Clancy Martin - The Philosophy of Deception-Oxford University Press (2009)

The Philosophy of Deception
This page intentionally left blank

EDITED BY
Clancy Martin
1
2009
1
Oxford University Press, Inc., publishes works that further
Oxford University’s objective of excellence
in research, scholarship, and education.
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 Thailand Turkey Ukraine Vietnam
Copyright © 2009 by Oxford University Press, Inc.
Published by Oxford University Press, Inc.

198 Madison Avenue, New York, New York 10016
www.oup.com
Oxford is a registered trademark of Oxford University Press
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,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of Oxford University Press.
Library of Congress Cataloging-in-Publication Data

The philosophy of deception / edited by Clancy Martin.
p. cm.
ISBN 978-0-19-532793-9
1. Deception. 2. Self-deception. 3. Truthfulness and falsehood. I. Martin, Clancy W.
BJ1421.P45 2009
177'.3—dc22 2008038438
987654321
Printed in the United States of America
on acid-free paper
For Robert C. Solomon (1942–2007)
beloved teacher, mentor, and friend
Contents
Contributors, ix
Introduction: The Intersections of Deception

and Self-Deception, 3
Clancy Martin
PART I. THE PRACTICES OF DECEPTION

AND SELF-DECEPTION
1 Self, Deception, and Self-Deception in Philosophy, 15

Robert C. Solomon
2 On Truth, Lies, and Bullshit, 37

Harry Frankfurt
3 Deceit in War and Trade, 49

William Ian Miller
4 On the “Existential Positivity of Our Ability to Be Deceived”, 67

Mark A. Wrathall
5 Self-Deception, Deception, and the Way of the World, 82

David Sherman
6 Duplicity Makes the Man, Or, Can Animals Lie?, 104

Kelly Oliver
7 Lie Catching and Microexpressions, 118

Paul Ekman
PART II. TRUTH, LIES, AND SELF-DECEPTION: THE THEORY

AND THE ETHICS
8 Deception and Trust, 139

Alan Strudler
viii Contents
9 Lying, Deception, and Related Concepts, 153

Thomas L. Carson
10 Deception and the Nature of Truth, 188

Michael P. Lynch
11 The Truth about Kant on Lies, 201

James Edwin Mahon
12 On the Supposed Duty of Truthfulness: Kant on Lying

in Self-Defense, 225
David Sussman
13 User-Friendly Self-Deception: A Traveler’s Manual, 244

Amelie Rorty
14 Have I Unmasked Self-Deception or Am I Self-Deceived?, 260

Alfred R. Mele
Index, 277
Contributors
THOMAS L. CARSON is the author of The Status of Morality (1984),

Value and the Good Life (2000), and Lying and Deception: Theory and
Practice (forthcoming). His many articles have appeared in Philosophy
and Public Affairs, Nous, Philosophy and Phenomenological Research,
and other leading journals. He is professor of philosophy at Loyola
University in Chicago.
PAUL EKMAN is a pioneer in the study of emotions and their relation

to facial expressions. He is the author of Telling Lies (2009), Emotions
Revealed: Recognizing Faces and Feelings to Improve Communication and
Emotional Life (2003), and, with E. L. Rosenberg, What the Face Reveals
(1998), among other books. He is professor emeritus at the University of
California, San Francisco.
HARRY FRANKFURT is the author of the bestselling On Bullshit (2005)

as well as its companion, On Truth (2006), among many other books.
Widely recognized as one of the most influential philosophers of his
time, he is currently professor emeritus of philosophy at Princeton
University.
MICHAEL P. LYNCH is professor of philosophy at the University of

Connecticut. He is the author of Truth in Context (1998), True to Life
(2004), and Truth as One and Many (forthcoming).
JAMES EDWIN MAHON is associate professor and chair of the Philosophy

Department at Washington and Lee University. He is the author of the
entry “Lying” in the Encyclopedia of Philosophy (2nd ed.) and the entry
“The Definition of Lying and Deception” in the Stanford Encyclopedia of
Philosophy.
ALFRED R. MELE is William H. and Lucyle T. Werkmeister Professor of

Philosophy at Florida State University. He is the author of Irrationality
(1987), Springs of Action (1992), Autonomous Agents (1995), Self-
deception Unmasked (2001), Motivation and Agency (2003), Free Will and
Luck (2006), and Effective Intentions: The Power of Conscious Will (2009).
ix
x Contributors
WILLIAM IAN MILLER has research interests in medieval history, especially

the sagas of Iceland, social and political theory, emotions, and vices and
virtues. He has published Eye for an Eye (2006), Faking It (2003), The
Mystery of Courage (2000), and The Anatomy of Disgust (1997), among
other books. He is a Thomas G. Long Professor of Law at University of
Michigan Law School.
KELLY OLIVER is W. Alton Jones Chair of Philosophy and professor of

women’s studies at Vanderbilt University. She researches ethics, social
and political philosophy, feminism, and theories of oppression. Her
publications include Women as Weapons of War: Iraq, Sex and the Media
(2007), Subjectivity without Subjects (1998), and Witnessing: Beyond
Recognition (1997), among many other books.
AMELIE RORTY is visiting professor of philosophy at Boston University and

lecturer in the Department of Social Medicine at the Harvard Medical
School. She has written widely on the philosophy of mind and history of
moral psychology, especially on the emotions, akrasia, and self-deception.
Her current project is On the Other Hand: The Ethics of Ambivalence.
DAVID SHERMAN is associate professor of philosophy at the University

of Montana, Missoula. He is the author of Camus (2008) and Sartre
and Adorno: The Dialectics of Subjectivity (2007) and is co-editor of The
Blackwell Guide to Continental Philosophy (2003).
ROBERT C. SOLOMON (1942–2007) was professor of philosophy at the

University of Texas, Austin. His prolific career produced such works as
From Rationalism to Existentialism (1991), The Passions (1993), and Not
Passion’s Slave (2003). He was named to the Academy of Distinguished
Teachers and received the President’s Associates Teaching Award in 1985
and 1996 and the Chad Oliver Award in 1998, among other awards.
ALAN STRUDLER is professor of legal studies and coordinator of the

Ph.D. program in ethics and legal studies at the Wharton School of the
University of Pennsylvania. His current research is in deception and
in corporate governance. His articles have appeared in the Journal of
Philosophy, Business Ethics Quarterly, Philosophy and Public Affairs, and
other journals.
DAVID SUSSMAN is associate professor of philosophy at the University of

Illinois, Urbana-Champaign. His research interests include ethics and
moral psychology.
MARK A. WRATHALL is associate professor of philosophy at the

University of California, Riverside. He has published extensively on
phenomenology and existentialism.
Introduction
The Intersections of Deception and Self-Deception
Clancy Martin
Sometimes we tell a lie, and there’s no doubt about it. The classic exam-
ple is the lie a child tells when asked by his mother if he broke the cookie
jar. “I don’t know who did it,” the boy says, and he is lying: he knows full
well that the jar fell from the counter and shattered when he was reach-
ing hopefully down and in for the last cookie. Similarly, there are cases
of self-deception that are so straightforward no reasonable person can
deny that here, at least, someone is lying to himself. To take an example
suggested by Amelie Rorty in her chapter in this collection: your enor-
mously accomplished, MacArthur-winning full professor friend comes to
you and says, “They just denied my husband tenure, but the fact is his
research is so much better than my own.” Here we see that whatever
her motives—as Rorty points out, and Alfred Mele would agree, those
motives may be considerably more complex than they initially appear—
your friend is simply lying to herself. The boy in the first example is lying,
and he knows it, he’s not in the least self-deceived about it; the woman in
the second example is thoroughly self-deceived, and she does not know it
(though we would say that at some level she must), and therefore she is
not lying when she reports her belief to you.
But most of the lies we tell, whether we are telling them to one another
or to ourselves, are not nearly so clear-cut. Lies and self-deceptions seem
to exist along a continuum, with cases like the extreme ones I just men-
tioned on either end, and in the middle the many cases where the lies we
tell others are inseparably mixed up with lies we tell ourselves. As Robert
C. Solomon says, “Deception and self-deception are mutually entangled
phenomena . . . to fool ourselves, we must either fool or exclude others;
and to successfully fool others, we best fool ourselves” (page 25, this vol-
ume). During the worst days of the second war in Iraq, the cover of the
European edition of the Economist showed President George W. Bush
walking arm in arm with British Prime Minister Tony Blair, with the title
“Sincere Deceivers.” The point, of course, was that it was neither the case
that Bush and Blair took themselves to be telling the truth during the
3
4 The Philosophy of Deception
run-up to the war and then found themselves to be mistaken (though this
was the official story both preferred), nor was it the case that they cyni-
cally lied in order to manipulate their countries into going to war. One
suspects that the truth was more complicated, more interesting, and more
frightening: what happened, we might hypothesize, is that both leaders
deceived themselves and deceived one another, believing their own lies
and the lies each was telling the other, for strategic reasons so that they
could achieve the goals they had in mind, and once they were involved
in this dizzying whirl of motivated deceptions and self-deceptions, they
suddenly found themselves in a position to sell their deceptions with real
sincerity to the public. The young Nietzsche put it nicely:
With all great deceivers there is a noteworthy occurrence to which they
owe their power. In the actual act of deception, with all its preparations,
its enthralling in voice, expression and gesture, in the midst of the scenery
designed to give it effect, they are overcome by belief in themselves Self-
deception has to exist if a grand effect is to be produced. For men believe in
the truth of that which is plainly strongly believed.1
The suspicion that motivated me in gathering together this unusual and

terrific collection of work from fourteen of the most interesting thinkers
working on deception today was this: that the study of lying and the study
of self-deception, which had been undertaken almost entirely indepen-
dently, could both benefit from a sustained examination of the many traits
they have in common, of the ways they work together, of similarities and
differences in their structure, their practice, their ethics. I first began to
entertain this idea when I was still a graduate student, writing my disser-
tation, “Nietzsche and Deception,” under Robert C. Solomon. And when
I read Harry Frankfurt’s remarks on love and lies in Shakespeare’s Sonnet
138, the need to explore the question of the connections between decep-
tion and self-deception became urgent for me. (Frankfurt’s remarks are
included here in a montage I have assembled with his permission and with
the help from his books On Bullshit and On Truth.) Several of the essays
here take on this question directly: for example, those by Robert C. Solo-
mon, Mark Wrathall, Amelie Rorty, William Ian Miller, and David Sher-
man. Other essays, such as James Edwin Mahon’s and David Sussman’s,
both about Kant on lying, or Mele’s on self-deception, are concerned more
or less exclusively with one or the other phenomena, but are nevertheless
very helpful when thinking about the larger problem of how deception
and self-deception intersect. Ekman’s essay, for example, while ostensibly
concerned only with the seemingly straightforward subject of why we are
poor at catching liars, both unravels and knots all sorts of different theoret-
ical strings of deception, lying, believing liars, and self-deception. More of
the essays focus on lying than on self-deception, but I am happy about that,
1. Friedrich Nietzsche, Human, All-Too-Human, trans. R. J. Hollingdale (New York:

Cambridge University Press, 1996), 53.
Introduction 5
since the literature on self-deception is much more complete than that

on lying, and the essays that deal explicitly with the connections between
lying and self-deception are very much concerned with how lying to oth-
ers influences our self-knowledge and selfhood. And, of course, all of the
essays here are very much concerned with the question of the nature and
the value of truthfulness and truth, and one—“Deception and the Nature
of Truth,” by Michael Lynch—examines the relationship between truth
and lie from the standpoint of the analytical epistemologist.
The book is dedicated to Robert C. Solomon, my friend and mentor,
who died suddenly last year. I begin the book with Bob’s essay, in honor
of Bob, and because he introduces many of the themes that are developed
in more detail in later contributions: the notion that deception and self-
deception are essential to self-maintenance; the suspicion that philoso-
phers place too high a price on the truth, and naively fail to recognize the
importance of false beliefs and even lies for human flourishing; the com-
plex nature of both deception and self-deception, and their importance
to communication; the observation that lies and self-deceptions are cru-
cial to our social interaction; the difficulties of defining lie and deception;
the suggestion that it may be the motives behind a deception (whether
of oneself or another), and not the deception itself that we find mor-
ally blameworthy (or even praiseworthy). All this said, Bob nevertheless
insists—as Nietzsche did before him, and as every writer in his collection
does in one way or another—that the truth is to be prized, and deceptions
are to be viewed as a kind of necessary evil.
In part I of this book, the essays tend to be concerned with what we might
call the how of deception: the ways in which deception, both of oneself
and others, is actually carried out in human life. This concern naturally
takes us into some quite technical theoretical territory: take the case of
Wrathall’s phenomenology of deception in sense-perception, for exam-
ple. But the larger emphasis here is generally on deception in our human
lives. The essays in part II tend to be more concerned with theoretical
philosophical debates in the literature on deception and self-deception.
Thomas Carson’s excellent and exhaustive analysis of the definition of
lying and other similar concepts, for example, covers all of the technical
philosophical terrain on that notoriously difficult project. Similarly, Mele’s
piece brings us completely up to date on the literature in self-deception.
I will now briefly canvass the individual chapters. As noted, Frankfurt’s
contribution is an assembly of work he has already published. It opens
with his claim that the harm that lies do us results from their interfer-
ence with our efforts to understand things as they truly are: “the real state
of affairs” (37). Lies thrust us into an imaginary world that we cannot
live in or rely on. Frankfurt goes on to discuss Kant’s and Montaigne’s
claim that lies undermine human society. Here Frankfurt argues, agree-
ing with Solomon’s position, that Kant and Montaigne have gone too far:
although it is true that lies can tear the social fabric apart, they can also
knit it together, and we are all quite familiar with the process of finding
our ways through the twists and turns of ordinary social deceptions. What
really bothers us about the lie, Frankfurt plausibly claims, is the harm a
lie does to us as individuals. We take lies personally, and it is the personal
betrayal that hurts.
Frankfurt develops his personal take on the harm of the lie with a
discussion of the poet Adrienne Rich. For Rich, the liar puts himself in a
place of terrible loneliness: by hiding his mind from others, he perilously
removes himself from human society. (This may also, of course, be a rea-
son to lie: the liar may enjoy, even cherish, the feeling of having a secret
place where only he may go, a place where he is wholly himself alone.)
This self-estrangement performed by the liar is part of what harms us
when we are told lies, Frankfurt argues: it is a kind of blow to us to be
pushed away. Still worse, he says, is the way a lie “leads one to feel a little
crazy” (40; here he is quoting Rich). That is, when we trust someone, we
deeply believe we can rely on what they say, and when we learn we have
been deceived our feeling of reality is interfered with. We learn that our
own ability to judge is unreliable, and we are not sure what we should
believe and what we should suspect.
In his marvelous, first-rate analysis of Shakespeare’s Sonnet 138,
where Frankfurt shows that it is possible to both know a lie is a lie and
yet believe it; to tell a lie to someone knowing that that person will
understand that it is a lie, and yet expect him or her to believe it; and
to understand that the lies being told are essential to a lovers’ intimacy
that cannot be created in any other way. Frankfurt’s contribution closes
with several excerpts from his now famous essay “On Bullshit,” outlin-
ing his definition of bullshit—using a story about Wittgenstein and Fania
Pascal—and explaining his claim that bullshit may be more dangerous
to the truth than lies.
William Ian Miller’s contribution is, as he says, “a genealogy on deceit
in war and trade” (50). He starts with deceit in Ovid and the Old Testa-
ment and works his way all the way up to the present day, considering
the deceptions of such famous tricksters as Odysseus, David, the Vikings,
Machiavelli, William the Conqueror, even Montaigne. He then consid-
ers the practices of some famous deceivers in our contemporary business
culture, such as Bernie Ebbers, Dennis Koslowski, and Kenneth Lay (and
their like). Miller concludes by asking: what is it in us that prefers the
trickster and deceiver to his dupe?
Mark Wrathall’s contribution is a phenomenological interpretation and
development of Nietzsche’s observation that “a perspectival, deceptive
character belongs to existence” (67). Wrathall uses Merleau-Ponty and
Heidegger to explore what it means to experience perceptual deception
and asks to what degree are we ourselves responsible for the deceptions
that occur in perceptual experience. For Wrathall, “the phenomenology
of deception . . . points us to the inherently meaningful structure of the
perceptual world” (77). Through studying how these deceptions occur,
Introduction 7
we realize how deeply we are mentally and creatively engaged with per-
ceptual phenomena.
David Sherman raises an attack against the usual ways of understand-
ing self-deception on the grounds that such ways operate with too thin
a notion of truth and true belief. He outlines a more adequate way of
understanding truth, deception, and self-deception through an account
of sociohistorical truth and sociohistorical deception that he develops out
of Kant and Hegel. Crucial to Sherman’s story is his division between first-
order belief formation and second-order belief formation: the space
between these two is, he argues, the space of self-deception. Like
Wrathall, Sherman uses Heidegger to bolster his account, and the essay
concludes with a strikingly original analysis of the relationship between
self-deception and Sartre’s notion of bad faith.
Kelly Oliver introduces the notions of psychoanalysis and the subcon-
scious to further complicate the problem of deception and self-deception.
She writes, “insofar as unconscious forces drive us beyond our control and
even beyond our knowledge, then we are all and always a bunch of liars”
(104). Oliver discusses the differences between humans and nonhuman
animals in the work of Lacan, posing the questions: To what degree is
humans’ capacity to lie a function of their capacity for speech? And is
there a difference between pretending (as nonhuman animals do) and
pretending to pretend (as we do)? Oliver follows Derrida in concluding
that at least one difference between human and nonhuman animals is
that “animals may be capable of deception, but man is the self-deceptive
animal” (114). (Interestingly, recent studies on bluffing suggest that there
may be an evolutionary advantage to self-deception, which many differ-
ent kinds of animals practice: see David Lingstone Smith’s book Why We
Lie: The Evolutionary Roots of Deception and the Unconscious Mind (2004)
for a variety of studies and examples). But Oliver and Derrida agree that
the distinction between pretense and lie that Lacan rests on is unsupport-
able: the answer to Oliver’s question is yes, nonhuman animals lie just as
well as we do.
Paul Ekman, who is probably the most famous expert on lying alive
today, considers two questions here: (1) Why are most people so lousy at
catching liars? (2) What are the micro expressions that give liars away, and
given that they are easy to learn, why do we have such trouble learning
them? An interesting part of Ekman’s piece, which begins with a discussion
of his definition of lying, is that he does not think one must speak in order
to lie, and he thinks one can lie without falsifying (I have to say that I think
he is right about this). Ekman then proceeds to canvass the research on
catching liars, which includes interesting information about why and when
people lie, and at what costs they are willing to lie. Among many other rea-
sons, Ekman argues that we are poor lie catchers (and poor liars) because
of our evolutionary history and our ancestral environment: Ekman thinks
that back in our early development as humans, lying was easily detected
and swiftly and severely punished. Another particularly delightful reason
is that “our parents teach us not to identify their lies” (126). And adding
weight to the central thesis of this volume, he argues that we are bad
at identifying liars because “we often want to be misled; we collude in
the lie unwittingly [or wittingly?] because we have a stake in not know-
ing the truth” (126). Ekman uses a famous exchange between Neville
Chamberlain and Adolf Hitler to illustrate the point.
As do the opening pages of Frankfurt’s piece, Alan Strudler’s essay
addresses the tough question of what makes lying wrong. Using Bernard
Williams’s idea that deception is wrong because (1) it involves a breach of
trust, and/or (Williams is unclear) (2) it is a manipulation of the dupe by
the deceiver, Strudler offers a brilliant analysis rich with terrific thought
experiments to argue that “not all manipulation in deception involves a
breach of trust, and that deception that involves a breach of trust may
involve a wrong that is distinguishable from that which occurs in other
deception (139). Strudler argues that deception is often a form of legit-
imate self-defense, and in those instances should be governed by those
norms. I consider this essay the best work I have ever read on deception and
trust, and I expect it to become a classic in the literature on deception.
Thomas Carson’s chapter is the most thorough canvass of the literature
on the definition of lying available today. As he says, he “offers defini-
tions of the concepts of lying and deception and explains the distinctions
between lying, deception, withholding information, ‘keeping someone
in the dark,’ bullshit, spin, and half-truths” (153). He shows that many
of the debates about the nature and morality of lying can be reduced
to confusions and disagreements about (sloppy) definitions of the dif-
ferent phenomena. He develops his own definition of lying: “deception
is intentionally causing someone to have false beliefs” (153). One of the
pleasures of Carson’s article, like Strudler’s, is working through his many
ingenious cases and thought experiments to see how subtle and precise
this definition is (I still disagree with it, as does Ekman, but that’s as may
be). Carson also does a terrific job of analyzing and refining Frankfurt’s
notion of bullshit, and adds yet another player to the philosophical game
here with his carefully defined notion of “spin.”
In his essay, Michael Lynch argues that thinking about deception can
teach us a great deal about the nature of value and truth, what we can
expect from a theory of truth, and why some truth theories are doomed
to inadequacy. He opens his piece with his own account of the nature
of lying and how it should be distinguished from deception (as an aside,
I think it shows how rich and difficult the problem of defining “lie” actu-
ally is that there are so many different and well-defended definitions of
“lie” within this single volume). He then reviews several ways in which
the connections between deception and truth show us what a workable
theory of truth would have to look like. Finally, Lynch offers a concise and
persuasive argument that there is something intrinsically valuable about
the truth, even if that intrinsic value is “merely” psychological. All things
being equal, we simply prefer the truth.
Introduction 9
The essays by James Edwin Mahon and David Sussman deal with
Kant’s notorious claim that when we lie we are always acting immorally,
whatever the circumstances may be. Mahon points out that the hysteria
of philosophers over this claim of Kant is not really warranted, given that
“there are three senses of a lie to be found in Kant’s moral philosophy [the
ethical, the juristic, and the sense of right], and three corresponding duties
not to lie” (202). Mahon also adds to the debate about the definition of
lying, pointing out that for Kant a lie is “the making of an untruthful
statement with the intention that it be believed to be true” (203). There
are thus many cases, Mahon points out, in which Kant would not take the
statement or deceptive action under consideration to be a lie (the cate-
gory “lie,” for Kant, is quite narrow). So, for example, the many formalized
lies of etiquette and politeness—such as writing “Your humble servant” at
the end of a letter—are, though untruths, not lies, since they are clearly
not made with the intention that the untruthful statement be believed to
be true. Juristic lies are a still narrower category than (un)ethical lies. By
the end of his essay, Mahon convinces the reader that Kant’s prohibition
on lying is not nearly so outrageous, sweeping, and difficult to defend as it
has almost always been taken to be.
Sussman begins his essay by admitting how strongly Kant does seem to
denounce lying and, indeed, self-deception. “Kant approvingly notes that
scripture presents evil coming into the world not through the first murder,
but through the first lie (MM 6:431), and considers our endemic propen-
sity to hypocrisy and self-deception to be at the heart of the ‘radical evil
in human nature’ that is the fundamental source of all moral corruption”
(225). Sussman further elaborates Kant’s attacks on lying, including his
famous claim that truthfulness is an unconditional duty, and he then com-
plains (you can hear the charming tones of Sussman’s teacher Christine
Korsgaard in this remark): “These outrageous conclusions are a gift to
Kant’s enemies and a calamity for his friends” (226). Sussman goes on to
argue that although the conclusions of Kant’s ‘A Supposed Right to Lie’
are “wildly implausible, they do have substantial motivation within Kant’s
practical philosophy” (230). Sussman then elaborates Kant’s discussion of
who has and who does not have a right to the truth and introduces a notion
of “defensive lie” quite similar to that of Strudler. But Kant is not willing to
endorse defensive lying, because, Sussman argues, for Kant, defensive lies
presuppose a principle at odds with the “quasi-contractual commitments”
that are the “necessary preconditions of any social order” (239). Sussman
concludes that Kant can justify a lie to the would-be murderer at the door
in at least some circumstances—if the murderer is putting one under pres-
sure to respond, for example—but that Kant’s prohibition against lying
remains much stronger than many of us might like.
The last two essays are on self-deception. Amelie Rorty presents a
summary of her many thoughtful, persuasive, and articulate defenses
of the practice of self-deception, and reviews forms of self-deception
about which we should be ambivalent and wary. Although “many
varieties of self-deception are ineradicable and useful” (244), it is not good

