over its muscular movements and so its actions. Correspondingly,
while the earliest operations of an instinct, being entirely reflexive, produce
actions of whose ends the animal is incognizant, later operations produce
actions whose ends it cognizes and, to the extent the actions fall within its
control, it consciously pursues. Accordingly, instincts qualify as motives of
these actions. As principles of or propensities to action that humans share
with other animals, they replace passions in James’s account of human
psychology.
James’s discussion of the instincts special to human beings confi rms this
observation. Human beings, James maintained, possess an enormous vari-
ety of instincts. He included among them tendencies to attack what opposes
one, to flee what threatens one, and to comfort one’s kin and conspecif-
ics. In each case, one perceives some exciting object, and the perception,
then, mobilizes one to action. James characterized the operations of these
instincts as anger, sympathy, and fear. Later in the discussion, he took up
the human tendency to acquire or appropriate things that please and noted
how this tendency can turn into the impulse to harm others who possesses
things that one covets. He characterized such impulses as impulses of envy
and jealousy. Finally, he added the tendency of parents, mothers, in par-
ticular, to cleanse and feed their young and characterized the operations of
this instinct as parental love. Each of these characterizations identifies an
instinct by the type of emotion that the action it tends to produce typically
expresses. These characterizations suggest a correspondence between the
feelings of bodily changes that the operations of certain instincts produce
and the different types of emotion that theorists of the mind since Plato
had distinguished and studied. If James had affi rmed this correspondence,
then his distinction between emotions and the instincts whose operations
produced them would have mapped onto Hume’s distinction between a pas-
sion’s degree of violence and its motivational strength or power to produce
action.
James, however, denied it. Specifically, he denied that one’s feeling an
emotion of one of these types—fear, anger, joy, or the like—necessarily
resulted from one’s undergoing a bodily movement or syndrome of bodily
movements that the operations of the corresponding instinct produced.42
Indeed, he denied that it necessarily resulted from one’s undergoing a bodily
movement or syndrome of bodily movements of any determinate type.
Because of the indefi nite variability among people in their reflex responses
to the perception of the same exciting object, James argued, no determinate

42
Ibid., II, 442
230 on emotions

type of bodily change or syndrome of bodily changes corresponds to any of

these types of emotion. A frightening object may induce fl ight in one per-
son and paralysis in another. Triumph may cause one person to jump and
another to prance. And a similar variability among people holds of auto-
nomic responses as well:

The moment the genesis of an emotion is accounted for, as the arousal

by an object of a lot of reflex acts which are forthwith felt, we imme-
diately see why there is no limit to the number of possible different emo-
tions which may exist, and why the emotions of different individuals may
vary indefinitely, both as to their constitution and as to objects that
call them forth.43

It follows, therefore, on James’s view, that motives of voluntary action origi-

nate in instincts that can have no more than a statistical association with the
emotions to which he referred homonymously. James’s separation of emotion
from the springs of action was thus even greater than the one Hume implied
in using ‘emotion’ to capture the feature of a passion that characterized how
violent it was as distinct from how powerful a motive it was.
Hume, in so using ‘emotion,’ identified a distinctive feature of passions,
their sensible turbulence or agitation, which, departing from the traditional
concept of passion, he took to be independent of a passion’s motivational
strength. He did not, however, treat this feature as a special object of psy-
chological study. His purpose in identifying it as a feature distinct from a
passion’s motivational strength was to secure his notion of a calm passion
whose operations were commonly mistaken for determinations of reason.
Passions were the object of his study. James, by contrast, went further. Not
only did he identify emotion with sensible turbulence or agitation that had
no motivational potential and was therefore independent of whatever had
motivational strength, but he also, by removing from the study of psychol-
ogy the type of mental state to which passions, on the classical empiricist
taxonomy, belonged, made such turbulence—turbulence directly produced
by bodily changes and not prior mental states—the object of psychological
study. A major consequence, then, of his applying his attack on the classi-
cal empiricist conceptual scheme for mental states to the empiricist program
for studying emotions was to redefi ne the study of emotions as the study of
such states of arousal. His redefi nition represents a significant change in how
emotions, as the object of scientific study, are conceived. He thus brought
into full form the transformation in how the object of the study is conceived
that Hume had begun.