all the time; and Rorty surveys the field of the many and various forms
of self-deception, good and bad. She gives us a long and helpful list of
what self-deception is not. Like Solomon, she also shows how philoso-
phers have (self-deceptively?) construed and misconstrued both the
theory and the practice of self-deception to suit their own interests by
insisting on the priority of the truth. While Rorty is known as one of our
great analysts of self-deception, she insists on the importance of deception
to understanding the phenomenon of self-deception: “Like deception, self-
deception is a species of rhetorical persuasion; and like all forms of
persuasion, it involves a complex, dynamic, and cooperative process. Suc-
cessful deceivers are acute rhetoricians, astute seducers who know how to
co-opt the psychology of their subjects Deception and self-deception
are not merely detached conclusions of invalid arguments: they are inter-
active processes with a complex cognitive and affective aetiology” (247).
But she is at her very best when she is defending self-deception, and
that is my favorite part of her essay, as she tears down, as thoroughly
as one could want, the Bernard Williams–style argument that the truth
is valuable because it is socially useful. She argues that it is not just
the self-deceiver who benefits from her own self-deceptions but all
those she lives with as well. In fact, she argues, we all encourage self-
deception in one another, because we know we need it for ourselves
and from the people around us. “It is virtually impossible to imagine
any society that does not systematically and actively promote the self-
deception of its members” (252). Robert C. Solomon agrees, making the
same point from the perspective of interpersonal deception: one “might
well hypothesize that deception, not truth, is the cement of civilization,
a cement that does not so much hold us together as it safely separates us
and our thoughts. We cannot imagine social intercourse without opac-
ity” (21).
In his essay, entitled “Have I Unmasked Self-Deception or am I Self-
Deceived?” (a reference to his well-known 2001 book Self-Deception
Unmasked ), Alfred R. Mele separates the problem of self-deception into
two component questions, one explanatory (“How does self-deception
happen?”) and the other conceptual (“What is self-deception?”). He
admits that he finds the explanatory question more interesting. (I think
he’s exactly right: In fact one can go further, as he does in his book, and
argue that the conceptual question is slightly confused and the explana-
tory one alone actually solves the so-called problem or paradox of self-
deception.) However, Mele says, because his critics worry that he has not
spent enough time addressing the conceptual question, he explains here
what motivates it and how it should be answered.
The key to Mele’s account of self-deception is what he calls his “defla-
tionary view.” Self-deception, he writes, does not entail “intentionally
deceiving oneself; intending (or trying) to deceive oneself; intending (or
trying) to make it easier for oneself to believe something; concurrently
Introduction 11
believing each of two explicitly contrary propositions” (261). This is

strong stuff—and the reader of this introduction will easily see that I have
a rather different view—but Mele defends these claims so well that one
is glad not to have to argue with him in a seminar room. There is helpful
discussion here of Mele’s now-famous notion of “twisted self-deception”
(first defended at length in his 1999 Philosophical Psychology article
“Twisted Self-Deception”): the phenomenon of the self-deceived person
believing something he or she wants to be false (think of Othello’s case,
for example). And Mele goes through several kinds of self-deception that
do not fit the “classic model” it (holding the true belief, not-p, and bringing
yourself to believe p). In fact, none of the authors in this collection hold
the “classic” view of self-deception, so perhaps now we should consider it
sufficiently debunked. Mele offers his own theory of self-deception, what
he calls the FTL theory (after J. Friedrich, Y. Trope, and N. Liberman, on
whose work he draws): “people enter self-deception in acquiring a belief
that ‘p if and only if p’ is false, and they acquire the belief in a suitably
biased way” (266). Or, in greater detail, here are his sufficient conditions
for self-deception:
S enters self-deception in acquiring a belief that p if:
1. The belief that p that S acquires is false,
2. S treats data relevant, or at least seemingly relevant, to the truth value of p
in a motivationally biased way,
3. This biased treatment is a nondeviant cause of S’s acquiring the belief that
p, and
4. The body of data possessed by S at the time provides greater warrant for
not-p than for p.” (267)
Mele then canvasses the opinions of various critics—Dana Nelkin,

Richard Holton, Dion Scott-Kakures, and Eric Funkhouser, along with
an imagined response from Amelie Rorty—who, he thinks, more or less
support his view but argue for further buttressing of it, which he thinks
is unnecessary or counterproductive. I close the book with Mele’s piece
because it is a first-class defense of his very influential view and because
although he thinks the analogy of self-deception with deception is an
unhelpful one, I think his essay shows us just the opposite: His under-
standing of self-deception can provide us with a more helpful analogy
with deception. Just as cases of simultaneously believing p and not-p are
rare and unhelpful in the discussion of self-deception, I think cases of
believing p and trying to persuade someone not-p are, relatively speaking,
rare and unhelpful when analyzing interpersonal deception. The interest-
ing cases, the complicated cases where the real philosophical work is to be
done, are in a domain in which beliefs are not as clear-cut as p and not-p.
When we lie, when we self-deceive, we distract, we confuse, we act, we
posture, we vacillate, we hesitate, we form and reform and distort and
assert and refute our beliefs and our attitudes. It is in this process, the way
the mind actually works, that we are human.

PART I
THE PRACTICES OF DECEPTION AND
SELF-DECEPTION
1
Self, Deception, and Self-Deception

in Philosophy
Robert C. Solomon
“I have done that,” says my memory. “I cannot have done that,”

says my pride, and remains inexorable. Eventually, memory
yields.
—Friedrich Nietzsche, Beyond Good and Evil
Nietzsche once asked, “Why must we have truth at any cost anyway?”1 It was
an odd question, coming from the philosopher who prided himself, above
all, on his brutal honesty, and it is an obscene question, in any case, for the
profession that sees itself as solely seeking the truth. Even those philosophers
who challenge the very idea of truth, not just Nietzsche and Nagarjuna but
Jacques Derrida and Richard Rorty, are scrupulous and unforgiving when it
comes to deception, misrepresentation and so-called creative misreadings, at
least of their own work.2 Philosophers in general insist on the truth even if
they do not believe in “the Truth.” They despise deception, and they ridicule
the self-deception of the “vulgar,” which it is their mission to undo.
Australian philosopher Tony Coady probably speaks for most philoso-
phers when he writes, “dishonesty has always been perceived in our culture,
and in all cultures but the most bizarre, as a central human vice. Moreover
the specific form of dishonesty known as lying has generally been scorned,
and the habitual liar treated with contempt. There are perfectly good rea-
sons for this.” But, he adds, “we should note that this perception is consistent
with a certain hesitancy about what constitutes a lie and with the more than
1. Nietzsche, Beyond Good and Evil, trans. W. Kaufmann (New York: Random House,1966).
2. Consider Nietzsche’s late lament “Has anyone understood me?” (Ecce Homo)
Consider too Derrida’s indignant response to widespread criticism of his work when he
was offered an honorary degree at Cambridge in the spring of 1992. “I have never written
any such thing!” he insisted to the press and against his critics, hardly indifferent to the
truth of the matter. This is not, of course, a refutation of the position (any more than
the perils of Pyrrhus constituted a refutation of ancient skepticism). But it is a pragmatic
paradox of considerable interest, what Bernd Magnus (following Fish) calls a
“self-consuming concept.”
15
16 The Practices of Deception and Self-Deception
sneaking suspicion that there might be a number of contexts in which lying

is actually justified.”3 Plato defended “the noble lie,” and the ultrarespectable
English ethicist Henry Sidgwick suggested that a “high-minded lie” in the
direction of humility might do us all a good deal of good.4
Philosophers have often fantasized whole cultures composed of liars, if
only as a possible counterexample to the categorical imperative or as a
source of delicious self-referential paradoxes. The neo-Marxist notion of
“false consciousness” and one common use of the word “myth” have rein-
forced the idea that a whole society could be in self-deception. But the pos-
sibility of such pervasive self-deception already presupposes some ideal and
independent criterion for the truth while at the same time giving consider-
able recognition to the legitimacy and the necessity of deception. In this
chapter, I want to further muddy these already treacherous waters without
denying what I take to be obvious, that in general—indeed, more than just
in general—we have to trust what people tell us and that lying, without
some further specification, is wrong.5 Furthermore, there is a great deal of
deception that is not so straightforward as lying, starting with misleading
gestures and the evident fact that people can be systematically misleading
in their overt behavior, suggesting in their actions what they might never
put into words. There are also half-truths, quarter truths, and “spin.” What-
ever the pronouncements of the philosophers, the case against deception
both in and out of philosophy is clouded, not only by questions about con-
sequences but by questions of culture and the intricacies of self-deception.
We could, of course, delimit the use of “deception” and especially the use
of “lying” to just those cases in which an untruth is knowingly and mali-
ciously told with the intention to deceive. In other words, we could eviscer-
ate this rich set of phenomena and eliminate a good deal of the complex
subject matter. In particular, it would eliminate what I shall suggest is a very
large proportion of cases in which deception and self-deception function
3. C. A. J. Coady, “The Morality of Lying,” in To Tell a Lie: Truth in Business and the
Professions (Sydney: St. James Ethics Center, 1992), 7–12.
4. Indeed, Sidgwick further suggested that philosophers might be well-instructed to
lie systematically to their readers. While he firmly believed in the truth of the doctrine of
utilitarianism, he also believed that public knowledge of that doctrine might have results
that would be disastrous. Accordingly, the promotion of the utilitarian doctrine in practice
required its systematic deception.
5. For example, one might deny that a person has a right to the truth in question. Do
Nazis have a right to know where their innocent victim is hiding? Does an eavesdropper
have the right to overhear only truths? Did Special Prosecutor Kenneth Starr have a right
to ask or hear the truth about his quarry’s private sex life? It is said that dishonesty is a
form of injustice, but in that case greater injustices may excuse or override the injustice of
a lie. One might even refuse to call unjust lies “lies.” This maneuver would presumably also
eliminate jokes and fictions as lies, at least where the audience does not expect to hear the
truth. One critical concern here, of course, is the case of “white lies,” often based on social
conventions. See C. A. J. Coady, Testimony (Oxford: Oxford University Press, 1992). Even
more interesting are those cases and cultures in which social convention is considered
mandatory and “trump,” even at the expense of what we would consider an outright lie.
Self, Deception, and Self-Deception in Philosophy 17
together and support one another. It would also eliminate all of those cases
in which cultural considerations clearly dictate deception in the name of
politeness, or appropriateness, or for the sake of face or feelings, whatever
the unvarnished and possibly rude truth may be. To define lying as wrong
or to limit “lying” to just cases of wrongful deception begs important ques-
tions, one of which is Nietzsche’s “Why must we have truth at any cost any-
way?” And as we already noted, Nietzsche is not alone. Plato and Sidgwick
both defend the high-minded lie, and one of the attitudes advocated for
novice monks, I am told, is the obvious falsehood “I am the worst person
in the world.” In Buddhism, the demand for truth and truthfulness seems
to include acquiescence in the face of such astounding precepts, all in the
name of their salutary affects. Nietzsche pursues a very different program,
of course, defending desirable untruths that are inspiring and conducive to
creativity, self-realization, and the “will to power.” But the point, it seems to
me, is the same: Truth is in the service of values, not the other way around.
“Dishonesty is a form of injustice, a vice,” Coady says, echoing the harsh
condemnations of Augustine and Kant. “It deforms the liar and debases the
currency of language.” But not all untruths are malicious, and not all decep-
tions are lies. The truth hurts, and sometimes it destroys. Lies can protect
and inspire, and deception can serve noble ends. Self-deception sustains
the illusions that sustain us, and though conducive to pathological dys-
function it is self-deception and not the truth alone that shall set us free.6
Indeed, in many if not most cases of self-deception (and deception too) the
question of truth can be a source of considerable consternation, not just
for the perplexing reasons long advocated by epistemological skeptics but
rather because of the self-fulfilling (and sometimes self-denying) features
of our beliefs about our selves and those aspects of the world that matter
most to us. A saintly man considers himself wicked. What is the truth of
the matter? A mass murderer with strong political beliefs, a “terrorist” in
the eyes of the press, considers herself a noble freedom fighter. Who is right
and who is wrong? A lover trusts and defends the beloved, no matter how
hideous the evidence to the contrary. Is this self-deception, or is it just—
love? The truth in such matters is rarely a matter of “the facts” alone.
DECEPTION IN PHILOSOPHY
Call me a truth-seeker, and I will be satisfied.

—Ludwig Wittgenstein (letter to his sister)
When it comes to the larger questions of philosophy—the meaning of life,

the nature of morality, the existence and personality of God and the teleology
6. Amelie Rorty, “Adaptivity and Self-Knowledge,” in Mind in Action (Boston: Beacon,

1990). “The Hidden Politics of Self-Deception,” in Self and Deception, ed. Roger Ames and
Wimal Dissanayake (Albany: State University of New York Press, 1996).
of nature, it is by no means clear what it means to seek “the truth.”7 Philo-

sophical doctrines seem to be more like professions of faith, perspectives
on reality, interpretations, conceptual sculptures, an art form, rather than
hypotheses or claims about truth as such. It was Hegel, following Plato, who
distinguished Philosophical Truth (“the Truth”) from the ordinary truths of
science and everyday life, defending the peculiar status of the former.8 And
yet, for over two thousand years, both East and West, the peculiar nature of
philosophical truth seems only rarely to have thrown into question the status
of truthfulness as a definitive moral and intellectual virtue in philosophy.
If this is a paradox, it is not a very interesting one, but it gives rise
to a fascinating if neglected set of questions about the self-aggrandizing
language of philosophy and the variety of deceptions and self-deceptions
among philosophers. There is no doubt, for example, that philosophers
have almost always deceived themselves if less often others about the
importance of philosophy, a fact made manifest only occasionally by some
iconoclast such as Nietzsche or Wittgenstein or a Zen master like Dogen.
On a more parochial level, philosophers generally deceive themselves
and try to deceive others about the superiority of this school or method
as opposed to that one, typically ripping one thread out of a fabric and
defending it alone as the whole truth. Personally, philosophers often
deceive themselves about their supposed love and pursuit of the truth—
not to mention wisdom—while ignoring their concern for their own rep-
utation and their status in the profession. Plato’s bully in The Republic,
Thrasymachus, has in fact remained as much of a presence in philosophy
as Socrates, though he is rarely recognized as who he is.9 Truth, like jus-
tice, remains in the hands of the strong (the tenured full professor, the
Distinguished Chair, the member of the more prestigious faculty).
Throughout the history of philosophy, deception has been assumed to
be a vice, honesty a virtue. Of course, one might tactfully suggest that the
very nature of the subject, namely, the articulation of profound truths,
requires such a commitment. If philosophers didn’t seek and tell the truth,
what would distinguish them from poets and myth-makers, apart from
their bad prose? Philosophers seek and tell the truth, the whole truth and
nothing but the truth. Or so they would have us believe. Diogenes strolled
the city looking for an honest man, not expecting to find another but never
doubting that he himself was one. He would not have fared much better,
we suspect, if he had toured the philosophers’ hall of fame. His predecessor
7. Wittgenstein’s statement in the epigraph is quoted in Ray Monk, Ludwig

Wittgenstein: The Duty of Genius (London: Cape, 1990), 14.
8. Or as Oxford philosopher J. L. Austin later put it, “in vino veritas, perhaps, but in a
sober symposium, verum.”
9. See Janice Moulton, “A Paradigm of Philosophy: The Adversary Method,”
in S. Hardng and M. Hinitikka, eds, In Discovering Reality: Feminist Perspectives on
Epistemology, Metaphysics, Methodology and Philosophy of Science, ed. S. Hardng and
M. Hinitikka (Dordrecht: D. Reidel, 1983), 149–164.
Socrates insisted that he was telling the truth when he claimed to know
nothing, an argumentative strategy that was doubly a lie. For many philoso-
phers and scientists too, we readily recognize that the search for truth may
be something of a cover, a noble facade for working out personal problems,
pleasing their parents, or pursuing personal ambition. Nietzsche suggested
that every great philosophy is “the personal confession of its author and a
kind of involuntary and unconscious memoir.”10 But unselfconscious self-
revelation is hardly the same as telling the truth, and when philosophers
such as Nietzsche go on to argue that there is, in fact, no truth, refusing
to tell the truth might become a kind of truthfulness and insisting on the
truth becomes a philosophically venal sort of lie.11
And yet, Socrates, we are told, died for the sake of his honesty. Epictetus
the early Stoic defended above all the principle “not to speak falsely.” In
more modern times, Immanuel Kant took the prohibition against lying
as his paradigm of a “categorical imperative,” the unconditioned moral
law.12 There could be no exceptions, not even to save the life of a friend.
Even Nietzsche took honesty to be one of his four “cardinal” virtues, and
Albert Camus praises his own invention, “the stranger” Meursault, as a
hero for the truth. Fellow “existentialist” Jean-Paul Sartre insisted that
deception is a vice, perhaps indeed the ultimate vice. 13 Sartre argued ada-
mantly on behalf of the “transparency” of consciousness, thus enabling
him to argue (against Freud) that all deception is in some sense will-
ful and therefore blameworthy. Today one reads American ethicists,
e.g. Edmund Pincoffs, who insists that dishonesty is so grievous a vice that
its merits cannot even be intelligibly deliberated.14 In this, unlike many
other matters, philosophy and common sense seem to be in agreement.
And whether philosophy merely follows and reports on the Zeitgeist or
actually has some hand in directing it, it would be safe to say that the
philosophical championing of honesty is an accurate reflection of popular
morality. Lying, for philosophers and laymen alike, is wrong.15
But what does it mean to insist that lying is wrong, and how wrong is it
really? The blanket pronouncements of the philosophers typically conceal
10. Nietzsche, Beyond Good and Evil.

11. Nietzsche, “On Truth and Lie in the Extra-moral Sense” (1873), in The Portable
Nietzsche, ed. W. Kaufmann (New York: Viking, 1954).
12. Kant, Grounding of the Metaphysics of Morals, trans J. Ellington (Indianapolis:
Hackett, 1981).
13. Jean-Paul Sartre, “Bad Faith,” in Being and Nothingness, trans. H. Barnes (New York:
Philosophical Library, 1956).
14. Edmund Pincoffs, Quandaries and Virtues (Lawrence: Kansas University Press, 1986).
15. Nor should this be assumed to apply only to “advanced” and philosophical cultures.
The ancient inhabitants of the island of Maui used to throw the umbilical cords of their new-
born infants into the (then active) crater of the volcano Haleakale to assure that their
children would grow up to be honest. On my way to the East-West conference on self and
deception, I was particularly struck by the fact that, of all the virtues, honesty was singled
out as exemplary.
more than they reveal, and the best questions are left under the covers. Is a
lie told to embellish an otherwise tedious narrative just as wrong as a lie told
in order to cover up a misdeed and avoid punishment? Is a lie told in desper-
ation any less wrong than a calculated, merely convenient lie? Is a lie told out
of self-deception more or less wrong than a clear-headed, tactical lie? (Is the
former even a lie?) Are all lies wrong—is lying as such wrong?—or do some
lies serve an important function not only in protecting one another from
harm (especially emotional harm) but in developing and protecting one’s
own sense of individuality and privacy? One might better think of lying
as diplomatic, as fortification, as essential protection for a necessarily less
than candid self. Or, one could just think of honesty as merely one among
many of the virtues, not a fundamental virtue at all. Stephen Carter tells the
story of a man on his deathbed who tells his wife about a long affair, just to
“cleanse his conscience.” Carter rightly questions whether such honesty is in
any way admirable, or rather just an inconsiderate cruelty.
It is worth noting that Aristotle, in his catalog of moral virtues, lumped
“truthfulness” together with “friendliness” and “wit,” important traits to
choose in a friend or colleague, to be sure, but hardly the cornerstone
without which the entire edifice of morality would fall down. Moreover,
what Aristotle meant by “truthfulness” primarily concerned the telling of
one’s accomplishments, “neither more nor less”—in contemporary terms,
handing in an honest resumé.16 He did not seem at all concerned about
social lies, “white lies” or, for that matter, even political lies except insofar
as these contributed to injustice or corruption. 17 Critics have often chal-
lenged Kant’s analysis of honesty as a “perfect duty,” appealing to our
natural inclination to insist that it is far more important to save the life
of a friend than it is to tell the truth to the Nazis who are after him. But
if there is even one such case in which it is right to lie and honesty can
be overridden, then the “perfect” status of the duty not to lie is compro-
mised, and the question is opened to negotiation.
It is in the light of such dogmatic (“a priori”) condemnation too that we
can understand the perennial controversy surrounding the seemingly inno-
cent “white lie,” the lie that saves instead of causing harm. And, to say the
obvious (though it is often neglected by philosophers), lies can also enter-
tain, as theater and as fiction, and not only on the stage or on the page.
Indeed, lies can also be useful and fascinating in philosophy. Not only do
they provide promising “heuristic” goads to further thinking, they provide
some of the essential subject matter as well. How many dozens of professors
are now employed because some Cretan, millennia ago, supposedly declared
that “all Cretans are liars” and thus generated the most basic paradox in logic
and philosophy. (If he told the truth, then he was lying, but if he was lying,
16. Aristotle, Nicomachean Ethics, trans. T. Irwin (Indianapolis: Hackett, 1944). Clancy
Martin, “Nietzsche on Deception” (Ph.D. diss., University of Texas (2003), and elsewhere,
has argued this thesis at some length.
17. Aristotle, Politics. trans. B. Jowett in The Works of Aristotle, ed. R. McKeon
(New York, Random House, 1941).
then. ) Is there anything wrong with a lie when it causes no harm? And is
it always true that we should tell the truth “even when it hurts”?
Behind the blanket prohibition of lying we can discern the outlines of a
familiar but glorious philosophical metaphor, the truth as bright, plain and
simple, standing there as the Holy Grail of Rationality, while dishonesty, on
the other hand, is dark and devious, the ill-paved path to irrationality and
confusion. In revealing the truth, we think of consciousness as transparent
through and through; in deception we detect an opacity, an obstacle, a wall
within consciousness. The honest man and the true philosopher know all
and tell all (except in Socrates’ case, since he insists that he does not know
anything). Nevertheless, Socrates’ student Plato offers to lead us out of the
shadows and into the light, even at great peril. The philosopher illuminates
that which the liar and the layman leave in the dark, including his or her own
inner soul.18 Truth and light are good; deception and darkness are bad or evil,
leading not only to ignorance and harm but to the degradation of rationality,
the abuse of language and the corruption of the soul. But philosophy, one
begins to suspect, has overrated these metaphors of clarity and transparency.
The obvious truth is that our simplest social relationships could not exist
without the shadows and darkness provided by deception and lies.
In his novel The Idiot, Fyodor Dostoevski gave us a portrait of a man
who had all of the virtues, including perfect honesty. 19 He was, however,
an utter disaster for everyone he encountered. More recently, Albert
Camus presented us (in The Stranger) with an odd “anti-hero,” “a hero for
the truth” who was unwilling to lie.20 It is not surprising that he comes off
as something of a monster, inhuman, “with virtually no human qualities
at all” (as the prosecutor points out at his trial for murder). On a more
mundane and “real life” philosophical level, one cannot imagine getting
through an average budget meeting or a cocktail party speaking nothing
but the truth, the whole truth and nothing but the truth. If one wished to
be perverse, he or she might well hypothesize that deception, not truth,
is the cement of civilization, a cement that does not so much hold us
together as it safely separates us and our thoughts. We cannot imagine
social intercourse without opacity.
Steve Braude, a philosopher who works extensively in parapsychol-
ogy, illustrates the importance of deception with a simple experiment. He
asks his audience if anyone would take a pill (which he has supposedly
invented) that will allow them to read the minds of everyone within a
hundred-yard radius. Not surprisingly, no one accepts the offer. We can all
18. Plato, Republic, trans. G. M. A. Grube (Indianapolis: Hackett, 1974) bk. 7.