43
Ibid., II, 454.
 The Emergence of Emotion as an Object of Scientific Study 231

IV
Unsurprisingly, James’s definition of emotion continued to be controversial for
decades after he first proposed it. The definition makes sense only when taken in
conjunction with James’s attack on the classical empiricist conceptual scheme
for mental states. This scheme had been a staple of Anglo-American writing on
the workings of the mind for nearly two centuries, and for this reason one should
expect philosophers and psychologists whose understanding of how the mind
works was defined by this scheme to find the definition deeply problematic. So
it was with many of James’s contemporaries and their students. Hence, their
considerable resistance to the definition was no doubt due in large part to their
inability to shift from seeing mental life through the lenses of classical British
empiricism to seeing it as James urged them to see it. A review of the objections
commonly raised to his treatment of emotions supports this hypothesis.44 Many
of them implicitly presuppose the very conceptual scheme James was attack-
ing. Still, while his definition remained controversial, the program of studying
emotions by studying exclusively the physiological reflexes assumed to be their
conditions became the dominant program in experimental psychology. Such a
program, of course, served the interest in applying the methods of experimen-
tal science to psychology: physiological reflexes are objectively measurable and
mechanistically explicable. But it still needed theoretical support, and James’s
treatment served this need far better than any of the more traditional accounts.
As in other areas of science, change in how the phenomena studied are conceived
occurred in tandem with change in the methods by which they were studied.
While the concept of emotions as states of arousal that James brought into
full form fits the methods of experimental psychology, the growth of interest
during the latter half of the twentieth century in cognitive theories in psychol-
ogy and the corresponding decline of interest in behaviorist theories put the
concept under new pressure. Specifically, as philosophers and psychologists
began increasingly to focus on the role of cognition in emotion, they came to
question the adequacy of the concept. In view of the cognitive character of emo-
tions, they argued, the concept’s coverage was too broad. James saw his defini-
tion of emotion as serving, in virtue of the breadth of its coverage, to break the
grip of the empiricist fi xation on taxonomy. But because his definition applies to
any feeling of reflex bodily movement that the perception of an object produces,
it is insensitive to the difference between states of arousal that involve atten-
tion to an object and states that do not. And the concept of emotion that the
definition helped to entrench in modern psychology inherited this problem.
Of course, experimentalists could continue to take the object of their study
to be a state of arousal that the perception of an object produces regardless of

44
See Dixon, pp. 214–216.
232 on emotions

whether it also involves attention to its stimulus or some other object. But their
doing so, once the concept they assumed came under criticism, was bound to be
problematic. For the term ‘emotion,’ even after the entrenchment in modern
psychology of this concept, continued to be used as a name for the same states
of mind that had long been referred to as passions, sentiments, affections, and
desires, and these states are all states in which the mind is directed at some object.
Grief, fear, joy, anger, hope, sadness, regret, and the many other states that writ-
ers on emotion before James devoted pages of their treatises to distinguishing
and classifying are all states directed at or toward some object. One grieves over
a death, fears a snake, feels joy at the team’s victory, hopes for sunshine, is angry
at being snubbed, regrets a misdeed, and so on. Plainly, an account of such states
that applied, mutatis mutandis, to states of arousal in which the mind was not
directed at some object would fail to account for the object-directedness of the for-
mer states. Hence, either the study of emotions in modern psychology was not a
study of the same phenomena that philosophers and psychologists since Plato had
developed theories of the passions, sentiments, affections, and the like to explain,
or the concept of emotion on which it proceeded was defective. In this way the
problem for those who had entrenched James’s concept in their work was set.
While James could unhesitatingly cite both feeling fear upon looking down from
a precipice and feeling sick and dizzy upon looking at horse blood dripping from a
stick as examples of emotion,45 once one notices the difference between the two,
that in the former the object of fear is falling and in the latter there is no object of
sickness and dizziness, his definition’s excessive breadth should be apparent.
The influence of such criticism in changing how a great many philosophers
and psychologists came to conceive of emotions in the latter half of the twentieth
century reflects the rapid rise of cognitive theories of the mind in these thinkers’
disciplines. In the study of emotions Bob Solomon was a leader in this movement.
In his book The Passions, he pioneered the cognitivist criticism of the concept of
emotion that prevailed in modern psychology.46 Those of us who have followed
him in this endeavor owe him an intellectual debt of immeasurable value.

45
James, Principles, II, 457. James gives this example in answer to the objection to his account
that perception alone, that is, without the intermediation of an emotion, cannot excite any bodily
changes. To this objection he answers, “The writer well remembers his astonishment, when a boy
of seven or eight, at fainting when he saw a horse bled. The blood was in a bucket, with a stick in
it, and, if memory does not deceive him, he stirred it round and saw it drip from the stick with
no feeling save that of childish curiosity. Suddenly the world grew black before his eyes, his ears
began to buzz, and he knew no more. He had never heard of the sight of blood producing faintness
or sickness, and he had so little repugnance to it, and so little apprehension of any other sort of
danger from it, that even at that tender age, as he well remembers, he could not help wondering
how the mere physical presence of a pailful of crimson fluid could occasion in him such formidable
bodily effects.”
46
Robert C. Solomon, The Passions: The Myth and Nature of Human Emotion (Garden City, NY:
Anchor Press/Doubleday, 1976).
 I N DE X
Action, 19, 143, 151, 163, 165, 210–11,
227–30
involuntary, 70, 110, 222, 227–30
voluntary, 120, 227–30
Aesthetics, and emotion, 86, 89, 103, 106,
111–12, 219
fi lm, 78–79, 82, 86, 92, 141–45, 167
music, 55–58, 123
theatre, 61–62, 73–74, 88–89, 141–45
Affectivity, 104–5, 107–8, 209, 219–21
See also Sensation
Amusement, 76–98
Anger, 9–11, 20–22, 24, 26, 34–36, 56, 67,
69–70, 87–88, 94, 103–4, 113, 143, 145,
183, 195, 222
Animals, 8, 26, 55, 57, 210, 228–29
Anscombe, G. E. M., 2
Aristotle, 16–18, 27, 68, 73, 76–77, 81,
99–103, 113–15, 120, 123, 134
Austin, J. L., 127, 130, 141
Autonomic response. See Reflex
Averill, James, 105–7, 109
Baier, Annette, 157
Bain, Alexander, 81–83
Barschack, Lior, 45
Behaviorism, 2–3
Belief, 3, 60, 72–74, 88–90, 143–45, 153–54
Benevolence, 17, 25, 27, 219–20, 222
Boruah, Bijoy, 59, 60n, 74
Bowlby, John, 172
Brain, 109, 199–200, 203–4, 211, 215–16
Briggs, Jean, 9–10
Brown, Thomas, 215–18, 225–26
Burke, Edmund, 133
Children, 9–10, 26, 56–57, 133, 153–54,
184–85, 193–94
Cognition, 6–7, 55, 57, 70, 76, 87–90, 92–97,
103, 108, 144–46, 231
See also Belief; Intentionality; Judgment;
Perception
233
Comedy, 32, 44–45, 48–51, 53–54, 66n, 77,
79–81, 84n, 88, 127, 154–55
Compassion, 16–17, 19, 22–27, 65n, 66–67,
71, 73
Courage, 68, 114–15, 120, 123, 129, 134
Curiosity, 99, 101, 110, 119, 121–22
Damasio, Antonio, 109, 199, 200n, 205–6,
208–9, 212
Da Ponte, Lorenzo, 31, 34–37, 41n, 42n, 43,
44n, 53
Darwin, Charles, 8, 106, 129, 228n
Death, 55–58, 159–78
Descartes, René, 132, 208, 218n, 222–23
Dispositions, 15, 68, 128, 200n
as virtues and vices, 17–23, 26–27, 68,
114–15, 137, 191n
Dixon, Thomas, 215–19
Doka, Kenneth, 163–65
Douglas, Mary, 167
Ekman, Paul, 8–11
Embarrassment, 65, 69, 136, 166
Emotion
of animals. See Animals
authenticity of, 141–46
contagion of. See Sympathy
control of, 87, 136–37, 142, 162, 224
expression, of, 8, 106, 143, 162–63, 165
false, 141–46
function of, 94–97, 118, 162, 175
object of. See Intentionality
responsibility for, 6–8, 142, 162, 183
subjective theory of, 4–7, 10
Empathy, 32, 62, 66–69, 82, 184, 186
Evaluation. See Judgment
Fear, 5, 15, 20–22, 25–26, 51, 56, 59–61,
63–64, 66–73, 81, 87–88, 89–90, 94,
100, 113–15, 142–45, 167, 184, 201,
223, 232
Feeling. See Affectivity; Sensation
234 Index