19. Fyodor Dostoevski, The Idiot, trans. Henry and Olga Carlisle (New York: New
American Library, 1969).
20. Albert Camus, The Stranger, trans. Stuart Gilbert (New York: Random House,
1946). Camus’s own commentary on his novel and his character Meursault (“a hero for
the truth”) was published over a decade later, in the preface to Germaine Greer’s 1955
edition. Camus’s judgment is compromised by the fact that Meursault does lie in the
novel, indeed, commits outright perjury, and his obliviousness to matter of morals make it
highly unlikely that he can be said to “refuse to lie.”
imagine the bored and restless thoughts flickering through a friend’s mind
as we describe our latest trauma or the adventure of the day, the distracted
and hardly flattering thoughts of our students as we reach the climax of
the lecture two minutes before the class bell rings, and the casual and not
at all romantic thoughts of a lover in a moment of intimacy. “What are you
thinking?” is an extremely dangerous and foolish question, inviting if not
usually requiring the tactical but flatly deceptive answer, “Oh, nothing.”
Philosophers have long whitewashed the threatening nature of the truth.
Many of them adopt, without irony, a secularized version of the religious
banner “the truth shall set you free.”21 But, against the philosophers, we all
know that sometimes the truth hurts and the harm is not redeemed, that
the truth is sometimes if not often unnecessary, that the truth complicates
social arrangements, undermines collective myths, destroys relationships,
and incites violence and vengeance. Deception is sometimes not a vice
but a social virtue, and systematic deception is an essential part of the
order of the (social) world. In many societies, social harmony is valued far
more than truthfulness as such, and to tell the other person what he or
she wants to hear rather than what one might actually feel or believe is
not only permitted but expected. In such circumstances, do we still want
to speak of “deception” at all? And could we not begin to see our own
enlightened emphasis on “seeking the truth at all costs” (as Ernst Jones
wrote admiringly of Sigmund Freud) as one more ethnocentric peculiar-
ity, another curious product of our strong sense of individualism and a
particularly unsociable conception of “the truth”?
DECEPTION, SELF-DECEPTION, AND THE SELF
The difficulty making such distinctions [between real and only appar-
ent truthfulness] is almost as great for liars as for their dupes, because
self-deception enters into such estimates to such an extraordinary degree.
Hypocrites half believe their own stories, and sentimentality makes fraud
take on the most innocuous tints.
—Sissela Bok, Lying
The true hypocrite is the one who ceases to perceive his deception, the one
who lies with sincerity.
—André Gide, Counterfeiters (1955)
It is too often assumed that deception is a peculiarly linguistic

activity, having to do with the assertion of false propositions, and that
self-deception is therefore a paradoxical if not impossible lie to oneself.22
21. It is perhaps not without intentional ambiguity that this originally religious
injunction (John 8:32) is engraved on the administration building of the University of
Texas at Austin.
22. This is the formulation discussed in the analytic literature, for example, in Brian
McLaughlin, “Self-deception and the Structure of the Self,” in Ames and Wimal
But, as I suggested earlier, I can effectively deceive someone by driving

off or walking in the wrong direction, without saying a word, and there
is good evidence that many animals systematically practice deception. 23
And it is wrong to suppose that self-deception is simply the application
of deception to oneself, not only because that view generates even more
severe paradoxes but because it represents a serious misunderstanding
of the phenomenon. If one explores the strange realm of self-directed
psychological attitudes (and various “self-”prefaced ascriptions of psy-
chological attitudes) what immediately becomes evident is that rarely
are self-ascription and ascription to others just different applications
the same psychological description. (Consider, just as a small sample, self-
love, self-pity, self-respect and self-loathing.) Indeed, if one were to think
of self-deception as deception about the self rather than deception
directed to the self, there would be less of a temptation to assimilate the
first-person cases to the third-person.24
What this also means is that the nature of the self is part of our inquiry
along with the various conventions and conceptions concerning decep-
tion. Insofar as the self is a social being and not merely a locus of self-
reference, the character of both the self and self-deception depend on
the character of the society or culture in question. Who and what we are
depends only in part on what we think of ourselves, and what we think
of ourselves is rarely free of the opinions of others and free of the ethi-
cal values of our society. We want to think well of ourselves, and so the
Dissanayake, Self and Deception, 31–52, and it is the point of departure even for those who
reject the paradox, e.g. Herbert Fingarette in his excellent little book Self-Deception (New
York: Humanities Press, 1969). It is also used by Jean-Paul Sartre at the beginning of his
famous discussion of “bad faith” (mauvaise foi) in Being and Nothingness, 86–116, but this is
unfortunate, as bad faith is clearly a much broader and richer concept than self-deception.
In fact, Sartre uses the paradox only to set up a show trial against Freud’s psychological
determinism and attack his notion of “the Unconscious.” The “knowing p and not-p”
paradox plays virtually no role in either his examples or his subsequent arguments. I would
suggest that the examples in the early (pt. 1) chapter on “Bad Faith” might better be
looked at in the light of the concepts of being-for-others and the all-important notions of
freedom and responsibility in pts. 3 and 4 respectively. Indeed, I would argue that Sartre’s
attack on Freud and his casual treatment of the paradoxes of self-deception are at most
secondary if not incidental to his overall aim, which is to attack what Kathleen Higgins
has called the “atmosphere” of irresponsibility that he perceived in Parisian society. (Of
course, as so often, the same charge has been turned in turn on Sartre himself, by Herbert
Lottman, in his detailed chronicals of the actual [lack of] involvement of Sartre and his
comrades in the Resistance and the War.)
23. E.g. Carolyn Ristau on broken wing displays by waterbirds: C. Ristau, “Aspects of
the Cognitive Ethology of an Injury-feigning Bird, the Piping Plover,” in Cognitive Ethology:
The Minds of Other Animals, ed. C. Ristau (Hillsdale, N.J.: Erlbaum, 1991). See also Robert
M. Sefarth and Dorothy L. Cheney, How Monkeys See the World (Chicago: University of
Chicago Press), on deceptive monkey shines among our fellow primates.
24. I owe this tentative suggestion to Annette Baier.
need to fool ourselves coupled with the strong temptation to deceive

others about ourselves is always with us. The various social conventions
that dictate the rules about lying and deception are the same conventions
that dictate the acceptable nature of one’s self. What gets praised as good
character and what gets condemned as deception are part of the same
system of valuations.
In so many discussions of deception and, paradoxically, self-deception,
it is simply assumed that in lying one is clear about the truth oneself
and then purposefully and directly misleads the other about its nature. 25
Lying, accordingly, is fully intentional and malicious, at least insofar as it
willfully deprives another of something extremely important, the truth.
But this presupposes a degree of autonomy, rationality and transparency
that just doesn’t hold up to scrutiny. There are, of course, cold-blooded,
self-interested lies, knowingly false answers to such direct questions as
“Where were you last night?” and “Who ate all the cookies?” But one
might consider the claim that such lies are the special case rather than
the rule, like cold-blooded murder-for-profit in the bloody complex of
accidental, negligent, desperate and passionate homicides. Our fascina-
tion with lying and deception will not be satisfied by the straightforward
cases. What we are after is a dynamic drama of truth and falsehood in the
complex social and emotional webs we weave, compared to which what
is often singled out as “the lie” tends to becme an ethical exception of
comparatively little interest.
Self-deception, like deception, is a dynamic social phenomenon, not
just an internal drama or a pathological condition. The “social” nature
of the phenomenon, however, is often less than obvious, but part of the
reason for this is that philosophers tend to think of self-deception as an
odd and even paradoxical version of deception, as a “lie to oneself,” not
involving other people in any way at all. Of course, the lie may well be
“about” other people—as in a lover’s self-deceptive vision of his or her
beloved, and other people may be affected by one’s self-deception, as they
themselves are deceived in turn. But a conception of self-deception that
begins with the idea that the dynamics of self-deception are individu-
ally self-contained will lose the essential connection between deception
and self-deception, namely, their shared role in our social and personal
relationships. So, too, it is important to get away from the static “know-
ing and not knowing” conception that characterizes many philosophical
studies of self-deception.26 As an integral part of an on-going relationship
both deception and self-deception are necessarily dynamic, unstable (or,
25. J. P. Sartre, Existentialism and Human Emotions (New York: Citadel Press), 1957;
Fingarette, Self-Deception.
26. Robert Audi, “Self-deception, Rationalization, and Reasons for Acting,” in
Perspectives on Self-Deception, ed. B. McLaughlin and A. Rorty (Berkeley: University of
California Press, 1988).
perhaps, what Sartre calls “metastable”)27 and a continuous effort of enor-

mous complexity.28
Deception and self-deception are mutually entangled phenomena.
Superficially, one involves other people, the other does not. But to treat
them as different versions of the same phenomenon in two very different
settings or to treat them as wholly different (as lying and lying to oneself,
respectively) is to miss the dynamic that motivates both of them. To fool
ourselves, we must either fool or exclude others; and to successfully fool
others, we best fool ourselves. Philosophical discussions of lying too often
take as the paradigm example the straightforwardly cynical, self-interested
lie and ignore the more common species of lying that includes self-deception
as well. Transparency to ourselves can be just as intolerable as transparency
to others and for just the same reason. The self, with its many flaws and fail-
ings (some of them only imagined), is too much in evidence.
The recognition of one’s own motives and the significance of one’s own
thoughts can be devastating to one’s self-image and sense of self. A good
part of the self is self-presentation and self-disclosure, but an aspect of
equal importance is the need to disguise, to hide, not to disclose, those fac-
ets of the self that are less than flattering, humiliating or simply irrelevant
to the social context or interpersonal project at hand. (Think about pos-
ing for a simple photograph.) To a certain extent, this is merely a matter
of attention, of editing, of selective self-presentation, but it is not just (or
even for the most part) in our own hands. The self is essentially a social
construct, and our sense of ourselves depends on other people, or what
Jean-Paul Sartre called (with more than a touch a paranoia) “our Being-
for-Others.”29 One can hide or refuse to disclose oneself to oneself in many
27. A term that Jean-Paul Sartre borrows from chemistry in Being and Nothingness, 99f.
Metastability has a tentative stability, an appearance of stability, but the slightest intrusion
or misstep brings about total disaster. Consider a waiter carrying an overly full tray of cups
of hot coffee. All goes smoothly until the first jiggle, and a single boiling hot drop touches
his bare skin. He flinches slightly, and ...
28. Ibid., 112ff.; see Marcia Baron, “What Is Wrong with Self-Deception?” in
McLaughlin and Rorty, Perspectives on Self-Deception.
29. Sartre, Being and Nothingness, pt. 3. Philosophers typically talk as if our sense
of personal identity were just an internal affair (of self-revelation or memory or the
transcendental unity of consciousness). Indeed, Sartre’s analysis of the self in Being and
Nothingness too readily appears to be what I will call an “internalist” account. (Even the
name “for-itself” has obvious Cartesian credentials.) But in our obligatory reading of Sartre
in the context of discussions of self-deception, I would like to urge that three textual
points be kept in mind: (1) that the famous, often-reprinted chapter on “Bad Faith” is
one (remarkably short) early chapter in an eight-hundred-page book, and the subject is
rarely mentioned again. (2) That Sartre insists the “being-for-others” is on an “ontological
par” with the other two modes of being, “being-for-itself” and “being-in-itself.” He is
not, therefore, a traditional Cartesian dualist and his concept of self is not an internalist
account. In fact, Sartre’s argument and his examples are quite at odds with the ontological
apparatus he has provided for us in that early chapter of his humongous book. The
examples, which have been rightly criticized as inadequate by Allen Wood, “Ideology,
ways, notably by ignoring or distracting oneself, but none of these ploys has
a ghost of a chance if others cannot be distracted or fooled—or at least put
off—as well. Deception and self-deception are intimately intertwined. We
fool ourselves in order to fool others, and we fool others in order to fool
ourselves. And to make it more complicated (as it should be), we do not
always know which is which, who is self and who is other.
Deception between persons is rarely so cynical that it does not involve
more than a trace of sincerity and belief, in most cases the belief that
even if this particular “fact” is false, the truth that the lie is protecting is
far more significant than the act of lying. Thus we have the lover who
lies to protect his love, or the scientist who fudges her results to “prove”
a hypothesis she just “knows” to be true. Sissela Bok rightly suggests that
there is a thin line at best separating the lie for the sake of the truth and
the lie that marks one a liar. Lying for the sake of the truth is a paradox
that already requires a considerable amount of self-deception. Deception
between persons is rarely if ever unmotivated, and even a mischievous
lie “for its own sake” (the familiar “shaving” of one’s age, for example) is
typically a cover-up for other lies, insecurities and distrust. Thus Samuel
Johnson wrote, of self-deceptive men who would be virtuous, “having
none to recall their attention to their lives, they rate themselves by the
goodness of their opinions, and forget how much more easily men may
show their virtue in their talk than in their actions.”
If we understand deception and self-deception as exemplifying an
essential aspect of self-consciousness rather than as a willful violation of
principle or antisocial act, we begin to lose that sense of blanket con-
demnation of “lying as wrong” and come to understand deception and
self-deception as part of the matrix of human relations, neither good nor
evil as such but inviting sympathy and understanding rather than blame.
Amelie Rorty recites the touching case of a talented young doctor who
refuses to recognize in herself all of the evident symptoms of cancer. Her
behavior and her training make it obvious that, in some sense, she does
know of her condition, but the explicit recognition would be devastating.
And so she pretends, to herself and others (though convincing no one but
herself). Is there anyone who would call such behavior blameworthy? 30
People tell lies not only to avoid punishment or to impress others but
because they need to define and protect themselves (their selves) and cope
False Consciousness, and Social Illusion,” in McLaughlin and Rorty, Perspectives on Self-
Deception, 133–271, involve “being-for-others” as well as the categories of “facticity” and
“transcendence” he explicitly employs there. Thus construed, they escape many of Wood’s
(and other traditional) objections. (3) In an earlier work, The Transcendence of the Ego,
Sartre insists that the self is not “in” consciousness but is “outside of us in the world, like
the consciousness of another.” His is therefore what I will call an “externalist” account of
the self, not a Cartesian account at all.
30. Amelie Rorty, The Deceptive Self: Liars, Layers and Lairs,” in Rorty and
McLaughlin, Perspectives on Self-Deception.
with difficult social situations. Within the limited realm of self-knowledge,

in particular, deception is almost always a matter of coping rather than a
celebration of falsehood as such. Indeed, what it means to be false to oneself
is a rather complex ethical problem; our knowledge of ourselves is not only
incomplete but undergoing continuous revision, often along the lines of ide-
als and ambitions that are themselves ill-conceived, inappropriate or merely
borrowed. It is within this continuing coauthorship of self and self-esteem
that both deception and self-deception must be appreciated, and even the
most cynical interpersonal intrigues are first of all shared productions of
the self, involving both conspiracy and vulnerability in more or less equal
measure. Consider, for example, the web of affections and deceptions in
Chodoros Laclos’s Liasons dangereuses, which deceptively presents itself to
us as an aristocratic game but soon reveals itself as a life-or-death theater of
mutual self-deception.31 And as in Liasons dangereuses, (whose author felt
it necessary to produce a lengthy preface morally denouncing and distanc-
ing himself from the psychology he so insightfully represented), what is too
often presented as a morality tale becomes a study in interpersonal psychol-
ogy and the mutual, surreptitious, social construction of the self. It is not as if
ethics is (or should be) absent from such a study, but our evaluations can no
longer be the Manichean mantra “Truth is good, deception is evil.”
Telling the truth can be a vice parading as a virtue. In the name of
integrity, one can use truth as a weapon and honesty as a strategy, as in the
Stephen Carter example above. Children and lovers often tell the truth
precisely in order to hurt and to humiliate, and such truth-telling can be
manipulative, even vicious. In Camus’s last novel, The Fall, an extremely
devious character named Clamence confesses to an acquaintance (the
reader, of course) the truth about his life, including first and foremost the
many lies he had always been telling himself.32 What becomes evident,
however, is that he is still deceiving himself by way of seducing the other,
and even his truths are only a ploy. What Clamence is after, we learn in the
last pages, is neither truth nor total disclosure but a subtle vengeance, and
his confession is a subversive expression of a deeply felt resentment. But
who is the victim, and who is the villain, in such tales of devious- ness?
Why do we think that victims and villains must be part of the structure of
deception? As often as not, deception and self-deception combine to form
the most sincere belief among coconspirators, not victim and villain.
Virtually every faith and religion is a large-scale example of such belief, but
so too is almost everyone’s self-image and every society’s sense
31. Chodoros Laclos, Liasons dangereuses (New York: Penguin, 1962).

32. Camus, The Fall, trans. Justin O'Brien (New York: Vintage, 1956). In many ways,
The Fall is the opposite of The Stranger. Meursault (the “stranger”) is the very portrait of
transparency, all experience and virtually no reflection or self-consiousness. Clamence, by
contrast, is all reflection and painful self-consciousness. One tells the truth because he is
too simple-minded to lie, the other because he wants to seduce his listeners. In what sense
is either of them “not lying”?
of itself, including the scientific and philosophical communities as well as

every ethnic group or culture. Nietzsche and later Jung wrote extensively
on our need for myths and warned against an age that would try to do
without them. But what is a myth if not an elaborate self-defining collec-
tive self-deception, and if all such deceptions are wrong then would there
be any truth that is ultimately worth defending?
IT TAKES TWO TO TANGLE: THE LIE AS A HOLISTIC

PHENOMENON
It takes two to lie, one to tell and one to listen.