Foot, Philippa, 2–3 Mill, John Stuart, 10, 48, 51, 99, 115,
Forgiveness, 17, 27, 192 134–35
Freud, Sigmund, 96, 129, 135, 169–70, Miller, Jonathan, 94–96
183–86 Miller, William, 130–31, 147
Friendship, 26, 37–38, 46, 154–55, 157–58, Mistake, 5, 60, 64–65, 72–74, 141, 145–46
160, 172 Montaigne, Michel de, 196
Frijda, Nico, 109, 111–13, 118 Morality, 23, 35–36, 80, 115, 136–37,
139–40, 150, 161–69, 180, 182–84
Gender, 32–38, 40–46, 49–51, 52, 55 Morreall, John, 90–91
Goff man, Erving, 130, 196 Morton, Adam, 122–23
Grief, 35, 56, 58, 63, 90, 159–78, 182, 188 Motivation, 22, 60, 63, 66, 69–70, 89–90, 143,
Guilt, 21, 24, 63, 179–88, 193, 196–97 220, 222, 227–30
Mozart, Wolfgang Amadeus, 31–33, 35–36, 39,
Hampshire, Stuart, 128–30 40n, 43–46, 48–49, 51–53, 55, 155–56
Hanslick, Eduard, 57
Happiness, 79, 84, 88, 96, 132, 134, 142, 167, Nagel, Thomas, 137, 181–82
183 Neu, Jerome, 6
and social life, 29–31, 39–40, 51–52, 54 Neuroscience. See Brain
and virtue, 24, 26, 100–1 Nietzsche, Friedrich, 129, 183, 198
Herder, Johann Gottfried, 46, 49–51 Nussbaum, Charles, 33, 53–55
Higgins, Kathleen, 33, 53n, 54n Nussbaum, Martha, 15–16, 64n, 65n, 66n,
Hobbes, Thomas, 77–79, 132 67n, 70–71, 77
Hume, David, 19, 25, 99, 156, 198–200, 202n,
217–19, 225, 228 Obligation. See Morality
Humor, 32, 44–46, 53–54, 66n, 76, 79–83,
85, 87, 89, 95–98 Pain
Hunt, Lynn, 45 bodily, 81–82, 108, 114, 131, 201, 207–8,
Hutcheson, Francis, 77, 79 221–22
emotional, 46, 56, 59, 67, 103, 114, 168,
Identity, 36–37, 125–27, 130–32, 138, 183, 192, 196
172–73, 176 Parfit, Derek, 198–209
personal, 198–209 Parkes, Colin, 161n, 165–68
social, 160–61 Pennebaker, James, 174
Imagination, ch. 3 passim, 82, 89, 141–42, Perception, 15, 21–23, 26, 55, 87–88, 91–92,
174–77 103–4, 112, 227, 229
Instinct, 110, 133, 219–20, 222, 228–30 See also Sensation
Intentionality, 2–3, 22–23, 55, 63–64, 72–74, Phenomenology, 1, 4, 10
87, 89, 91, 146, 209 Physiology, 2, 8, 14, 88, 92–94, 103, 142–43,
Interest. See Valuing 199–200, 209, 227, 229
See also Sensation
James, William, 4, 10, 92–93, 99–100, Plato, 6, 16, 76, 96, 99, 106, 129, 135, 141,
103–12, 142, 199, 209, 213, 218, 225–32 216, 218, 229, 232
Joy. See Happiness Pleasure, 2, 38, 108, 220–22
Judgment, 4–7, 9–11, 14–15, 23, 55, 57, 64–65, aesthetic, 86, 90, 94, 106, 108
72–74, 91–92, 94, 112, 153–54, 209 intellectual, 83–86, 90, 94, 96–97, 99,
Justice, 14, 16–27, 50–51, 54, 67n, 94, 108, 101–3, 106, 108, 215
139–40, 157, 206 Pride, 3, 37, 77, 91, 123, 146, 219, 222, 225
Priestly, Joseph, 218
Kant, Immanuel, 48, 51, 136–40, 183–84, Projection, 3–6, 142
192, 199, 208
Kindness. See Benevolence Rawls, John, 158, 192n
Klein, Melanie, 186 Reflex, 25–26, 145
Regret, 63, 70, 157, 181–82, 232
Langer, Suzanne, 57 Reid, Thomas, 223–25
Laughter, 44–45, 66n, 74, 77–79, 88, 94, 96 Ribot, Théodule, 110–11, 118
Locke, John, 132, 217, 228 Rousseau, Jean-Jacques, 32, 46–49, 51, 126,
Love, 31–35, 38, 40n, 41–42, 46–49, 51–55, 56, 132
62, 68, 71, 90, 108, 149, 153–55, 157–58, Russell, Bertrand, 2, 4
159, 169–78, 185–86, 188, 222, 229 Ryle, Gilbert, 2–3, 126
MacIntyre, Alasdair, 173
McCrackin, Janet, 176–77 Sadness, 5, 67, 89–90, 94, 142
Memory, 87–88, 93, 169–70, 200–3, 206–8 See also Grief
 Index 235