—Homer Simpson
If deception and self-deception are to be understood first of all as inter-

related dynamic interpersonal and social phenomena, then it is a mistake
to try to understand them in terms of one or another artificially isolated
aspect of the relationship. For example, in most utilitarian discussions of
lying, much of the focus has been on the alleged victim, the person who is
misled or betrayed by the lie. The calculation thus tends to trace out the
obvious and not so obvious effects of even the “whitest” lie, its ability to
undermine trust and render the victim helpless when the truth might well
have allowed some significant action. Sissela Bok, for example, pursues such
a quest in wonderful detail, tracing the consequences of professional lies,
political lies, loving lies, paternalistic lies, therapeutic lies, experimental lies,
etc.33 Bok also examines at length the complications of authoritarian decep-
tion and the manufacture of excuses, including the notorious slippery slope
argument from the very plausible claim that “the whole truth” is impossible
to tell down to the insidious thesis that the truth is not necessary.
That is the challenge and the fun of “thick” philosophical investigations
of lying; first we recognize the obvious immediate consequences: hurt
feelings, a tragically un- or ill-informed patient (client, friend, public).
Then the devastating penalties for an unsuccessful “cover-up” become evi-
dent. Finally, there are the more subtle implications of spreading distrust,
increased cynicism and consequent withdrawal, a corruption of language
and public discourse. What gets left out of many of those discussions of
deception, however, is the need to focus on the liar and not just the con-
sequences. For if deception and self-deception are so intimately involved,
then the assumption that the perpetrator of the lie is not also its victim
becomes less plausible. The lie is a matter of mutual engagement and not
just a malevolent act perpetrated by one person upon another. But also,
this carries us away from utilitarian analysis to something quite different.
When philosophers have fixed their gaze on the nature of the lie instead of
its consequences, however, they have often tended to deny the
33. Sissela Bok, Lying (New York: Random House, 1978).

interpersonal and social nature of deception. Kant in particular was ada-

mant about the logical inconsistency of the “maxim” of any and every lie,
established by the fact that one could not universalize the permissibility
of lying without undermining the very possibility of language (assuming,
that is, that the primary purpose of language centers around such activities
as describing true facts and making promises). (He also complained that
lying was using the other as a means, but it was still one’s own act that was
in question, not the role of the recipient.) Because lying is (by definition)
the intentional telling of a falsehood, some attention must be focused on
the liar who has and exercises that intention. But Kant quite explicitly
ignores the concrete motives and the character that lie behind the lie, pre-
ferring to emphasize the immorality of lying as such rather than under-
stand the psychological and social dynamics of the relationship. But even
as ethics, it is certainly not unimportant what motivates lying and what
kind of characters we are dealing with when we point our fingers at liars.
Is it immoral to lie to murderers? How about other liars?
Here is where “virtue ethics” gains its hold. It is a conscientiously
thick examination of not only this lie or falsehood but the context in
which it is told, the aims and intentions behind its telling, and, most
important, the personality and character of the prevaricator. But to
overemphasize the character of the liar is just as misleading as an iso-
lated emphasis on the lie or its consequences. Deception is, to employ
that overused and much abused word again, a holistic phenomenon, one
that embraces both the liar and the lied to as well as (often) everyone
else who may be within earshot or is affected by the lie. One cannot
break it up into pieces and expect to understand its vital organic unity.
One cannot try to understand or evaluate the lie, the liar, the victim,
the audience, the “stakeholders,” and the consequences and then put
these together in some “multidimensional” analysis which adds up to an
adequate understanding.
One of the most distinctive and most neglected features of lying is
that it is surprisingly hard to do. As anyone who has tried to protect even
a small casual lie can tell you, the amount of thought and care that is
required to keep in mind all of the logical implications and possible con-
tradictions (“If I was at Sam’s place, then I couldn’t have seen Thelma
at the Casino, but if I didn’t see Thelma then how could I have known
about the party at Shelby’s house?”). It is always easiest, the old adage
tells us (with considerable wisdom), to tell the truth. But next easiest is
to believe your own lie, to become so submerged in its network of details
and implications that the continuation of the lie—as Aristotle argued for
honesty—becomes but second nature, without further thought or delib-
eration. In either case, however, neither ease nor difficulty is a dependable
mark of morality, and one might (like a novelist or any other story teller)
delight in the intrigue and self-conscious tension that artful lying requires.
Part of the pathology of compulsive liars may well be the high-adrenalin
challenge of holding a number of lies together as a high-risk acrobat might
juggle a number of brightly lit torches or razor-sharp knives—along with

the neurotic need to cover up not just something but (by logical implica-
tion) almost everything. But all of this hiding and masquerading is aimed
at one end, to protect the self.
In virtue ethics (as in Kant’s deontology) there is always the temptation
to scissor off the liar from any particular lie or any particular audience,
but lying by its very nature is a public performance. It may also be part
of a rich pattern of self-deception, a way of getting other people to affirm
one’s own (false) opinions or, conversely, a way of “sincerely” relating to
other people. So, too, with more innocent and straightforwardly strategic
lies. Lying involves a complex logic that reaches across and cuts through
our various social relationships and sometimes with great difficulty weaves
a portrait of the self and its relations. And even in self-deception, it is the
inconsistencies in our stories discovered or discoverable by other people
that motivate our continued efforts at duplicity. After all, if self-deception
were a matter of mere internal consistency, would anyone but a logician
feel compelled to avoid inconsistency at all costs? Would “cognitive dis-
sonance” ever become an issue much less a motivational force if it did not
also become subject to the scrutiny of others?34
Imagine yourself on a inter-city bus or a short-hop plane ride next to a
somewhat tedious but inescapable fellow passenger who insists on know-
ing “what do you do?” One can readily imagine offering up the most banal
and boring answer as an alternative to an utterly offensive reply, or, alterna-
tively, one can with slightly more effort imagine constructing a fascinating
but wholly false account of one’s life as a K.G.B. double agent or a Texas
Ranger. In the first case, one might get a chance to get some reading or
sleeping done while in the second, there would seem to be no harm done
but rather a welcome entertainment for both of you during an otherwise
tedious voyage. There is, of course, the odd chance that one’s fellow passen-
ger may (contrary to all expectations) show up again at your hotel, wreak-
ing the sort of havoc that only comic movies can fully convey, and it is true,
no doubt, that every lie opens one up to possible complications of this sort.
But this is hardly a moral objection to lying, and in the absence of harm
such elaborate lies seem unobjectionable. (So, too, one could argue, for the
“big lies” that hold most cultures and religions together—myths of origin,
shared fantasies of moral right and manifest destiny, illusions of favored
status in the eyes of the divine, delusions of grandeur.)35 But here again
34. Leon Festinger, Cognitive Dissonance (Evanston, Ill.: Row Peterson, 1957).
35. In Fiji, before the arrival of the British, Viti Levu was considered, with some
reason, the center of the earth, its largest land mass, surrounded by ocean and a few
hundred modest islands. In the nineteenth century, when first confronted with a map of
the world, Fijians reacted first with predictable denial, then humiliation. It was then that
the story of the great African canoe Kannitow became established among the Fijians,
despite its dubious veracity. Our British commentator scoffs. But then thinking of the
standard stories of our own culture (Columbus’s discovery, the conquest of the West, the
beacon of democracy and freedom in the world), are our stories any more reasonable?
the attention should be on the social context and the relationship—in the
above case essentially transient—and not on the lie or the consequences of
the lie exclusively.
It is a mistake to think about and condemn deception and self-deception
sui generis. There are legitimate lies in literature, heuristics in science,
myth in religion and philosophy. These are not just isolated fictional
frames with at most metaphorical connections to considerations of self or
extremely tenuous “expressions” of ourselves and our relations with other
people. Quite the contrary, these are the “myths and metaphors we live
by,” according to many authors from the ancients to our contemporaries. 36
Once we give up the philosophical tendency to generalize about decep-
tion and self-deception in the abstract and focus instead on the particular
phenomena of lying, the intentions and motives behind them, the context
as well as the consequences and the interpersonal relationship between
the participants, it becomes increasingly obvious that most lies are not
merely lies but also self-deception and part of a larger matrix of beliefs
and emotions that define not only this relationship but a community or
a culture.
The lies of love (or pretended love) depend for their credibility on a
remarkable institution that defines and gives structure as well as elabo-
rate discourse to a seemingly “primitive” emotion.37 Consider how much
cultural apparatus goes into the simple but (when it is false) vicious lie
“I love you.”38 The individual passion can be more or less faked, but the
words take on a force of their own. This because that affirmation, and its
requisite response (“I love you too,” is part of the culture, one of those
programmed exchanges that once initiated, can only be followd by the
response “I love you, too,”—or result in a calamitous breakdown. But
apart from outright lies about love, how often are we confused, or filled
with doubts, and the saying of it provides a way of clarifying those doubts,
by way of a “leap” (not of faith, but of commitment of sorts)? Is that a lie?
Self-deception, like deception, is not always motivated by self-interest,
cold and calculating, but by our engagement in an emotionally charged
world in which things matter to us, in which the truth is by no means clear
and wishful thinking and the expectations of others weigh more on us
than that abstract metaconception known to us as “the Truth.”
Deception and self-deception are part and parcel of our engagements
in the world including, not least, in the development and maintenance of
our image and sense of ourselves. Deception is first of all a way of relating,
a not entirely accurate presentation of self to others and to oneself. There
is no single point where presentation becomes deception, where pretense
36. George Lakoff and Mark Johnson, Myths We Live By (Chicago: University of
Chicago Press, 1980).
37. Roland Barthes, A Lover’s Discourse (New York: Farrar Straus and Giroux, 1977).
38. Jeffrey Masson has a delightful book with the title Dogs Never Lie about Love. The
implication, of course, is that we often do.
becomes sincerity, where play becomes the real thing. Self-presentation is

always deception, but whether or not it counts as deception or whether
or not it is blameworthy depends on the context, the performance, the
expectations.39 Some deception is harmful and even immoral, but some of
it is neither. Indeed, an extremist might even argue that there is no such
phenomenon as lying as such, only various ways in which we relate to
one another as insecure social creatures surrounded and infiltrated by an
inevitably equivocal language. We are, perhaps, not only capable of lying
but virtually incapable of not doing so.40 Deception and self-deception,
according to such a kinky view, may not be perversions so much as they
are the very stuff of human intercourse.
ON BEING TWO-FACED: THE DUPLICITOUS SELF OF

SELF-DECEPTION
Hypocrisy is the homage that vice pays to virtue.

—La Rochefoucauld
It has often been pointed out, with various degrees of obfuscation in cur-
rent “Continental” philosophy, that the self is double. This can be put in a
number of ways, beginning with the obvious, that we see ourselves “from
the inside” but can also be seen (and sometimes see ourselves) “from the
outside,” from a second- or third-person (misleadingly called “objective”)
point of view. But the complex relations between these two (or three)
viewpoints gives rise to some deep suspicions and the seemingly outra-
geous charge that the self is intrinsically duplicitous, not only double but,
as we say, “two-faced.” In The Fall, Camus has his character Clamence
confess this ambiguity, which he goes on to demonstrate in his progressive
seduction of his almost mute interlocutor and the reader.41 The face pre-
sented to the world is at odds with the face that smiles knowingly inward.
39. This is the point where proposition-minded philosophers too readily isolate
false verbal self-ascriptions of such presentations and point to them as the paradigm of
deception—lying, in effect, about who one is. (Aristotle treats “truthfulness” this way,
Nichomachean Ethics 4.9.) But a more typical and more interesting case is the person
with some but still quite limited knowledge who acts as if he or she is an expert without
ever claiming to be so. As casual conversation and social self-presentation, this is innocent
enough. But in a context in which a real expert is or becomes urgently needed, it becomes
not only deception but fraud and betrayal. So, too, it is not deception to “act friendly” with
a person for whom one has no special affection. Nor is it even obvious that proclaiming
one’s friendship in such circumstances, as encouragement or a friendly gesture, for
example, is deception or in any way blameworthy. It is only when the expectations of the
so-called friend (or in rare cases, other people) are such that the true devotion of a friend
is called for and not forthcoming that, in retrospect, the entire performance gets indicted.
40. Nietzsche, “Truth and Lie in the Extra-moral Sense”; Jean Baudrillard, Selected
Writings (Palo Alto: Stanford University Press, 1988).
41. Camus, The Fall.
But, then again, perhaps the face that smiles inward is itself an illusion,
only a reflection of the looks it absorbs from others.42
With this in mind, let me distinguish between two different models
of self-deception, which in turn depend on two quite different models of
the self. In dull and often ambiguous analytic jargon, the two might be
referred to as the internalist and the externalist models of self-deception,
and they are based respectively on similarly internalist and external-
ist accounts of the self. An internalist views self-deception as a relation
between a person and a set of beliefs (broadly construed, which may or
may not involve the contradictory “believing p and not-p” paradox that
preoccupies so much of the analytic literature). Accordingly, internalist
models of self-deception tend to be concerned with the internal structure
or architecture of the self. Thus Freud divides the self into consciousness
and an unconscious, a troubled house with an inaccessible basement, and
later into “agencies.” Amelie Rorty ingeniously suggests that the self is
something like a medieval city, a virtual labyrinth of pathways and neigh-
borhoods functioning independently and without knowledge of the oth-
ers.43 Less architectural theorists look for ways of gerrymandering belief
or bypassing belief altogether, by way of “avowals” or other acknowl-
edgments.44 The problem for the internalist, perhaps but not necessarily
made more intractable by paradox, is to understand how one can in some
sense “know and not seem to know.” Self-deception, in other words, is
first of all an epistemological problem.
The externalist, on the other hand, sees self-deception as a social
phenomenon. It has to do not so much with a person and his or her
beliefs as it has to do with a person and his or her roles and relationships.
Self-deception is a consequence of wanting to be thought of and treated
in certain ways and not others by other people. One’s self-conceptions
are the product and not the source of the opinions of others, and self-
deception is thus an attempt to manipulate those opinions and not just
one’s own. I think the idea that we fool ourselves in order to fool others
more often than not gets it backwards, and the idea that in self-deception
we first of all lie to ourselves is just plain wrong. Self-deception begins
and continues by playing a part, by acting the good husband, the respon-
sible citizen, the competent, healthy professional. This may or may not
be accompanied by self-avowal. It may or may not be accompanied by
rationalization or denial. Self-deception is thus first of all a performance,
if sometimes a performance enjoyed only by oneself. Thus Sartre argues
that being-for others is an essential ingredient of the self, even when oth-
ers are not around.45
42. Suggestions of this sort abound in the wonderful work of Erving Goffman, notably
The Presentation of Self in Everyday Life (New York: Doubleday, 1959).
43. Rorty, “Deceptive Self.”
44. Fingarette, Self-Deception.
45. Sartre, Being and Nothingness, pt. 3, ch. 1, esp. sec. 4, “The Look.”
It is also within the perspective of being-for others that we can under-

stand why Sartre comes down so hard on “sincerity,” which is only a “dis-
play” of the truth, a show, another self-presentation, a mode of deception
and not genuine self-reflection.46 Here, too, we can understand why Roger
Ames claims that the Chinese have no conception of self-deception.47
He is appealing to the clearly externalist model of the Chinese self and
with it an externalist account of self-deception. But, then, why should we
assume that the internalist model so accurately portrays the European
self, or that the externalist view is all that foreign to us? On the externalist
model, self-deception is contextual, and the context is paradigmatically a
social context. But isn’t this ultimately true of our “Western” conception
of self and self-deception? At one end of this view, there is no self without
others.48 Or as St. Exupery writes, “Man is but a network of relationships,
and these alone matter to him.”49 According to this externalist model, one
might say, only half ironically, that the “inner self” is social.
I want to endorse a version of this view. What does it mean, however,
to say that the self is social? I mentioned earlier that it is one thing to
claim that the self is socially construed, another to insist that it is socially
constructed, still another to maintain that it is socially constructed as a
social self, three claims often conflated. To say that the self is socially con-
strued is by far the weakest of the three, the one understandably adopted
by most social scientists, and it leaves entirely open to what extent the
nature of the self is determined, for example, by biological and noncogni-
tive social forces and attitudes. To have a construal of the self is to adopt
a perspective, to look at it a certain way, but this has minimal ontological
commitments. It might be, for example, that as a teacher I construe the
selves of my students as “student selves,” that is, defined in terms of their
behavior, preparation for and performance in class. I do not pretend that
they do not have other, most likely more personal and pressing concep-
tions of self themselves, as I find out when I meet them in the local pub.
To claim that different cultures construe the self in different ways is to
make a minimal claim, one that is just as true of various subcultures in
our own culture. Self-deception, so understood, would consist in part of
adopting an inappropriate construal of self in certain contexts.
To say that the self is socially constructed, by contrast, is to make a
much more radical claim. It is to insist that there is no self apart from its
construction in particular social contexts (although one might distinguish
46. Sartre, Being and Nothingness, pt. 3, ch. 1, esp. sec. 4, “The Look.” 110–11.
47. Roger Ames, “The Classical Chinese Self and Hypocrisy,” in Ames and Dissanayake,
Self and Deception, 219–40.
48. E.g. Hegel, in the Phenomenology, “Lordship and Bondage,” P. F. Strawson
in Individuals, and grudgingly even Sartre, in “The Reef of Solipsism,” in Being and
Nothingness, 303 ff.
49. Pilote de Guerre, quoted in Merleau-Ponty, The Phenomenology of Perception
(London: Routledge and Kegan-Paul, 1962).
here between a very general claim that one will have no concept of self
at all unless he or she grows up in some society and the more particular-
ist claim that a person learns to cultivate a particular self—a Texan or
a Chuangzou or Maori self, for example—only by growing up in Texas
or Chuangzou or Tanderoa [New Zealand]).50 To insist that the self is
constructed, however, is not yet to claim that the construction is of any
particular typology, in particular, it does not mean that the self so con-
structed is a social self. The American “rugged” individualist is a socially
constructed self, as is the infamously antisocial Ik.51
To say that the self is socially constructed as a social self is to make a
very particular kid of claim, which must be distinguished from the other
two. The Chinese self is socially constructed as a social self, an interde-
pendent self, but, again, we need to make at least one further problematic
distinction. The social construction of self proceeds in part by way of
a language, a language which includes a certain vocabulary of selfhood.
In this language we learn to talk about ourselves in certain ways. A self-
consciously individualist culture will naturally emphasize the importance
of such notions as individuality, autonomy and independence. A self-
consciously communitarian culture will naturally emphasize the impor-
tance of such notions as community, loyalty, duty, and kinship. But how
we talk about ourselves is only a partial indication of how we think about
ourselves and how we actually behave. The social construction of self also
proceeds without language, without self-description, by way of ten thou-
sand non-verbal cues and examples, everything from a mother’s refusal to
leave her infant alone even for a minute and the fact that people tend not
to look each other in the eyes to the waging of war and the celebration of
religious rituals. It is perfectly possible, therefore, for a culture to cultivate
a way of talking about themselves and the self that is somewhat at odds
with the ways in which they actually conceive of themselves and their
relationships to one another. Indeed, in times of social tension or disloca-
tion or in order to distinguish oneself from an alien culture, such diver-
gence of self-conception from practice may be extremely common.52
The resultant complexity of the self—as opposed, say, to the relatively
simplistic Cartesian model of the self—obviously invites self-deception,
deception about the self. If there is no self other than that which is con-
strued, socially constructed, and constructed as social, it might be thought
dubious in what sense there is a self to be deceived about. There are, of
course, any number of logical and deconstructionist tricks to blow up
50. Both the general and the particularist claim are developed, for example, by Clifford
Geertz, Interpretation of Cultures (New York: Basic Books, 1966). Both are suggested, at
least, by Hegel in “Lordship and Bondage.”
51. Colin Turnbull, The Mountain People (New York: Simon and Schuster, 1974).
52. For two good philosophical discussions of this, see Alasdair MacIntyre’s After
Virtue, especially the opening chapters, and Nietzsche’s classic discussion of the Greeks in
his Birth of Tragedy, trans. W. Kaufmann (New York: Random House, 1967).
these generalized schemes into seemingly intractable paradoxes, but that

is not where the interest in self and deception lies. It is rather in the
increasingly rich web of personal and mutual conceptions of self with
which we now all must navigate in our dealings with the rest of the world.
It is a bewildering but absolutely essential learning experience, but, from
the available evidence, we are not doing terribly well. Ever more fanatic
individualism and retreat seems to be spreading as a global response, even
in societies that once were models of social harmony, in philosophy if
not in fact. The current fantasies with extraterrestrial visits and invasions,
I would suggest, is not an extension of our multicultural challenge but a
somewhat pathological distraction from that challenge. (How often those
extraterrestrials are either “just like us” or so viciously hostile that no
interaction is possible.)53
CONCLUSION: TWO CHEERS FOR THE TRUTH
Deception and self-deception, I have argued, are among the many ways
in which we navigate our way in the social world. Self-deception, in par-
ticular, is a way that we attend to our navigating, and in our increasingly
complex world, it is by no means clear that deception and self-deception—
understood as mechanisms for getting along in the world—are always
vices. It is one of the joys of philosophy that comes from realizing that
our ideas about ourselves and our place in the world are not truths but
experiments, whose outcome is always tentative. But all of this existential
freedom comes along with a cost, all too evident in current American
politics. If deception and self-deception can be instrumental in further-
ing one’s dreams and fantasies, they can also be the ruination of all of us.
There are limits to existential freedom. And that makes it imperative, as
we weigh the skeptical conclusions of this essay, to come down firmly
in favor of the truth. There are two many lies and self-deceptions in the
world that are just not forgivable.
ACKNOWLEDGMENT
Portions of this essay are adapted from “Self, Deception, and Self-Perception in
Philosophy,” a chapter I wrote for Self and Deception, ed. Roger Ames and Wimal
Dissanayake (Albany: State University of New York Press, 1996), 91–122.
53. E.T. and Independence Day are two “blockbuster” examples.

On Truth, Lies, and Bullshit
Harry Frankfurt
[I]
How do lies injure us? Actually, as everyone knows, there are many famil-
iar circumstances in which lies are not truly injurious to us at all. They
may sometimes be, on the whole, genuinely beneficial. For instance, a lie
may protect us in one way or another from becoming aware of certain
states of affairs, when no one (including ourselves) has anything in par-
ticular to gain from our being aware of them and when our awareness of
them would cause us or others serious distress. Or a lie may divert us from
embarking on a course of action that we find tempting but that would in
fact lead to our doing ourselves more harm than good. Clearly, we must
sometimes acknowledge that, all things considered, having been told a lie
was actually helpful to us.
Even so, we often feel at such times that there was surely something
bad about what the liar did. In the circumstances, it may be reasonable for
us to be grateful for the lie. Whatever good the lie may turn out to have
done, however, we believe at bottom that a better alternative would have
been for the beneficial effects of the lie to have been achieved by sticking
to the truth without any recourse to lying.
The most irreducibly bad thing about lies is that they contrive to inter-
fere with, and to impair, our natural effort to apprehend the real state of
affairs. They are designed to prevent us from being in touch with what
is really going on. In telling his lie, the liar tries to mislead us into believ-
ing that the facts are other than they actually are. He tries to impose his
will on us. He aims at inducing us to accept his fabrication as an accurate
account of how the world truly is.
Insofar as he succeeds in this, we acquire a view of the world that
has its source in his imagination rather than being directly and reliably
grounded in the relevant facts. The world we live in, insofar as our under-
standing of it is fashioned by the lie, is an imaginary world. There may be
worse places to live; but this imaginary world won’t do for us, at all, as a
permanent residence.
37
Lies are designed to damage our grasp of reality. So they are intended,
in a very real way, to make us crazy. To the extent that we believe them,
our minds are occupied and governed by fictions, fantasies, and illusions
that have been concocted for us by the liar. What we accept as real is
a world that others cannot see, touch, or experience in any direct way.
A person who believes a lie is constrained by it, accordingly, to life “in his
own world”—a world that others cannot enter, and in which even the liar
himself does not truly reside. Thus, the victim of the lie is, in the degree of
his deprivation of truth, shut off from the world of common experience
and isolated in an illusory realm to which there is no path that others
might find or follow.
[ II ]
Some philosophers insist, with considerable vehemence, that lying deci-

sively undermines the cohesion of human society. Immanuel Kant, for
example, declared that “without truth social intercourse and conversa-
tion became valueless” (Lectures on Ethics). And he argued that because
lying threatens society in this way, “a lie always harms another; if not
some particular man, still it harms mankind generally” (“On a Supposed
Right to Lie from Altruistic Motives”). Michel Montaigne made a similar
claim: “Our intercourse being carried on solely by means of the word, he
who falsifies that is a traitor to society” (“Of Giving the Lie”). “Lying is
an accursed vice,” Montaigne declared. And then he added, warming to
his subject with rather extraordinary intensity, “if we did but recognize
the horror and gravity of . . . [lying], we should punish it with flames more
justly than other crimes” (“Of Liars”). In other word, liars—more than
criminals of any other sort—deserve to be burned at the stake.
Montaigne and Kant certainly had a point. But they exaggerated. Effec-
tive social intercourse does not strictly depend, as they maintained, on
people telling one another the truth (not as, say, respiration strictly depends
on oxygen, being altogether impossible without it); nor does conversation
really lose all its value when people lie (some real information might
come through, and the entertainment value of the conversation might
even be increased). After all, the amount of lying and misrepresentation
of all kinds that actually goes on in the world (of which the immeasurable
flood of bullshit is itself no more than a fractional part) is enormous, and
yet productive social life manages somehow to continue. The fact that
people often engage in lies, and in other kinds of fraudulent behavior,
hardly renders it impossible to benefit either from living with them or
from talking with them. It only means that we have to be careful.
We can quite successfully find our way through and environment of
falsehood and fraud, as long as we can reasonably count on our own abil-
ity to discriminate reliably between instances in which people are mis-
representing things to us and instances in which they are dealing with us
On Truth, Lies, and Bullshit 39
straight. General confidence in the truthfulness of others is not essential,

then, as long as we are justified in having a certain sort of confidence in
ourselves.
To be sure, we are rather easily fooled. Moreover, we know this to be
the case. So it is not very easy for us to acquire and to sustain a secure
and justifiable trust in our ability to spot attempts at deception. For this
reason, social intercourse would indeed be severely burdened by a wide-
spread and wanton disrespect for truth. However, our interest in shielding
society from this burden is not what provides us with our most funda-
mental reason for caring about truth.
When we encounter people who lie to us, or who in some other way
manifest a disregard for truth, it tends to anger and upset us. But it does
not primarily do so, as Montaigne and Kant would presumably have had
it, because we fear that the mendacity we have encountered threatens
or encumbers the order of society. Our main concern is clearly not the
concern of a citizen. What is most immediately aroused in our response to
the liar is not public spirit. It’s something more personal. As a rule, except
perhaps when people misrepresent matters in which serious public inter-
ests are directly involved, we are dismayed far less by the harm liars may
be doing to the general welfare than by their conduct toward ourselves.
What stirs us against them, whether or not they have somehow managed
to betray all of humankind, is that they have certainly injured us.
[ III ]
Truth and caring about truth concern us, then, in ways that do not bear
simply on our quotidian practical interests. They have a deeper and
more damaging significance as well. One of the most rewarding of con-
temporary poets, Adrienne Rich, offers an account of the malign effect
that lying inevitably has—apart from its harmful effect on the person
to whom the lie is told—on the liar herself. With poetic exactitude, she
observes that “the liar leads and existence of unutterable loneliness,” 1
The loneliness is precisely unutterable because the liar cannot even reveal
that she is lonely—that there is no one in her fabricated world—without
disclosing, in doing so, that she has lied. She hides her own thoughts,
pretending to believe what she does not believe, and thereby she makes
it impossible for other people to be fully in touch with her. They cannot
respond to her as she really is. They cannot even be aware that they are
not doing so.
The liar refuses to permit himself, to the extent that he lies, to be
known. This is an insult to his victims. It naturally injures their pride. For
it denies them access to an elementary mode of human intimacy that
1. “Women and Honor: Some Notes on Lying,” in Adrienne Rich, Lies, Secrets and
Silence (New York: 1979), 191.
is normally taken more or less for granted: the intimacy that consists in
knowing what is on, or what is in, another person’s mind.
In certain cases, Rich notes, lies may cause an even more profound sort
of damage. “To discover that one has been lied to in a personal relation-
ship,” she says, “leads one to feel a little crazy.”2 Here again, her observa-
tion is perspicuous. When we are dealing with someone whom we hardly
know, we have to make a more or less deliberate assessment of her reli-
ability in order to satisfy ourselves that what she tells us coincides with
what she actually believes; and this assessment ordinarily pertains only to
certain specific assertions that she has made. With our close friends, on the
other hand, both of these conditions are usually relaxed. We suppose that
our friends are generally honest with us, and we take this pretty much for
granted. We tend to trust whatever they say, and we do so, mainly, not on
the basis of a calculation establishing that they are currently telling us the
truth, but because we feel comfortable and safe with them. As we famil-
iarly put it, we “just know that they wouldn’t lie to us.”
With friends, the expectation of access and intimacy has become natu-
ral. It is grounded not in a calculated judgment but in our feelings—that
is, in our subjective experience, rather than in any intellectual assessment
based on pertinent objective data. It would be too much to say that our
inclination to trust our friends belongs to our essential nature. But it could
be properly enough said, as we sometimes do in fact say, that trusting
them has come to be “second nature” to us.
That is why, as Rich observes, discovering that a friend has lied to us
engenders in us a feeling of being a little crazy. The discovery exposes to
us something about ourselves—something far more disturbing than merely
that we have miscalculated, or that we have made an error of judgment.
It reveals that our own nature (i.e., our second nature) is unreliable, hav-
ing led us to count on someone we should not have trusted. It shows us
that we cannot realistically be confident of our own ability to distinguish
truth from falsity—our ability, in other words, to recognize the differ-
ence between what is real and what is not. Successfully deceiving a friend
implies, needless to say, a fault in the one who tells the lie. However, it also
shows that the victim of the deception is defective, too. The liar betrays
him, but he is betrayed by his own feelings as well.
Self-betrayal pertains to craziness because it is a hallmark of the irra-
tional. The heart of rationality is to be consistent; and being consistent, in
action or in thought, entails at least proceeding so as not to defeat oneself.
Aristotle suggested that an agent acts rational insofar as he conforms his
actions to the “mean”—that is, to a point midway between excess and
deficiency. Suppose that, for the sake of good health, someone follows a
diet that is either so meager or so indulgent that it not only fails to improve
her health but actually leads her to become less healthy than she was.
2. “Women and Honor: Some Notes on Lying,” in Adrienne Rich, Lies, Secrets and
Silence (New York: 1979),186.
Aristotle urges that it is in this defeat of her own purpose, in this self-
betrayal, that the practical irrationality of the person’s divergence from
the mean consists.
Intellectual activity is undermined, similarly, by logical incoherence.
When a line of thought generates a contradiction, its further progressive
elaboration is blocked. In whatever direction the mind turns, it is driven
back: it must affirm what it has already rejected or it must deny what it
has already affirmed. Thus, like behavior that frustrates its own ambition,
contradictory thinking is irrational because it defeats itself.
When a person discovers that he has been told a lie by someone in
whose reliability he had found it natural to have confidence, this shows
him that he cannot rely on his own settled feelings of trust. He sees that
he has been betrayed, in his effort to identify people in whom he can have
confidence, by his own natural inclinations. These have led him to miss
the truth rather than attain it. His assumption that he could guide himself
in accordance with his own nature has turned out to be self-defeating,
and hence irrational. Since he finds that he is by nature out of touch with
reality, he may well fell that he is a little crazy.
[ IV ]
However penetrating and illuminating Rich’s thoughts about lying in per-

sonal relationships may appear to be, there is in this matter, as of course
there is in almost every matter, more than one side to the coin. Another
wonderful poet—perhaps, in fact, the greatest of all—has a rather different
tale to tell. Here is Shakespeare’s charming and provocative Sonnet 138:
When my love swears that she is made of truth,
I do believe her though I know she lies,
That she might think me some untutored youth,
Unlearn’d in the world’s false subtleties.
Thus vainly thinking that she thinks me young,
Although she knows my days are past the best,
Simply I credit her false-speaking tongue:
On both sides thus is simple truth suppressed.
But wherefore says she not she is unjust?
And wherefore say not I that I am old?
O, love’s best habit is in seeming trust,
And age in love loves not to have years told.
Therefore I lie with her and she with me,
And in our faults by lies we flattered be.
There is a widely accepted dogma according to which it is essential for

lovers to trust each other. Shakespeare is doubtful. His observation in the
sonnet is that the best thing for lovers—“love’s best habit”—is actually
not genuine trust. Mere “seeming trust” is just as good, he suggests, if not
sometimes even better.
The woman in Shakespeare’s poem professes to be utterly truthful—

she “swears that she is made of truth”—yet she lyingly pretends to believe
that the man is younger that she knows him to be. The man knows that
she doesn’t really believe this, but he decides to accept her characteriza-
tion of herself as truthful. So he brings himself to think that she really
does believe the lie about his age that he has told her, and thus that she
really does consider him to be younger than he actually is.
She lies to him about how honest she is, and about believing his
account of his age. He lies to her about how old he is, and about whether
he accepts her representation of herself as thoroughly truthful. Each of
them knows all of this: each knows that the other is lying, and both know
that their own lies are not believed. Each lyingly pretends to believe, how-
ever, that the other is flawlessly straightforward. This collection of lies
enables the two lovers, united in “seeming trust,” to believe that their self-
flattering lies about themselves—as impeccably honest, or as engagingly
youthful—have been accepted. And thus, lying with each other in this
way, they conclude lying happily together.
I suggested earlier that part of the fault in lying is that the liar, by deny-
ing access to what is truly in or on his mind, forecloses an elementary and
normally presumed mode of human intimacy. That foreclosure is surely
not a feature of the situation Shakespeare describes. The lovers in his son-
net each know not only what is in the other’s mind but what lies behind
it as well. Each knows what the other is really thinking. And each knows
that the other knows this: they lie egregiously to each other, but neither
is fooled. Each knows that the other is lying, and each is aware that his or
her own lies are seen through.
Neither of the lovers is actually getting away with anything. Both com-
prehend what is really going on in the mirrored and layered complex of
attempted deceptions that they have severally contrived. Everything is
reassuringly transparent to them. Both of the lovers are secure in their
awareness that their love is undamaged by their lies. They can see, through
all the lies they have been told, and through all the lies they themselves
have told, that their love survives even knowing the truth.
My guess is that the intimacy these lying lovers share, in virtue of
recognizing each other’s lies and in virtue also of knowing that their
own lies have not successfully deceived, is especially deep and enjoy-
able. The intimacy they achieve spreads to corners of themselves that
they have made specific and potentially costly efforts to keep hidden.
Despite it all, however, they see that they have seen through each
other. The hidden corners of each have been penetrated. The realiza-
tion by each of them that each both occupies and is occupied by the
other, and that this mutual penetration of their lies has marvelously
led their exercises of deception to the truth of love, must be wonder-
fully delicious.
I do not ordinarily recommend or condone lying. In most cases, I am all
for truth. Nevertheless, if you are confident that you can lie yourself into
a situation like the one Shakespeare sketches in his sonnet, my advice to

you is: Go for it!
[V]
Wittgenstein once said that the following bit of verse by Longfellow could
serve him as a motto:3
In the elder days of art
Builders wrought with greatest care
Each minute and unseen part
For the Gods are everywhere.
The point of these lines is clear. In the old days, craftsmen did not
cut corners. They worked carefully, and they took care with every aspect
of their work. Every part of the product was considered, and each was
designed and made to be exactly as it should be. These craftsmen did not
relax their thoughtful self-discipline even with respect to features of their
work that would ordinarily not be visible. Although no one would notice
if those features were not quite right, the craftsmen would be bothered
by their consciences. So nothing was swept under the rug. Or, one might
perhaps also say, there was no bullshit.
It does seem fitting to construe carelessly made, shoddy goods as in
some way analogues of bullshit. But in what way? Is the resemblance
that bullshit itself is invariably produced in a careless or self-indulgent
manner, that it is never finely crafted, that in the making of it there is
never the meticulously attentive concern with detail to which Longfellow
alludes? Is the bullshitter by his very nature a mindless slob? Is his prod-
uct necessarily messy or unrefined? The word shit does, to be sure, suggest
this. Excrement is not designed or crafted at all; it is merely emitted, or
dumped. It may have a more or less coherent shape, or it may not, but it
is in any case certainly not wrought.
The notion of carefully wrought bullshit involves, then, a certain inner
strain. Thoughtful attention to detail requires discipline and objectiv-
ity; it entails accepting standards and limitations that forbid the indul-
gence of impulse or whim. It is this selflessness that, in connection with
bullshit, strikes us as inapposite. But in fact it is not out of the question
at all. The realms of advertising and of public relations, and the nowadays
closely related realm of politics, are replete with instances of bullshit so
unmitigated that they can serve among the most indisputable and clas-
sic paradigms of the concept. And in these realms there are exquisitely
sophisticated craftsmen who—with the help of advanced and demand-
ing techniques of market research, public opinion polling, psychological
3. This is reported by Norman Malcom, introduction to Recollections of Wittgenstein, ed.

R. Rhees (Oxford: Oxford University Press, 1984), xiii.
testing, and so forth—dedicate themselves tirelessly to getting every word

and image they produce exactly right.
Yet there is something more to be said about this. However studiously
and conscientiously the bullshitter proceeds, it remains true that he is also
trying to get away with something. There is surely in his work, as in the
work of the slovenly craftsman, some kind of laxity that resists or eludes
the demands of a disinterested and austere discipline. The pertinent mode
of laxity cannot be equated, evidently, with simple carelessness or inatten-
tion to detail. I shall attempt in due course to locate it more correctly.
Wittgenstein devoted his philosophical energies largely to identify-
ing and combating what he regards as insidiously disruptive forms of
“nonsense.” He was apparently like that in his personal life as well. This
comes out in an anecdote related by Fania Pascal, who knew him in Cam-
bridge in the 1930s: “I had my tonsils out and was in the Evelyn Nursing
Home feeling sorry for myself. Wittgenstein called. I croaked: ’I feel just
like a dog that has been run over.’ He was disgusted: “You don’t know
what a dog that has been run over feels like.’ ”4
Now who knows what really happened? It seems extraordinary, almost
unbelievable, that anyone could object seriously to what Pascal reports
herself as having said. That characterizations of her feelings—so inno-
cently close to the utterly commonplace “sick as a dog”—is simply not
provocative enough to arouse any response as lively or intense as disgust.
If Pascal’s simile is offensive, then what figurative or allusive uses of lan-
guage would not be?
So perhaps it did not really happen quite as Pascal says. Perhaps
Wittgenstein was trying to make a small joke and it misfired. He was only
pretending to bawl Pascal out, just for the fun of a little hyperbole; and
she got the tone and the intention wrong. She thought he was disgusted
by her remark, when in fact he was only trying to cheer her up with some
playfully exaggerated mock criticism or joshing. In that case, the incident
is not incredible or bizarre after all.
But if Pascal failed to recognize that Wittgenstein was only teasing, then
perhaps the possibility that he was serious was at least not so far out of the
question. She knew him, and she knew what to expect from him; she knew
how he made her feel. Her way of understanding or of misunderstanding
his remark was very likely not altogether discordant, then, with her sense
of what he was like. We may fairly suppose that even if her account of the
incident is not strictly true to the facts of Wittgenstein’s intention, it is
sufficiently true to her idea of Wittgenstein to have made sense to her. For
the purposes of this discussion, I shall accept Pascal’s report at face value,
supposing that when it came to the use of allusive or figurative language,
Wittgenstein was indeed as preposterous as she makes him out to be.
Then just what is it that the Wittgenstein in her report considers to
be objectionable? Let us assume that he is correct about the facts; that
4. Fania Pascal, “Wittgenstein: A Personal Memoir,” in Rhees, Recollections, 28–29.

is, Pascal really does not know how run-over dogs feel. Even so, when she
says what she does, she is plainly not lying. She would have been lying
if, when she made her statement, she was aware that she actually felt
quite good. For however little she knows about the lives of dogs, it must
certainly be clear to Pascal that when dogs are run over, they do not feel
good. So if she herself had in fact been feeling good, it would have been a
lie to assert that she felt like a run-over dog.
Pascal’s Wittgenstein intended to accuse her not of lying but of misrep-
resentation of another sort. She characterizes her feeling as “the feeling of
a run-over dog.” She is not really acquainted, however, with the feeling to
which this phrase refers. Of course, the phrase is far from being complete
nonsense to her; she is hardly speaking gibberish. What she says has an
intelligible connotation, which she certainly understands. Moreover, she
does know something about the quality of the feeling to which the phrase
refers: she knows at least that it is an undesirable an unenjoyable feeling,
a bad feeling. The trouble with her statement is that it purports to convey
something more than simply that she feels bad. Her characterization of
her feeling is too specific; it is excessively particular. Hers is not just any
bad feeling but, according to her account, the distinctive kind of bad feel-
ing that a dog has when it is run over. To the Wittgenstein in Pascal’s story,
judging from his response, this is just bullshit.
Now assuming that Wittgenstein does indeed regard Pascal’s charac-
terization of how she feels as an instance of bullshit, why does it strike
him that way? It does so, I believe, because he perceives what Pascal says
as being—roughly speaking, for now—unconnected to a concern for the
truth. Her statement is not germane to the enterprise of describing reality.
She does not even think she knows, except in the vaguest way, how a run-
over dog feels. Her description of her own feeling is, accordingly, some-
thing that she is merely making up. She concocts it out of whole cloth;
or, if she got it from someone else, she is repeating it quite mindlessly and
without any regard for how things really are.
It is for this mindlessness that Pascal’s Wittgenstein chides her. What
disgusts him is that Pascal is not even concerned whether her statement is
correct. There is every likelihood, of course, that she says what she does
only in a somewhat clumsy effort to speak colorfully, or to appear viva-
cious or good-humored; and no doubt Wittgenstein’s reactions—as she
construes it—is absurdly intolerant. Be this as it may, it seems clear what
that reaction is. He reacts as though he perceives her to be speaking about
her feeling thoughtlessly, without conscientious attention to the relevant
facts. Her statement is not “wrought with the greatest care.” She makes it
without bothering to take into account at all the question of its accuracy.
The point that troubles Wittgenstein is manifestly not that Pascal has
made a mistake in her description of how she feels. Nor is it even that she
has made careless mistake. Her laxity, or her lack of care, is not a matter of
having permitted an error to slip into her speech on account of some inad-
vertent or momentary negligent lapse in the attention she was devoting to
getting things right. The point is rather that, as far as Wittgenstein can see,
Pascal offers a description of a certain state of affairs without genuinely
submitting to the constraints which the endeavor to provide an accurate
representation of reality imposes. Her fault is not that she fails to get
things right, but that she is not even trying.
This is important to Wittgenstein because, whether justifiably or not,
he takes what she says seriously, as a statement purporting to give an
informative description of the way she feels. He construes her as engaged
in an activity to which the distinction between what is true and what is
false is crucial, and yet as taking no interest in whether what she says is
true or false. It is in this sense that Pascal’s statement is unconnected to
a concern with truth: she is not concerned with the truth-value of what
she says. That is why she cannot be regarded as lying; for she does not pre-
sume that she knows the truth, and therefore she cannot be deliberately
promulgating a proposition that she presumes to be false: her statement is
grounded neither in a belief that it is true nor, as a lie must be, in a belief
that is not true. It is just this lack of connection to a concern with truth—
this indifference to how things really are—that I regard as the essence of
bullshit.
[ VI ]
What bullshit essentially misrepresents is neither the state of affairs to

which it refers nor the beliefs of the speaker concerning that state of
affairs. Those are what lies misrepresent, by virtue of being false. Since
bullshit need not be false, it differs from lies in its misrepresentational
intent. The bullshitter may not deceive us, or even intend to do so, either
about the facts or about what he takes the facts to be. What he does nec-
essarily attempt to deceive us about is his enterprise. His only indispens-
ably distinctive characteristic is that in a certain way he misrepresents
what he is up to.
This is the crux of the distinction between him and the liar. Both he
and the liar represent themselves falsely as endeavoring to communicate
the truth. The success of each depends on deceiving us about that. But
the fact about herself that the liar hides is that she is attempting to lead
us away from a correct apprehension of reality; we are not to know that
she wants us to believe something she supposes to be false. The fact about
herself that the bullshitter hides, on the other hand, is that the truth-
values of her statements are of not central interest to her; what we are not
to understand is that her intention is neither to report the truth nor to
conceal it. This does not mean that her speech is anarchically impulsive,
but that the motive guiding and controlling it is unconcerned with how
the things about which she speaks truly are.
It is impossible for someone to lie unless he thinks he knows the truth.
Producing bullshit requires no such conviction. A person who lies is
thereby responding to the truth, and is to that extent respectful of it.