Sartre, Jean Paul, 6, 125–26, 138–40, 142, Taylor, Charles, 137–38

145–46, 176 Trilling, Lionel, 127, 128n, 131, 136
Schopenhauer, Arthur, 57, 80
Scruton, Roger, 85, 91–92 Ulpian, 18
Seneca, 183 Universality, 8–10, 25–27, 56
Sensation, 26, 56–57, 88, 143, 209
of bodily change, 1, 4, 92, 103, 143, 199n, Valence, 162, 209
227, 229 Valuing, 5, 15, 19–22, 25, 55, 60–61, 70–71,
Sentiments, 8–10, 26–27 119–20, 123, 172
Shame, 5, 30, 34–37, 39, 41–42, 44–45, 48, subjective vs. objective, 6–10
52, 63, 69, 136, 180, 192, 196 Vengeance, 26–27, 29, 32, 34, 36, 38, 40n,
Smith, Adam, 25, 218 49, 222
Solomon, Robert, 3–12, 14–19, 21, 23–27, 32, Virtue. See Dispositions
59–61, 64–68, 70–72, 74, 82, 97, 142, Volition, 222–23, 224
147, 149, 155, 159–63, 165–66, 170, 172, See also Emotion, control of
173, 177, 179, 232
Spinoza, Baruch, 6, 128–29, 183 Walter, Tony, 174–76
Steinberg, Michael, 32–33, 35, 52, 54 Walton, Kendall, 73n, 142–45
Stoics, 5, 8, 16, 64n, 182–83, 191, 195 Williams, Bernard, 2–3, 18, 128, 156, 181–82,
Suffering, 24–25, 50, 74, 181 200–1, 206–7
Sundararajan, Louise. See Frijda, Nico Wittgenstein, Ludwig, 2
Sympathy, 59, 66, 68, 82, 88, 117 Wolterstorff, Nicholas 26