When an honest man speaks, he says only what he believes to be true; and
for the liar, it is correspondingly indispensable that he considers his state-
ments to be false. For the bullshitter, however, all these bets are off: he is
neither on the side of the true nor on the side of the false. His eye is not
on the facts at all, as the eyes of the honest man and of the liar are, except
insofar as they may be pertinent to his interest in getting away with what
he says. He does not care whether the things he says describe reality cor-
rectly. He just picks them out, or makes them up, to suit his purpose.
In his essay “Lying,” Augustine distinguishes lies of eight types, which
he classifies according to the characteristic intent or justification with
which a lie is told. Lies of seven of these types are told only because they
are supposed to be indispensable means to some end that is distinct from
the sheer creation of false beliefs. It is not their falsity as such, in other
words, that attracts the teller to them. Since they are told only on account
of their supposed indispensability to a goal other than deception itself,
Augustine regards them as being told unwillingly: what the person really
wants is not to tell the lie but to attain the goal. They are therefore not
real lies, in his view, and those who tell them are not in the strictest sense
liars. It is only the remaining category that contains what he identifies as
“the lie which is told solely for the pleasure of lying and deceiving, that
is, the real lie.”5 Lies in this category are not told as means to any end
distinct from the propagation of falsehood. They are told simply for their
own sake—that is purely out of a love of deception: “There is a distinc-
tion between a person who tells a lie and a liar. The former is one who
tells a lie unwillingly, while the liar loves to lie and passes his time in the
joy of lying The latter takes delight in lying, rejoicing in the falsehood
itself.”6
What Augustine calls “liars” and “real Lies” are both rare and extraor-
dinary. Everyone lies from time to time, but there are very few people to
whom it would often (or even ever) occur to lie exclusively from a love
of falsity or of deception.
For most people, the fact that a statement is false constitutes in itself a
reason, however weak and easily overridden, not to make the statement.
For Augustine’s pure liar it is, on the contrary, a reason in favor of mak-
ing it. For the bullshitter, it is in itself neither a reason in favor nor a rea-
son against. Both in lying and in telling truth, people are guided by their
beliefs concerning the way things are. These guide them as they endeavor
either to describe the world correctly or to describe it deceitfully. For this
reason, telling lies does not tend to unfit a person for telling the truth in
5. “Lying,” in Treatises on Various Subjects, in Fathers of the Church, ed. R. J. Deferrari,

vol. 16 (New York: Fathers of the Church, 1952), 109. Augustine maintains that telling
a lie of this type is a less serious sin than telling lies in three of his categories and a more
serious sin than telling lies in the other four categories.
6. Ibid., 79.
the same way that bullshitting tends to. Through excessive indulgence in
the latter activity, which involves making assertions without paying atten-
tion to anything except what it suits one to say, a person’s normal habit of
attending to the ways things are may become attenuated or lost. Someone
who lies and someone who tells the truth are playing on opposite sides, so
to speak, in the same game. Each responds to the facts as she understands
them, although the response of the one is guided by the authority of the
truth, while the response of the other defies that authority and refuses
to meet its demands. The bullshitter ignores these demands altogether.
She does not reject the authority of the truth, as the liar does, or oppose
herself to it. She pays no attention to it at all. By virtue of this, bullshit is
a greater enemy of the truth than lies are.
Deceit in War and Trade
William Ian Miller
It is believed by a good portion of humankind, and not disconfirmed by

the experience of more than a few birds, fish, and insects, that without
deceit and dissimulation there would be no sex, no reproduction of the
species. Even the gods need tricks to bed their objects of desire. Zeus
becomes a swan, a bull, a snake, a stream of gold, an eagle, or even a lowly
shepherd, to seduce or rape as the case may be.1 The first time between
mates requires a seducer’s arts (or alcohol), even if, and this is a constant
theme of many trickster tales, the seducer was himself seduced by arts
superior to his own, and the only success he can claim as a deceiver is the
self-deception that has him believing that his own artfulness determined
the course of events. Add deceits like breast implants, Viagra, hair color,
botox, and liposuction to the mix, and we have more people tricked than
was ever possible in the good old days. Maybe not: in days of yore mere
clothing was felt to be deceitful. Clothes covered an ugly truth beneath
and “made the man” out of whole cloth. In Hebrew, the word for clothing
and the word for deceit, betrayal, and treachery come from the same root:
B-G-D.2 But are such cover-ups, props, and enhancements still deceitful
if openly owned up to? Sure they are, for falseness owned up to is one of
falseness’s most common tricks.
It is hard to imagine the world getting started without trickery and
dissimulation. If the snake were not in the garden, there would be neither
history nor a plausible imagining of human psychology. We would not be,
or if we were to be, we would not be the least bit interesting. For us post-
lapsarian souls, the trick is to find the proper amount of wariness to keep
the costs of our foolishness down without becoming paranoid, and, per-
haps, to develop the proper amount of wiliness so as to get our way with
others without being discovered, or if discovered, discovered in such a
1. See Ovid, Metamorphoses 6.113–14.

2. Hebrew gets you every which way, naked or clothed: the word for “naked” (‘rum)
applied to the innocent Adam and Eve in the last verse of the second chapter of Genesis
is a perfect homonym with the word for “cunning,” rendered as “subtil” in the Authorized
Version, used to describe the serpent in the next verse beginning Genesis 3.
49
way as to become known either as lovable rogues or as persons to be reck-

oned with. As with love, so with war, and so, too, with the more modest
exchanges of the marketplace. What follows in this essay is a genealogy
on deceit in war and trade placed mostly in premodern times, a sketchy
genealogy at that, with some bastard children admitted to the inheritance
and other legitimate ones excluded.
Sun Tzu, fifth century B.C., says the art of war is the art of deception:
“all warfare is based on deception; therefore when capable, feign incapac-
ity; when active, inactivity. When near make it appear that you are far
away; when far away that you are near. Offer the enemy a bait to lure
him.” Easier said than done, for more than the obvious reason that your
enemy is reading Sun Tzu too. The reason Sun Tzu is so categorical is that
he knows warriors are jealous of their courage, which will lead them to
prefer ill-advised charges and refuse practical retreats, lest they be blamed
as cowards. The problem is not so much enduring the taunts of the enemy
as those of one’s companions or the songs your women will sing in mock-
ery.3 Strategies and tactics of deception have to be managed so as not to
appear to be motivated by cowardice. Honor requires a certain amount of
risk seeking for the sake of risk seeking, within limits to be sure, for one
of the great rewards of honor is to live to enjoy the reputation for having
it. So courage has to concede some space to guile, and guile has to pay
some deference to courage. Odysseus, no less than Achilles, gets his epic.
The problem, as always, is how to manage practically. Does one employ
feigned flights, ambushes, night attacks? Does one wear the uniform of
the enemy? Does one stab in the back, violate a safe conduct? Just what,
if anything, is not fair in love and war?
SKILL
Let me start with a frolic and a small detour. Take the case of boxing,
football, or any sport that involves facing off against an opponent rather
than, say, track or golf, which involves tests that require comparisons
of discrete individual performances. 4 Football is all about making your
opponent think you are going to pass rather than run, run here rather
than there, throw to this receiver rather than that one. Receivers develop
moves to fake out defenders. Boxers and wrestlers feint, set up their
opponents. Moves are the name of the game. And while we delight in the
player who is so strong and powerful that he need not develop the skills
3. One might secure a reputation for courage with one grand performance early in
one’s career and thereby gain enough credit to make room for considerable prudence
thereafter, but even then there would be gossiping. See, e.g., Polybius (10.3) on Scipio
Africanus.
4. Though races longer than four hundred meters involve strategizing, and positioning
tactics, and golfers may try to get into each other’s heads, the rough distinction survives.
Deceit in War and Trade 51
of trickery and fakery and need do nothing more than charge straight
ahead and overcome with brute strength, he is the exception; even he
will eventually confront someone who is his match in sheer strength and
then the one who can trick, feint, and fake will defeat the other. There is
no moral problem here. And the analogy with war works.
What if, however, the big strong brute, a Goliath, gets taken down
from afar by someone who brings a different technology to bear, not quite
within the rules of the game Goliath thinks he is playing, but within the
rules of the game his opponent thinks is the real game at hand? A moral
problem begins to emerge, rather different from the one that arises when
a man nine cubits and a span tall decked out in armor from head to toe
bearing a spear crafted with the latest Iron Age metallurgy goes against
an untested youth without armor still using Stone Age technology. Asym-
metry of force raises different moral problems from those raised by guile
and deceit, though one of the goals of deceit may be to gain a local advan-
tage so that you become, relatively speaking and for a brief exploitable
moment, the one holding the winning hand.
For us, drugging one’s opponent is clearly outside the game, yet more
than a few athletes think that drugging oneself can do nice work if you
can sneak it in: thus the tawdry tales that are the daily fare of baseball,
swimming, cycling, cross-country skiing, track and field, even golf. The fact
that we recognize a distinction between the deceit that is openly part of
the skill set of a particular game and the deceit of juicing oneself up with
undetectable chemicals has a history. Witness the Iliad, and many a medi-
eval romance, in which it is not yet clear that actions like bribing judges,
enlisting the gods on your side, sticking poles in the spokes of the other’s
wheels, poisoning your sword in a duel, using magic weapons, and claim-
ing a privilege that entitles you to win a race by virtue of your social rank
rather than your speed were not clearly outside the game. Though some
advances have been made, sometimes it is one step forward, two back.
The boundaries separating one game from another, or the inside from
the outside, are often in dispute. Archers, though very effective at kill-
ing at relatively small risk to themselves, like Paris or the English long-
bowmen, were looked down on by the men who fought mano a mano,
like Hector or knightly men-at-arms. That David could stay well out of
reach of Goliath’s javelin and hit him from afar with a force-multiplying
sling—is that a move, a tactic in the game; or not the same game? Bald
trickery and cheating if you are a Philistine, a glorious triumph if you are
a Judean. In the age of chivalry, a thirteenth-century French legal compi-
lation defines treachery in war, “treason” in his idiom, as hitting someone
who could not see what hit him.5 This was meant to give a bad name to a
5. “Treachery is when one attacks a man so that he cannot see the blow coming ...or
when one surprises another and strikes him when he cannot defend himself”; Li livres de
Jostice et de Plet, ed. Rapetti [sic], in Collection de Documents Inédits sur l’Histoire de France
(Paris: Firmin Didot, 1850), bk. 19, ch. 19, p. 297. (Thanks to Stephen D. White for this
citation, and see his “Alternative Constructions of Treason in the Angevin Political World:
knight who snuck up from behind, but mostly it meant woe for the poor
crossbowmen a hundred meters behind the general melee who ran out of
bolts. Bowmen were not held for ransom, as a captured knight would be,
but were butchered for their “treachery,” though of course one made sure
to have as many as one could afford serving one’s own cause. 6
There is an anxiety, though fainter in our day than in earlier times,
that lurked in the notion of skill and deftness that is evidenced by an
insistent “duplicity,” a conflicted moral tone, in more than a few of the
words in English meaning skill. The word cunning in Middle English was,
in fact, the word for technical skill such as a craftsman might possess; a
cunning silver smith was one who did beautifully intricate and admirable
work. By the end of the sixteenth century, cunning had begun to acquire
a pejorative sense that ended up driving out the morally neutral and posi-
tive senses it had had.7 There was also the pun and confusion in Middle
and early Modern English caused by two mysteries of different origin, one
Greek, the other Latin. The former was the mystery of religion, the mys-
teries of the faith, which we still employ secularly today when reading a P.
D. James mystery; the second mystery was the mystery 8 of craft, with craft
bearing the sense of skill, art, trade, as, for instance, Chaucer’s Reeve, who
“in youthe ...hadde lerned a good myster; / He was a wel good wrighte,
a carpenter.”9 This mystery was confused with mastery, for fairly obvious
reasons of near homonymity, and by the coincidence that a person who
passed beyond being a journeyman in his "myster" became a "master" in
his craft. The result is a triple pun, each form suggesting in its own way
a sleight of hand: sleight, by the way, like cunning, bears the competing
senses of both trick and skill, though the latter sense was already rare by
the nineteenth century.10 And then add to that the double meanings of
the word craft just used,11 whose bad and good senses coexisted from the
Traïson in the History of William Marshal,” e-Spania, 4 (décembre 2007), 1–47, mis en ligne
le 21 décembre 2007: http://e-spania.revues.org/document2233.html. Consulté 10 avril
2008.) I discuss various revenge cultures in which it was perfectly acceptable to dispatch
one’s vengeance target without him knowing what or who had hit him in Eye for an Eye
(Cambridge: Cambridge University Press, 2006), 150–51.
6. On butchering crossbowmen, see Matthew Strickland, War and Chivalry: The
Conduct and Perception of War in England and Normandy, 1066–1217 (Cambridge:
Cambridge University Press, 1996), 180–81. Nonetheless, knights in chivalric battle employed
nearly all the tricks advised by Sun Tzu; the most usual one being to avoid battle altogether
and substitute for it the plundering of one’s opponents’ peasants; see John Gillingham,
“War and Chivalry in the History of William Marshal,” in Gillingham, Richard Coeur de Lion:
Kingship, Chivalry and War in the Twelfth Century (London: Hambledon Press, 1994), 227–41.
7. OED, cunning sb. For an engaging dissection of cunning see Don Herzog, Cunning
(Princeton: Princeton University Press, 2006).
8. OED, mystery 2, from Medieval Latin misterium, altered form of ministerium.
9. Canterbury Tales, General Prologue, vv. 613–14.
10. OED, sleight sb. 1.
11. OED, craft sb. 1.
twelfth century, and one sees a deep mistrust of the moral quality of skill
and those who were skilled. They were felt to be uncanny, as if they were
so many otherworldly tricksters.
Not only technical skills but intelligence itself raised grounds for suspi-
cion and was of uncertain moral warrant. The serpent, after all, “was more
subtil than any beast of the field.” Many felt intelligence was in league
with the devil: the “cunning man” or the “cunning woman” was also the
specialist in black arts, which is another sense cunning took on. The mor-
alists hailing reason as a virtue ran up against the popular understanding
that reason was nothing more than a stick in the bundle of tricks clerics
and lawyers used to make sure the game being played was one favoring
their own skill set: wordplay rather than swordplay.12
EXPEDIENCE AND SMART TOUGH GUYS
Heroes of the sort who were known as crafty, subtle, wily, “wise,” an Odys-
seus or a David, would always test the boundaries and ultimately accept
expedience as the chief arbiter of their consciences. That Odysseus goes out
one night to kill sleeping allies of the Trojans is part of what it meant to be
wise in his world, but that the battle-brave, traditionally heroic Diomedes
does the actual butchering, while Odysseus drags the bodies away so as not
to worry the horses into making too much noise to spoil the fun, is conven-
tionally understood to show that Homer is critiquing the ugly business that
not merely compromised the heroic ethic, but in some sense sustained it.13
But perhaps we are making Homer subtler than he meant to be by mak-
ing him more congenial to our own deep ambivalence regarding honor
and the heroic ethic. Why assume that, in the night-killing episode, he is
undermining Odysseus or Diomedes, morally or otherwise? For Homer,
it is mostly a comic interlude anyway, an excuse to get raucous laughter
at the expense of the groveling coward, Dolon, captured by Diomedes
and Odysseus. Homer may be rather indifferent to the moral implica-
tions, if any, of spearing sleeping soldiers if doing so made tactical sense,
especially when it prompts a good laugh. He may merely have meant to
indicate that heroes get to charge ahead and act heroically by day—no
mean feat—as long as they have someone minding the purely practical
side of things, taking advantage where advantage is to be had, at night.
And one person, a thoroughly respectable person at that, could play both
roles: upstanding man of courage by day, a sly and sneaky killer by night.
Besides, whatever the moral cost that attends night killing is rather neatly
balanced by the moral failure of the sleeping victims: they die for their sin
of culpable complacency.
12. See Fredric Cheyette, “The Invention of the State,” in B. K. Ladner and K. R. Philp,
eds. Essays on Medieval Civilization (Austin: University of Texas Press, 1978), 143–78.
13. Iliad 10.480–510.
An assassin by night, Diomedes even makes out like a bandit in gift-

exchange in broad daylight, as a trader of sorts. When Diomedes meets
Glaukos in battle, Diomedes, appealing to their grandfathers’ close
friendship, takes off his armor and offers it to Glaukos, suggesting that
they exchange armor as their ancestors had exchanged gifts two genera-
tions earlier. The trouble for Glaukos is that his armor is gold, whereas
Diomedes’ is bronze, and the difference in value is, as Homer hastens to
tell, one hundred bulls to nine. To Homer’s dismay, Glaukos agrees to the
exchange: “Zeus must have stolen Glaukos’s wits away.” 14 But what was
poor Glaukos to do? Diomedes traps him in a web of piety and takes him
to the cleaners. And remember, Diomedes is not even famed for cunning;
that is Odysseus’s honor.
There is a strange moral economy here. Surely, those who believe in
face-to-face battle fair and square are not all operating in bad faith; but
they are not about to disown the advantage of having some Odyssean
sorts on their side, and they surely do not want such sorts in the high com-
mand of the enemy. Then, too, the cunning types, like Odysseus or David,
are not weaklings either. Is it that they (and we) tolerate Odysseus and
David, and even the wily Jacob, because they are tough guys, too? Odys-
seus and David are hardly cowering poltroons. Both are consummate war-
riors; they stop at nothing. And even Jacob, though quaking in his sandals
when informed that Esau and his band are approaching, is, like Odysseus,
a very good wrestler.15
The Icelandic sagas add more nuance. They show that if one was a
cagey trickster, one need not be much of a warrior: one could be sneaky
and wily and even enjoy a certain threat advantage that came with that
reputation. As long as they did not scare easily, these wily souls would
even be respected, feared, and admired, and be counted men of honor. 16
But they still needed steely nerves.
The ancient moralists counseled that fortitudio needed an assist from
sapientia, that without the latter the first would kill itself with grand
charges and heroic action, and without the former the wise would be
more akin to university professors than to either Odysseus or David.
14. Iliad 6.232; Robert Fitzgerald. trans. (Garden City, NY: Anchor, 1974). Homer’s
joke at Glaukos’s expense is more complex than it appears at first glance. The difference
in value between Diomedes and Glaukos’s armor would only be one hundred to nine
in an armorer’s shop, and that is part of the wit of the joke, for it wryly introduces hard
mercantile concerns that both fund and are the object of much glory seeking. Back in
the Trojan camp, the bronze armor of Diomedes, because it had belonged to the great
Diomedes, would have a premium attached to it. In the Greek camp, the gold armor
gains little for having been Glaukos’s, though its value might still be enhanced beyond
comparable gold armor by its now having the comic tale of its acquisition attached to it.
15. See for fuller discussion my The Mystery of Courage (Cambridge: Harvard
University Press, 2000), 162–67.
16. Thus Snorri the Priest in Eyrbyggja Saga and Laxdæla Saga; Sturla in Sturlu Saga.
Gudrun in Laxdaela Saga. Compare the easier-to-scare Mord Valgardsson in Njál’s saga.
How much sapientia was needed, though, if the Trojans, after holding out
behind walls and in battle for ten years, could fall for the Trojan horse?
But then the Trojans had as yet no proverb saying “Fear Greeks bearing
gifts,” until Laocoön coined it on the spot, though the spot was in the
Aeneid, and thus too late to help Troy; moreover, in the Aeneid it did not
have the force of a proverb but of an impious novelty, and within an hour
Laocoön and his sons were devoured by sea-serpents.
How could a trick like that work? Virgil has to devote considerable
effort to make the tale told by Sinon, the Greek “deserter,” convincing
enough for the Trojans to appear as something less than idiots for having
believed him. Virgil could hardly wish his Romans to be descended from
Trojan fools, no matter how heroic the escapee Aeneas might be. Surely,
the Trojans had their own trickster tales that expressed the same prover-
bial wisdom as the tale of the horse. Everyone does. The Trojans had their
moralists and comedians who either warned against or complained about
the wiliness of their slaves, their women, and their enemies and coined
proverbs about the poison that lurked within the benignest of gifts, but
still the deceit worked.
Deceit works as well as it does because humans are so easy to set up
for the fall. Warnings of doom go unheeded. And why shouldn’t they?
Cassandra raises the same problems all prophets raise: it is very hard to
tell the false doomsayer from the true, for even the false one gets it right
some of the time, given that we get sick, have accidents, grow old, and
die. It might thus be rational not to listen to Cassandra, unless she can
come up with some hard evidence that the straits are more dire than
usual.17 Besides, doomsayers are a dime a dozen. Why should Caesar have
been wary about the Ides of March, when he no doubt had been treated
to years of being warned to beware the Nones of February, the Kalends
of June, and so on. He had to ignore doomsayers, or he would never have
left the house, and would never, for certain, have been the world histori-
cal figure he was.
But even if the Trojans had the story of the Trojan horse or believed
Cassandra, the fact is that the same trick can work again and again. Like
the plots of novels, there are a limited number of tricks, and they merely
have to trick themselves out in slightly different costumes to work in
every generation. The very predictability of known deceits that have
been employed for the umpteenth time give the trickster the advantage
of his or her mark’s complacency with regard to that particular trick: no
17. Obviously, reason alone does not block our ears to Cassandra’s panicky predictions.
Our deafness to her doomsaying is also a sign of what in the sixteenth century was called
the sin of security, a culpable looking on the bright side of things, “a forgetting mortalitie,”
in Thomas Nashe’s words; see my Faking It (Cambridge: Cambridge University Press,
2003), 28. But we also may, not unjustly, mistrust the doomsayer as motivated by a
dispositional pessimism that resents the pleasure of those who are eating, drinking, and
being merry, who may be doing so not out of a culpable sense of security but because that
is how they are choosing to meet their ends, not how they are choosing to live their lives.
one is going to get me on that one, and so we lower our defenses and get
fooled again.18
But war raises the stakes: if the standard trickster makes us pay for our
greed and our vanity, the emotional backdrop of war is dominated by fear
for life. Lighting more campfires than you have men to warm in order to
make the enemy believe you have a larger force has been employed hun-
dreds of times in scores of wars. It is not as if the enemy does not know
the trick (they have employed it more than a few times themselves, and
perhaps at the same time you are employing yours); it is just that they are
scared and likely to err on the side of nonconfrontation. And on occasion,
they have lit fewer campfires than needed to entice you into battle when
they have the advantage in numbers. So what do those campfires mean?
If you are being tricked into more wariness than it was wise to have at a
particular moment, and you thus let a small enemy force slip away under
the cover of darkness, you at least are alive to get it right, or wrong again
perhaps, the next time. But even without fear as the backdrop, in mere
matters of anticipation and reading another’s intentions, as in a basket-
ball game, for instance, the same crossover move works again and again,
against people who are skilled players themselves and have seen it a thou-
sand times: it is not the move itself but the timing of it that bears a good
portion of its guile.
Though both sapientia and wisdom are words that indicate a virtue and
bring virtuous associations to mind, the “wise” in wisdom, even the sapiens
in sapientia, can end up partaking of the more dubious world of feints
and double moves—that is, of craft, cunning, and sleights—as much as
of the world of wise counsel, prudence, and subtle strategy. The kind of
“wisdom” that was required in the midst of ancient battle shared more
with the skill of crossover dribbles and how to anticipate them, when to
fake right and go left, than with high matters of strategy. Thus, in the Bat-
tle of Maldon—an Anglo-Saxon poem of extraordinary power composed
shortly after the battle (991 A.D.) it commemorates—the English leader
is called fród, a word bearing the meanings “wise, skillful, sage, old,” for
spearing a Viking through the neck in close combat, after that same Viking
has wounded him. Fród applies to him in part as an epithet for being the
leader of the English troop, an elder, but the nearest referent for fród in
the text is the spearing. The Old English says the hand of the fród warrior
“guided” his spear through the neck of the Viking (“Fród wæs sé fyrdrinc”;
he let his “francan wadan / þurh ðæs hysses hals, hand wísode”).19 He is fród
18. Some betrayals are so openly negotiated that it boggles the mind that they can
work; one needn’t even build a horse. In eleventh-century Byzantium, we see people who
are guarding city walls having “cautious” but obviously quite loud conversations with
besiegers below about opening the gates, and this after the besiegers have walked around
shouting up to various defenders to find out who could be bought; see Anna Comnena,
The Alexiad 2.9, trans. E. R. A. Sewter (London: Penguin, 1969), 95–96.
19. The Battle of Maldon, vv. 140–42.
because he knows how to make the moves to maneuver a spear through

soft, exposed neck tissue in the heat of the action, while wounded, rather
than running his spear up against a shield or armor. In short, he fakes the
Viking out of his linen undergarments, deceives him as to some subset of
his intentions.
Here is why philology is seductive, for it tricks me into letting it bear
too much argumentative weight, but that does not mean I will not con-
tinue to be beguiled by it (or disown being beguiled when I recognize that
I am). The word in Old English that in this passage means “to guide,” “to
aim,” is, in its infinitive form, wísian,20 and is formed from the same root
that yields the word wise in Modern English. So the wise person is he who
knows the way to guide a spear to its target, so as to guide another to his
doom: wisdom here can encompass skill with spears in close combat as
well as wisdom in counsel.
Wise, to return to a theme presented earlier, has the same dark side to
it as craft, cunning, and sleight do: thus wise, in the early modern period, no
less than cunning, meant skilled in magic, particularly in the black arts, so
that the village “wise woman” was the witch, 21 whom one consulted about
one’s impotence or infertility, until one decided to blame the disorder on
her cunning and then betray her to the magistrate for burning. Nor should
it be surprising that in Middle English philosopher meant, in the words of
the Oxford English Dictionary, “an adept in occult science, as an alchemist,
magician, diviner of dreams, weather-prophet, etc.”22
THE HALF-MACHIAVELLIANISM OF MONTAIGNE
Surely, in the golden age when men were men and met in battle the way
Hector met Achilles (though Hector started running) and not quite as
David met Goliath, there was no perfidy, no tricksterism. Why? Because
our forefathers were too honorable, too virtuous—or too dumb—to set
the other one up with a stab in the back or with poison in the night.
Or was it only that they liked to tell stories of grand virtue, which
on occasion they aspired to, but, mercifully, still kept a few tricks up
their sleeve, about which they also told tales? Montaigne takes on these
issues right at the start of his Essays, devoting essays 5 and 6 of Book I
to them.
Essay 5 bears the title “Whether the Leader of a Besieged Place Should
Go Out to Negotiate.” The answer is no, by gum, don’t do it! unless you
want to see your walled town breached while you are talking peace out-
side. Hence the title of essay 6, which continues the theme of 5: “The
Hour of Negotiations Is Dangerous.”
20. OED, wise v.1, 2b.

21. OED, wise a. 2b.
22. OED, philosopher 2.
This leads Montaigne to musings on fair play versus deceit and expedi-
ence. First he takes on the issue of a glorious past, when men were too
honorable to be so treacherous as to win wars while pretending to make
peace. He tells of a Roman legate, Marcius,23 who secured a truce with
his opponent, Perseus, the king of Macedonia, in order to buy time to
reinforce his troops. But the Senate disapproved of Marcius’s behavior
as contrary to the ancient practice of their fathers, which was “to fight
with courage, not with sleights, not by surprise or night ambushes, nor by
feigned flight and unexpected about-faces, nor to have undertaken a war
unless it was first declared, and often designating the hour and place of
battle.” To behave as Marcius did was to act, they said, with “Greek crafti-
ness or Punic cunning, for whom it was actually less glorious to win by
force than by fraud.”
The view attributed to the Senate was not all tough talk, easy for old
men safe in Rome to engage in. Some policy was alleged for the old vir-
tuous way of no feints and sleights: “Only those recognize themselves
defeated who know they have been beaten by valor, troop by troop, in a
fair and just war.”24 In other words, you gain threat advantage by tromping
the smithereens out of an enemy when they are ready for you and are giv-
ing their best. They will then come away with no easy excuses to prime a
self-deceiving belief in their own superiority. They will be very reluctant
to try again and will accept defeat as the order of things. Winning on the
sly with deceit is not as destructive of the other side’s will to take up the
cause again.
A point before continuing with my theme: it appears to be an oft-
occurring, anxious fantasy of the powerful to believe that some subset of
peoples whom they have defeated and abused are all cunning and craft,
and defiling, loathsome and dangerous for being so. Thus in the Roman
view the Carthaginians and the Greeks were all guile and deceit, and
thus, though sometime later, in just about everyone’s paranoid fantasies,
the Jews, who for more than a thousand years have served as the univer-
sal cunning and guileful Other. Guile gets you only so far, as I suppose
the old men of the Senate well knew. Guile may get you a victory here
and there and at lower cost than if it were gained without guile but we
need only look at the lot of the Greeks, Carthaginians, and Jews, that is,
at the big-time wily losers. Guile needs more than guile: it also needs
23. Montaigne, Oeuvres Complètes, ed. Maurice Rat (Paris: Gallimard, 1962).
Montaigne actually has Lucius Marcius, which must be an error for Quintus Marcius. It
was Lucius Aemilius Paulus who defeated Perseus in the Third Macedonian War, who
had the highest reputation, not Quintus Marcius, who was known for deceitfulness.
Translations are mine.
24. A nearly opposite argument is made in Thucydides 4.19: an outnumbered party
suggests to the opposition that an honorable peace will bring much better consequences
than one side overwhelming the other, for then the overwhelmed will want to get even,
whereas honor will require them not to break a peace they negotiated. The argument is,
not surprisingly, made by those who are about to get overwhelmed.
a durable fist to secure the advantage guile gains (unless, like the more
conventional small stakes conman, you skip town before the dupe knows
he has been fleeced).25 Though the guileful might have contempt for the
lack of subtlety of the mighty whom they see as big dumb brutes, the
mighty brutes are not so dumb; that they are is a fantasy of the defeated.
The mighty rightly suspect—precisely because they are smart—that the
cunning fear brute toughness, more rightly than the mighty do wiliness.
The cunning weak have rather less margin for error than the mighty. Nor
are the mighty victors averse to recruiting the cagiest of the losers as, yes,
trusted advisors.
It seems, Montaigne says, that “these good men [the Senators] had
not yet learned of this belle sentence [quoting Virgil]: ‘Trick or courage . . .
between enemies, it doesn’t matter.’ ”26 Virgil notwithstanding, the Flo-
rentines more than a thousand years later would give their enemies, says
Montaigne, a month’s notice before engaging in battle.27 Now, he observes,
we are harder-nosed realists, we are less “superstitious,” and we believe
that he has “the honor in war, who has the gains Where the skin of the
lion doesn’t work, better sew on a bit of the fox’s.” And no time is riper
for deceit than during peace negotiations, especially for those under siege,
who might have to leave the security of their walled fortress to parley:
“and for this reason all military men in our times give voice to the rule
that a governor of a place under siege should never, himself, go outside to
negotiate.” Mistrust and cynicism are current, and the proverbial wisdom
that sustains them is ancient.
’Twas ever thus, whether among Homeric heroes, the virtuous Romans
of the Republic, or the knights at the dawn of the chivalric age. William
of Malmesbury (c. 1120) says of William the Conqueror early in the
Conqueror’s career: “How can I do justice to the incredible courage and
self-confidence he showed in never stooping to a surprise attack, always
naming the day beforehand, as though his proud spirit disdained the nor-
mal practice of our times.”28 Similarly, most people were pretty sharp about
making sure hostages were exchanged before negotiating, but then even
25. Before Jews acquired a reputation for cunning, they first tried taking on the
Romans the good old Roman way, by force. The Romans incurred considerable losses in
finally crushing Bar Kochba’s rebellion in 135 CE. That was the last time a Jewish army
would take the field until 1948. The Punic example notwithstanding, one would almost be
tempted to see a correlation between the attribution of cunning to an entire people and
their lack of an army.
26. Aeneid 2.390.
27. Giving a month’s notice is not a grand gesture against Florence’s interest: though
it may give the enemy time to prepare, it also gives time for at least one political faction
among the enemy to betray their cause or time to undermine the courage of those who
have a month to sleep on the prospect of a fight. The month’s warning, of course, is really
only available to the side that already has the threat advantage.
28. William of Malmesbury, Gesta Regum Anglorum [The history of the English kings],
trans. R. A. B. Mynors (Oxford: Clarendon Press, 1998), sec. 234.
these exchanges could be tricks, as when the hostage you thought you
were getting, a person of high value and closely connected to the com-
mander on the other side, turned out to be one of low value, about whom
they could care less if he got butchered when they violated the truce,
as they had always meant to do. Despite the cynicism and mistrust and
despite the ancient wisdom of wolves donning sheep costumes, guards
were dropped yet again and again.29
Even guard-dropping may not be irrational, for wariness can be exces-
sive. One needs to decide how costly one’s wariness is as against the risk
of passing up golden opportunities, only some of which may turn out to
have been thinly gilded. Yet the deceits that work are lamented precisely
because one was so stupid not to have seen through them, not to have
taken precautions, because in fact the trick would have been so cheap to
insure against, so easy to discover—please make sure the sentry is a natu-
ral insomniac, and do not out go out to parley if you are the leader of the
fortress. Send a proxy, preferably a distinguished man of your entourage
whose wife you covet.
Montaigne continues the discussion in the next essay, giving an exam-
ple from his own neighborhood during the religious wars. One party com-
plained of the treachery of the other, but Montaigne seems to think that
given the norms of the time, each was foolish to have trusted the other.
He comes pretty close to embracing, or at least accepting, the hard truth
that this is how the game is played, so one had better wise up: “For it is not
said that, at a given time or place, we are not permitted to take advantage
of the stupidity of our enemies just as we do of their cowardice.”
If you can slaughter cowardly enemies, why not also the fools who
trust your white flag, who believe your overtures of peace? 30 As regards
the enemy’s gullibility, his trust in appearances, his susceptibility to being
tricked, Montaigne gives the practical view: in “a given time or place,”
stupidity should get no special treatment from our consciences. Exactly
when and where these given times and places are he does not say. He
means to suggest that the privilege granted to take advantage of stupidity
is not as broad as the right to benefit from the enemy’s cowardice, but it
is not much narrower either. Yes, to take advantage of the enemy’s stupid-
ity is ugly business, and we have bad words for it, like deceit, treachery,
guile, fraud, but we also have to contend with winning versus losing; and
the stakes are so high: our very lives, our freedom or our slavery, which
means mostly the enslavement of our wives, mothers, and daughters. So
you pass up the opportunity to steal the war because your opponent was
stupid enough to trust you without taking precautions. Will you gain your
29. Thus, to some, it is naively complacent to seek negotiated assurances from North
Korea or Iran not to build nuclear weapons.
30. Considering no mercy was shown to cowards in one’s own ranks, it was hard to
expect it to be shown to ones in the enemy’s. Culpable stupidity is the sin of commanders.
The private soldier’s stupidity hardly matters, unless he is a stupid, sleepy, and gullible sentry.
enemy’s gratitude, or just his sigh of relief while he thinks you an even
greater fool for not cashing in on the opportunity than he was for so
carelessly providing you with it? One posts sentries as a routine matter
against the deceit of night attack;31 why, too, shouldn’t figurative sentries
be posted during peace parlays or when the enemy comes out waving a
white flag?
TRADE
Deceit was a necessary player in war, but it was not officially supposed to
be there. Deceit ran against war’s high moral grain; it was dishonorable. If
courage alone could be counted on to win, then courage was morally pref-
erable to winning by deceit (unless you were, in Roman eyes, a Greek or
a Carthaginian who preferred to win by deceit even when you could win
by force). War was, by one moral theory, supposed to be a testing ground
of the purest virtue—courage. In fact virtue—virtù—was courage itself,
before the word got expanded and generalized to cover goodnesses of a
more peaceful stripe. Never mind, as noted, that wiser heads were more
than able to work around the official ideology of courage—but I would
be surprised if even those heads were consistently cynical. Odysseus cares
no less than Ajax that people respect his fighting ability, his straight-up
martial skills.
Trade was quite the other way around. Deceit was the name of the
game. And no virtue was held to inhere in chaffering, except ones like
prudence that suspiciously mimicked cowardliness if you scraped away
the makeup and paint. Long before it fell to Marxian academics to sneer
at traders, the role of sneerer was played by aristocrats, from the high-
caste warriors of Homer, to medieval nobles, to fainéant impoverished
lords at fancy watering holes in the nineteenth and early twentieth centu-
ries. Traders have gotten even worse treatment from lowly peasants, who
occasionally rose up and murdered them in pogroms, whether they were
Chinese in Southeast Asia, Germans in the Slavic hinterlands, Armenians
in Turkey, or Jews pretty near anywhere.
There was much self-serving self-deception in the ancient aristocratic
views of the cowardly trader, especially about those traders who were
involved in long-distance trade. A merchant needed to be a tough guy back
then to defend his goods, since the warriors who sneered at him tended
to be, not to put too fine a point on it, pirates; 32 while other traders, your
competitors, were not averse to piracy themselves, and that meant you
31. The Trojan allies whom Diomedes butchered are said to have counted on the
Trojans to post a watch. The children and wives of these allies were safe far away, so they
felt less sense of urgency to post their own sentries, trusting to the Trojans, who had more
at stake. The Trojans, however, were keeping a very lax watch; Iliad 10.415–22.
32. See, e.g., Iliad 1.366.
weren’t either. In the Scandinavian north, it was often hard to tell from
one week to the next whether a Viking ship carried traders or raiders. If
the place they were approaching looked well defended, they put on their
trading hats; if it looked ripe for plundering, they put on their helmets
(not horned, despite nineteenth-century depictions to the contrary).33
What they plundered in one place became the cargo—often women—
they sold in the next. One way or another, raiding involved some trading.
In war, the victim of a stratagem or deceit was “surprised.” Indeed, the
noun surprise first comes into English with the specifically military mean-
ing of being taken unawares by the enemy. 34 In trade, a buyer could not be
“surprised” as a man of courage might be in war, because the buyer
expected to be cheated by the seller, and the seller in his turn to be stiffed
by the buyer. The mutuality of the feeling, however, was not quite congru-
ent. Sellers who purveyed to the nobility could end up stiffed, with the
noble, if he cared to give an answer to someone so beneath him, claiming
that no payment was justified because the seller had cheated him. Some-
times a buyer showed up in force to “buy” the goods with muscle, coming
very close to piracy, but would leave behind a payment at a “fair” price he
determined.35 But let’s leave kings and barons, raiders and long-distance
traders out of it and take instead a simple transaction in the county mar-
ket on market day. There the belief was that sellers held all the cards.
Buyers expected, quite simply, to be tricked; that is what it meant to buy
until only rather recently.
If the satirists and moralists of the fourteenth through the seventeenth
century are to be believed,36 the odds were that the merchant’s scale was
biased in his favor, that he kept one set of weights to measure the goods
bought and another to measure the goods the buyer handed over in pay-
ment.37 Were his goods watered down, adulterated, shoddy? Probably, but
33. See Egil’s Saga, chs. 47–48.

34. OED, surprise sb. 1. Earlier Anglo-Norman French usage of the word was also
military. Surprises were night attacks, ambushes, stealing a march; see, e.g., the early
thirteenth-century L’Histoire de Guillaume le Maréchal, ed. Paul Meyer (Paris: Renouard,
1891), vv. 189, 400–11, cited in Gillingham, “War and Chivalry in the History of William
Marshal,”; (though his citation is off by two hundred lines); see also, notes 5–6 above.
Strickland (War and Chivalry, 41) notes an agreement, c. 1150, between Rannulf, earl of
Chester, and Robert, earl of Leicester, in which it was stipulated that neither would “for
any cause or chance lay snares” for the other unless he gave fifteen days notice. “Guile
and ruse in warfare were thereafter considered perfectly, legitimate, but there was mutual
concern to avoid falling victim by surprise to a damaging ‘first strike.’ ”
35. See my discussion of rán or strong arm purchases in the sagas: Bloodtaking and
Peacemaking (Chicago: University of Chicago Press, 1990), ch. 3.
36. Deceit, fraud, gulling the mark, and hypocrisy make up much of the substance of
Piers Plowman, more than a few of the Canterbury Tales, and Ben Jonson’s comedies.
37. Medieval weights, even the official ones maintained by governmental authorities,
were notoriously inaccurate; for a general introduction, see Bruno Kisch, Scales and
Weights: A Historical Outline (New Haven: Yale University Press, 1965).
then so were everyone else’s. At least everyone thought that such was the
case. Was the meal wormy, the meat tainted? If it wasn’t, the price would
have been still higher. When prices—obeying inexorable laws of supply
and demand—went up, buyers felt sellers were cheating and gouging, but
when abundance drove prices down, buyers gave sellers no credit whatso-
ever. The feeling of being had was programmed into the buyer’s position.
This is not the deceit of a sneak attack during a truce, in which one had
a rosier view than the situation turned out to warrant. In the regime of
brutal scarcity that characterized preindustrial economies, deceit in trade,
unlike deceit in war, was expected, and perceived whether it was there or
not. It was the product of a distinctly unrosy view that characterized any
commerical transaction. Deceit was the devil’s way, and thus the way of
the world, for the fallen world was the devil’s domain. Where there was
plenty, it was possessed not by the just but by the unrighteous.
Several factors helped add fuel to the fire of these suspicions, even
beyond their confirmation by fact:
1. Prices did go down when supplies increased relative to demand,
and then lower prices set a benchmark from which any upward
variation meant deceit.
2. Aristotle, Christianity, and the law (Islam was no better), all self-
deceptively clueless about the time value of money, raised
impossible expectations that loans and credit should be interest
free, that interest was unnatural and a swindle.
3. There was a sense, fostered by the ubiquity of poverty, that
transactions were zero-sum at best. A buyer could see that some
people were doing better than he was, and they were not distant lords
and ladies but people close by who were selling or extending credit
to him. In fact, these sellers were not much better off than he, but he
believed they were getting more than their fair share of the very small
pie. We might condemn this as envy, but given the poverty levels, envy
might have been the form a desperate life-force was forced to take.
4. There were fantasies of plenty, but they just frustrated the buyer.
For if there had been an Eden, and there was somewhere a Land
of Cockayne, he was not about to be an invitee. Better to engage in
petty thievery, secretive and unneighborly, to be a cheat yourself.
Mostly, though, the peasant’s deceit took the form not of taking
from but of keeping from: keeping his best cheese from the tithe-
hungry priest, his largest eggs and fattest hen, if such he had, from
his lord’s bailiff when he came to collect rents in kind. And if the
peasant unloaded his shoddiest goods on the priest and the bailiff,
well, then he, too, could smirk the smirk of the guileful. Such,
such, were the joys.
And what of the seller, who knew that the buyer loathed him? He
might maintain a pretense of honesty; might even be honest and kind and
extend credit to hard-strapped buyers, hoping that when the next bad
harvest came, as it inevitably did, they would refrain from burning him in
his home. Buyers, he knew, refused somehow to recognize that the seller
had himself been a buyer when dealing with his purveyor—that he had
incurred costs. And like any buyer, the seller believed he was being cheated
by his vendor. The baker knew that the prices he charged buyers, who
believed he was cheating them, were a function of what he himself had
to pay the miller, who he in turn believed was cheating him. The laws of
supply and demand were thus both known and not known by the same
person, depending on whether he was situated as a seller or a buyer.
To this day, it takes several undergrad economics courses to come to
understand that when gasoline prices jump, it is not the gas-station owner,
or even some evil corporation, swindling us (though it might be the effect
of the OPEC cartel) but that it is the predictable effect of ups and downs
in supply caused by hurricanes, war, refining capacity, and increased or
decreased demand in China and India. The politician believes that the
voters think like our medieval buyers, so he claims the gasoline purveyors
are gouging. Is the politician deceiving or self-deceiving when he does so?
Like the medieval seller, he seems to have only an intermittent under-
standing of supply and demand. Notice, too, that for us, the politician is
occupying a position the modern seller has largely vacated: that of the
structural deceiver, the sleight-of-hand chiseler, the devil’s familiar—
though we excuse some if they are charming. And no differently from the
medieval buyer, we do little more than grumble and occasionally, but very
rarely, rise up in a modest jacquerie and throw the bums out; mostly we
just get fooled again.
But a miracle happened: if deceit did not quite vacate the domain of
trade, it ceased to be its presiding deity. Finally, buyers’ fantastic expec-
tations of the bargain they felt justly entitled to started to be met for a
widening segment of the population in some parts of the world. We might
start with seventeenth-century Holland or eighteenth-century England,
but by the mid–twentieth century in the now rich Western world, mem-
bers of a very broad middle class, which included significant numbers of
those by other estimations deemed to be working class, accepted sellers,
with no resentment at all, as respectable members of their community.
We now believe in Best Buy, Tesco, and IKEA, and we are not fools for so
believing. We can even return what we have bought and get our money
back at our whim.38
Of course, all that wealth spontaneously generated myriad swindlers
and deceits the costs of which were to be measured in amounts heretofore
unthinkable.39 In an earlier draft of this piece, before the collapse of world
financial markets I could write this: “But with wealth and abundance,
38. For items of greater economic moment, like cars, lemon laws came to the rescue.
39. In the mid–nineteenth century, Herman Melville, no different from a medieval
moralist, could envisage American society as a series of nested and infinitely generative
cons; Melville, The Confidence Man (1857).
people came to think of swindlers and conmen as an aberrant subset of

sellers, a deviation from a new norm: the honest businessman. We in the
middle-class West are rightly more relaxed about buying and selling not
just because goods have improved and competition has made prices rather
favorable to consumers but also because as a general matter, we as buyers
are so far above the margins of dire straits that medieval people and even
our grandparents endured that we can afford to be tricked now and then.
The tapster watering down her ale to clear an extra tuppence at the end
of the day was thus felt as more immediately evil than Bernie Ebbers,
Dennis Kozlowski, Kenneth Lay, and their ilk, who stole untold millions
in the wave of corporate rip-offs of the early twenty-first century.”
There is a lesson to be drawn from my own culpable complacency: that
is, I got fooled again. I got taken in by the classic knaves who had tricked
themselves out in slightly varied versions of the garb of their earlier ava-
tars. And though I want to lead a jacquerie and burn a few summer homes,
I also recognize that like the Trojans before me, I kind of deserved to be
taken in, like not a few of the Icesave depositors got taken in by descen-
dants of Vikings. When a deal looks too good to be true, it is rarely going
to be true, unless you learn the ways of Mr. Ponzi, and get in early and out
before the vast assortment of fools, like the author of this piece, are left
holding worthless paper. And though we may still rightly trust Best Buy, it
may be unwise to trust not only Mr. Potter, but also Jimmy Stewart.
WHO DO WE ROOT FOR?
To go back to the standard trickster stories, in fairy tale, myth, in movies

delighting in cons, big and small: whose side is the audience supposed to be
on? It is not always clear. The Gnostics thought the hero of the Eden story
was the serpent. I bet a good chunk of the time, it is not the fool who got
swindled who gets our sympathy; I bet he very rarely does, unless it is the
self-pity I feel for my own having been snookered. We are not especially
tender toward fools, and as long as he it not a hedge-fund manager or sub-
prime lender, we indulge the rogue who outsmarts the mark by playing to
the mark’s pretensions, vanities, vices. The fool has it coming. And it helps
when the swindler does so with panache and charm, or if he is played
by Paul Newman or Robert Redford. Or even perhaps by televangelist
Jimmy Swaggart, whose talents a smug secularist who is swindled weekly
by his therapist must, despite himself, acknowledge. (That Swaggart and
the therapist both believe in their product may aid the swindle, indeed be
part of the swindle.) And when one of those ridiculous Nigerian emails is
discovered to have hit its mark, though we don’t admire the artlessness of
the scam, we cannot help but be tickled that such transparent mendacity
has worked on some perfect idiot.40
40. See Herzog, Cunning, 61–64, on his correspondence with such an emailer.
But then, we are all ripe to be deceived by others. Our own commit-
ment to good manners requires that we believe the acts others are putting
on. “Social life would not last long if men were not taken in by each other,”
says La Rochefoucauld. Goffman suggests that the primal social virtue of
tact requires that we go along with less than perfect performances, lest
we make a scene.41 Thus we accept as true, if only for the moment, the
pretenses of politeness and feigned pleasure others take in our company,
the excuses for why someone is late or has refused our invitation. This
“going along” is a moral requirement of sorts, and it gives the conman his
wedge even against people who are not fools but are simply doing what
well-socialized people must do. And that is when we turn against him.
We shift our sympathies when the swindler preys on the virtues of his
marks rather than on their vices, when they are plucked because they are
kind, charitable, hospitable, selfless, mannerly (or maybe just trying to
save for their retirement or kids’ college education or retirement). And
we mostly continue to sympathize, even when we begin to think that
that well-mannered and charitable soul who just got taken should have
been a little more alert to the risks of his goodness. If you want to help
real down-and-out characters, better get some street smarts to assist your
charity. And so when he has been fleeced for perhaps the second or third
time, then, though we still think the swindler a worm, we begin to find
the kindly soul’s naiveté, even his goodness, culpably stupid.
Whom we sympathize with seems to be subject to cyclic variation.
Thus the law swings from little sympathy for buyers, caveat emptor, to
aggressive solicitude—caveat vendor—no matter how stupid the buyer
may be. Stick it to the manufacturer; impose strict liability, punitive dam-
ages. Attorneys thus advise manufacturers to put warning labels on their
products. But then the inevitable shift of sympathies sets in, not because
people feel sorry for the nameless shareholders of a big corporation but
because the sufferance of fools can only go so far. My eleven-year-old son
takes great delight in reading warning labels and exploding in laughter:
“Hey dad, look at this package of sliced turkey pepperoni: ‘Warning: do
not eat packet’”—not the packaging, which even a fool knows not to eat,
but the sealed packet of preservative, which apparently looks too good to
some people to pass up. That the commercial showing a guy driving a car
off a cliff has the subtitle “professional stunt driver, do not attempt” tickles
him to no end. And he sits innocently mystified by the erectile dysfunc-
tion commercials during football games that conclude with hastily read
warnings about what to do if your erection lasts three or more days. Now,
if ever a fool had it coming. And damn, if that is not an unintended pun.
41. Compare the deceit that is so artful that the deceiver apparently wants credit
for the excellence of his show as a show, and may blow his own cover to get the proper
recognition of his art. La Rochefoucauld suggests that here, too, it would be bad form not
to fall victim: “some disguised deceits counterfeit truth so perfectly that not to be taken in
thereby would be an error in judgment”; Maxim No. 282.
On the “Existential Positivity of Our

Ability to Be Deceived”
Mark A. Wrathall
Illusory experiences have played and continue to play a significant role

in shaping philosophical accounts of perception. By and large, the need
to account for perceptual errors of various sorts has greased the skids for
the slide into representationalist theories of mind. But the experience of
perceptual errors—illusions, deceptions, and even hallucinations—has
pushed the existential-phenomenological tradition in a very different
direction. When I speak about the “existential-phenomenological tra-
dition,” I mean the tradition of philosophers influenced by Heideg-
ger, Sartre, and Merleau-Ponty. This tradition has its deep roots in
Nietzsche.
Nietzsche insisted that “a perspectival, deceptive character belongs
to existence” (Kritische Gesamtausgabe VII-3.180). At the same time, he
argued that “it is no more than a moral prejudice that truth is worth more
than mere appearance; it is even the worst proved assumption there is in
the world.” Indeed, he believed that when it comes to appearances, we
ought to question the supposition “that there is an essential opposition
of ‘true’ and ‘false’ ”: “is it not sufficient,” he asked, “to assume degrees of
apparentness and, as it were, lighter and darker shadows and shades of
appearance—different ‘values,’ to use the language of painters?” (Beyond
Good and Evil, sec. 34).
For Nietzsche, the world of experience, “the world which matters to
us,” is not an objective state of affairs, but something in which we are
involved and to the constitution of which we contribute. This world, he
argued, “is no matter of fact, but rather a composing and rounding up
over a small sum of observations; it is ‘in the flow’ as something becom-
ing . . . that never approaches the truth; for—there is no ‘truth’ ” (Beyond
Good and Evil, sec. 34).
With these claims—that the world of experience is not an objective
world, that deception belongs to perceptual experience,and that perception
ought not in any event to be thought of in binary terms as “true” or “false”—
Nietzsche prefigured the work of twentieth-century phenomenologists.
67
In this essay, I would like to explore the existential-phenomenological

treatment of the phenomenon of perceptual deception.
Phenomenology adheres to the principle that “everything which is
up for discussion regarding objects must be dealt with by exhibiting it
directly and demonstrating it directly” (Sein und Zeit H. 35).1 Ultimately,
then, phenomenology aims to convince by directing its audience to their
own experience of phenomena, and allowing the “things themselves” as
they show themselves to demonstrate the accuracy of the phenomenolog-
ical description. Thus, in dealing with instances of perceptual deception,
phenomenologists do not base their account on, for example, positing
the existence of hallucinations, understood as nonveridical experiences
of nonexisting objects or events or states of affairs, qualitatively indistin-
guishable from veridical experiences of existing objects or events or states
of affairs. Few, if any of us, ever have such experiences. 2 Instead, phenom-
enologists typically start with the kind of errors we do or can commit
in the normal course of events. For example, while walking through the
park, I walk slowly and quietly to avoid startling a deer on the path ahead
of me, only to discover as I draw closer that the “deer” is a shrub. I bite
into my bagel, which is covered, it seems to me, with a smoked salmon
shmear, and realize after a moment of shock that the pink shmear is actu-
ally flavored with strawberry, not smoked salmon. As I’m walking down
the path, I seem to see a stone ahead, which turns out merely to be a
patch of sunlight on the path. Or finally, we might consider the experi-
ence of a rather special case like Zöllner’s illusion, where “objectively”
parallel lines appear to be converging.
I will refer to such cases in general as “deceptions”—errors produced by
the fact that we don’t simply make a mistake, but rather we are taken in by
the way things present themselves. An issue to consider is whether some
or all of these deceptions are properly categorized as perceptual errors. One
might, for instance, maintain that they should be understood as errors of
judgment rather than perception—that, on the basis of appearances, we
draw a wrong conclusion about the nature of the objects we encounter. As
we will see, existential phenomenologists maintain that such a description
of these experiences is unsupported by the phenomena.
I don’t intend to review or critique nonphenomenological accounts of
perceptual deception in any detail. But before turning to the phenomeno-
logical account, I do want to note a few strategies for categorizing and ana-
lyzing such phenomena that the phenomenological tradition would reject.
1. Page citations from Sein und Zeit refer to the “H” numbers, which are based on the
pagination of the original German edition (Verlag Max Niemeyer, 1927), and which can be
found in the margins of both English language translations of Being and Time, as well as in
the margins of the Gesamtausgabe edition of Sein und Zeit (Klostermann, 1977).
2. Thus, as Komarine Romdenh-Romluc points out, even when addressing cases
of hallucination, Merleau-Ponty draws on “actual cases of hallucinatory experience” as
described in the clinical literature. See “Merleau-Ponty’s Account of Hallucination,”
European Journal of Philosophy (forthcoming).
On the “Existential Positivity of Our Ability to Be Deceived” 69
First, one might be tempted to draw a sharp distinction between verid-

ical and nonveridical experiences, and to reserve perceptual categories
(“seeing,” “hearing,” “smelling,” “tasting,” “feeling,” etc.) for those cases in
which we succeed in grasping things as they objectively are. When I look
at Zöllner’s illusion, for example, it doesn’t seem right to say of me that
I see converging lines, even though they look like they’re converging to
me. Or when I mistake a bush for a deer in the park, it doesn’t seem right
to say of me that I see a deer, even though it looks like a deer to me. So
one might feel compelled to draw a clear distinction between things look-
ing a certain way, or our experience having a certain phenomenal charac-
ter, or the mere appearing of things, and a genuine perceptual experience.
Or, more precisely, one might feel compelled to treat the mere appearing
as a genuine perceptual experience only if it is veridical. (Allowing for the
possibility of deviant causal chains, one would have to say that veridicality
is a necessary but not sufficient condition of a genuine perceptual experi-
ence.) In the genuine perceptual experience, the phenomenal character of
things corresponds to the way things actually are. One then accounts for
deceptions by treating them as the presentation of a certain phenomenal
character in the absence of the objects necessary to make that presenta-
tion true.3
This points us to a second temptation—that of assuming that there is
some determinate, objective fact of the matter about the character of the
things in the world that we perceive.4 Of course, this is a hard assump-
tion to avoid making—it seems that there either is a deer in the woods on
the path in front of me or there is not. It either is a salmon schmear on the
bagel or it is not. We successfully perceive things only if the way things
seem agrees with the way things objectively are.
And this, in turn, points us to a third temptation—the temptation to
treat our experiences as if there is a determinate fact of the matter about
what we are experiencing, as if it is possible to specify, at least in principle,
how it is that the world seems to us to be.
I suspect these three temptations hang together and reinforce one
another. It is only because we believe in a set of determinate, objective
facts about the perceived world, and only because we believe that the
way the world seems to us is equally objective and determinate—that it
3. It is not at all clear how this move can accommodate the many, perhaps prevalent,
cases in which I perceive something slightly wrongly. I buy a green tie to go with my green
suit, for example, only to find when I get home that the tie was in fact brown. What did I
see in the department store? I saw a tie—there can be no denying that much. But it seems
wrong to say that I saw a green tie, given that the tie was brown. And yet, if I had seen a
brown tie, I wouldn’t have bought it.
4. Resisting this temptation would not necessarily require one to deny that there is
some determinate, objective fact of the matter about the makeup of the physical universe.
But it would require one to acknowledge a possible distinction between the physical
universe—what Merleau-Ponty sometimes refers to as the “world in itself” (see, e.g., PP 10
and 39) and the perceptual world.
makes sense to treat the success or failure of perception as a matter of

truth or falsity.
These temptations also might lead one into what I would call an
unequal division of labor in accounting for perceptual deception. By this
I mean that the responsibility for the deception tends, unjustly, one might
suppose, to fall on the deceived party. When there is a mismatch between
the way the world seems to us to be and the way the world actually is, we
are at fault. One reasons, for instance, that we have drawn a false infer-
ence from the evidence about the world with which we are presented in
sensation, or that we have hastily judged that such and such is the case on
the basis of flimsy evidence. But what makes cases I’ve described instances
of deception as opposed to mere error is the sense that the deceived party
didn’t really do anything wrong. One’s perceptual systems may have been
working properly. One may have been proceeding with due care. And yet
one gets taken in.
The existential-phenomenological approach, however, does not find
itself tempted by the experience of deception to think about percep-
tion in these ways. Indeed, the phenomenology of deception is actually
thought to reinforce our ability to resist these temptations. In particular,
as I hope to show in what follows, deceptions such as these help one
to see perception as having not binary success conditions, but of suc-
ceeding to greater or lesser degrees—one can see the scene in better or
worse ways. But it rarely makes sense to say that I either perceived truly
or falsely. Second, deception helps us to recognize that the perceptual
domain is not the objective universe of physics. And finally, it helps us
recognize the indeterminate quality of our experience of the perceptual
domain.
THE PHENOMENOLOGICAL ACCOUNT

OF PERCEPTUAL DECEPTION
The starting point for the existential-phenomenological account of per-

ceptual deception is the recognition of what Heidegger calls in Being
and Time the “existential positivity of our ability to be deceived” (Sein
und Zeit, H. 138). The point is that deception doesn’t show a momen-
tary failing or accidental shortcoming in us, but rather points the way to
understanding something fundamental about us, the world, and our rela-
tionship to things in the world. As Heidegger explains, “every deception
and every error” should be seen “as a modification of original being-in”
(Sein und Zeit, H. 62). By this, Heidegger means that errors and decep-
tions are not mere mental events, nor do they consist in the possession of
false representations about the world. Instead, they are particular ways
of being out in the world and involved with things. In a related man-
ner, perception itself is not “measured against the idea of an absolute
knowledge of the world” (Sein und Zeit, H. 38)—that is, Heidegger denies
that veridicality, the measure of knowledge, is an appropriate category

for thinking about perception. Heidegger, for example, tends to speak
of “genuine” and “deceptive” perceptions (Echt- and Trugwahrnehmun-
gen), rather than “true” and “false” perceptions. This, in turn, leads to the
view that deception shows us something essential about the nature of
the world and the things we encounter in the world—namely, that they
are not objective and determinate. “It is precisely in the unstable seeing
of the ‘world,’ a seeing that flickers with our moods, that the available
shows itself in its specific wordliness, which is never the same from day
to day” (Sein und Zeit, H. 38).
Merleau-Ponty agrees. Writing in the context of thinking about hallu-
cinations (although the point applies broadly), he notes:
all the difficulties arise from the fact that objective thought, the reduction
of things as experienced to objects, of subjectivity to the cogitatio, leaves no
room for the equivocal adherence of the subject to preobjective phenom-
ena. The consequence is therefore clear. We must stop constructing halluci-
nation, or indeed consciousness generally, according to a certain essence or
idea of itself which compels us to define it in terms of some sort of absolute
adequation. ([PP] 336)
The experience of deception points us toward the unsteady, flickering

nature of the perceptual world, to the equivocal experience of “preob-
jective phenomena,” for were experience always clear and the world of
perception populated with determinate objects, we would not be taken
in by deceptive appearances.
Before going on, I should emphasize the tendentious nature of these
existential-phenomenological claims. For many, sense experience is to be
measured in the same way cognition is. It is either true or false, and it is
true by, to put it loosely, representing the way that the world is. Only
true sense experiences can qualify as perceptions. Deceptions, illusions,
hallucinations fail to represent the world, and therefore, there is no posi-
tive role to be played by perceptual deception in disclosing the world to
us. The source of the error must, therefore, be traced somehow back to
us—for example, an error of judgment, a false conclusion drawn from the
evidence of the senses.
So the existential phenomenologist can’t rest content with this descrip-
tion. We must confront the question: how does existential phenomenology
account for error? If we’ve abandoned the thesis of an objective, deter-
minate world, what basis is there for distinguishing between successful
and unsuccessful experiences of the world? And if not veridicality, than
what is the criterion for success? To answer these questions, I want first
to reconstruct some paradigmatic existential-phenomenological descrip-
tions of deception. I’ll then consider how it is that, the existential phe-
nomenologists suggest, these descriptions help us to resist the temptation
to think about our perceptual encounter with the world in the three ways
outlined earlier.
I turn first to Heidegger’s account of deception, offered most extensively

in two Marburg lecture courses: the 1923–24 course in Gesamtausgabe vol-
ume 17 (GA 17): Einführung in die phänomenologische Forschung, and the
1925–26 course in GA 21: Logik. Die Frage nach der Wahrheit. Let’s look
first at the kind of example Heidegger draws on. Heidegger writes5: “I am
walking in a dark forest and I see between the fir trees something coming
toward me—‘a deer,’ I say. The assertion does not need to be explicit. Upon
coming closer it turns out that it is a shrub, toward which I am heading”
(GA 21:187). How are we to understand this error? What allows me to be
deceived by the shrub? First of all, Heidegger emphasizes that the error
is not simply one of having said the wrong thing about what I have seen,
or having wrongly judged that there was a deer between the trees. Rather,
my fundamental error, he says, is that I have “comported myself in such a
way as to cover up” (GA 21:187). Heidegger uses the term “comport,” to
carry oneself or behave oneself, in order to emphasize the primarily practi-
cal dimension of our perceptual engagement with the world. Perceiving
wrongly isn’t believing something false, for Heidegger; it is acting in the
world in such a way that the true nature of things is covered up.
Heidegger proposes that there are three “structural conditions” of our
everyday comportment in the world that we need to focus on in thinking
about deception. The suggestion is that it is the very conditions of our
ordinary engagement with things in the world that makes us susceptible
to being deceived.
The first structural condition of comportment that Heidegger analyzes
is the fact that our comportment has an inherent “tendency to discover
something” (GA 21:187), and does this on the basis of “the always already
prior disclosure of the world” (GA 1:187). By this he means that we are
always already poised for things to show up to us, and we encounter them
as meaningful things in terms of our understanding of our world. So as
I walk through the park in the dark, my skills for park-walking are acti-
vated. As Merleau-Ponty puts it, “if there can be, in front of [my body],
important figures against indifferent backgrounds, this occurs in virtue of
its being polarized by its tasks, of its existence towards them, of its col-
lecting together of itself in its pursuit of its aims” (PP 101). So the first
thing that sets us up to be deceived is the way we are always disposed or
primed, through the aims implicit in what we are doing, to find things in
such and such a way.
5. I should note that in these passages, Heidegger’s ultimate goal is to understand how
it is possible to say something that is deceptive, rather than something that is simply false.
A false assertion need not be deceptive if it couldn’t possibly induce you to believe it.
So Heidegger tackles the problem of the lie by first asking how it is that we can perceive
erroneously, since it is such a perception that ultimately makes the lie believable. I note
this only because Heidegger introduces language into the discussion at certain points, and
I’m going to completely ignore those for my purposes. I don’t think that by systematically
ignoring that side of Heidegger’s analysis I’m doing any violence to his account of
deceptive perceptual experiences.
This leads us to the second structural condition of our comportment. This

condition has to do with the kind of entities we encounter in our everyday
dealings in the world: “the entity itself must have its being constituted in
such a way that, as the entity that it is, it offers and calls for the possibility
of a togetherness with others, and it does so on the basis of its being. That
is, it only is what it is in the unity of such a togetherness” (GA 21:185). The
entities we are primed or disposed to discover in comportment are entities
that are not what they are in themselves alone, irrespective of the relation-
ships they bear to other entities. Instead, entities are what they are holisti-
cally in virtue of the way they exist together with other entities. The classic
example of this is Heidegger’s ubiquitous hammer: the hammer is what it
is only because of the way it relates to nails and boards. The “togetherness”
that Heidegger mentions is, I take it, the meaning or significance of a thing,
where to be meaningful is to lead those who grasp the meaning from one
thing to another. An entity is the entity it is in terms of the way it directs us
to the context of other entities and activities within which it belongs. The
togetherness is, in turn, makes an entity the thing it is only to the degree
that it “offers” and can “call for,” that is, affords6 and solicits, us to be directed
from the entity to the things and activities with which it is involved. The
world is the organized totality of such relationships of offering and calling
for us to move from one thing and one situation to the next. And something
only is an entity insofar as it presents us with a “unity of togetherness,” that
is, shows up as holding a more or less coherent and organized place in such
a meaningful structure. We comport ourselves in the world by responding
to the significations that the world affords and solicits.
Together, the first and second structural conditions mean that we
always encounter the things in the world in terms of something else. We
never encounter something that is meaningless: “in the field of everyday
experiences, I don’t just stand there—for example, in the forest—and
simply have something before me. That is a purely fictitious situation.
Instead, I am always encountered in an unexpressed way by something
that I already understand, something that is laid out in advance as some-
thing, and that in this way is accepted and expected in the comportment
of coping with the world” (GA 21:187). So when I mistakenly see a deer,
for instance, it is because certain features of the scene in front of me draw
on my abilities to identify and respond to deer solicitations.
Finally, the third structural condition Heidegger identifies is the fact
that within the range of possible significations in terms of which we
6. Although I borrow the language of affordances from J. J. Gibson, there is one

important difference between Heidegger’s notion of what the world offers and Gibson’s
notion of environmental affordances. For Gibson, an affordance is a physical fact about
what the environment “offers,” “provides,” or “furnishes” an organism of such and such
a type. See Ecological Approach to Visual Perception (Hillsdale, N.J.: Lawrence Erlbaum
Associates, 1986), 127. For Heidegger, however, affordances for Dasein—the kind of
beings we humans are—are world dependent. That is, is a function of not just the kind of
organism we are but also our way of being in the world.
encounter things, the situation within which we find ourselves disposes

us to respond to certain solicitations rather than others. In the forest, for
example, nothing could solicit us to see “the cubed root of sixty-nine
coming towards me” (GA 21:188). Even though it is logically possible
that we could see the Shah of Iran coming through the forest, we will not
be motivated to see this in the Black Forest of Germany either (see GA
21:188). But both deer and shrubs are live possibilities.
To briefly review, then, Heidegger observes that our ordinary ways of
engaging with the world have the following structural conditions:
1. We are always poised to have meaningful entities show up for us;
2. these entities are meaningful insofar as they offer us a certain way
of relating them to other entities and activities (they present us
with affordances), and, in fact, they also call for us to follow up
those affordances (they solicit us to act on the affordances); finally,
3. the world presents us with a meaningful context of entities and
activities that disposes us to encounter some things but not others.
These conditions are not just the conditions of everyday comport-
ment—the conditions under which we are able to smoothly and fluidly
deal with things. They are also the conditions which make it possible for
us to be deceived by things. How so?
Consider the example of the salmon schmear. It is because I ordered
a salmon, not a strawberry schmear, and because, in the context of bagel
shops, one’s order is generally fulfilled, that I am primed for my bagel to
come with a salmon schmear. The pinkish color of the schmear in that
context leads me to anticipate the fishy flavor of a salmon schmear. But it
is also the case that the significations in the context lead me to experience
the color in a particular way (in fact, once I realized that I had the wrong
schmear on my bagel, the color thereafter looked strawberry pink, not
salmon pink). So the deception arose through a confluence of my disposi-
tions, the world-context, and the color of the entity itself, all conspiring
to indicate the existence of something that wasn’t there. But the decep-
tion was also uncovered as such through the course of further perceptual
comportment—it was the sweet, creamy strawberry flavor that changed
the way I was disposed to see the color and, consequently, let me see the
schmear for what it was.
But, as Heidegger points out, there is a distinction between a percep-
tual error and merely failing to see something—between, for example,
seeing the bush as a deer, and not seeing the bush. This distinction paral-
lels the distinction between calling someone by a pseudonym and calling
him or her by the wrong name.7 A pseudonym is “a designation behind
7. Heidegger thinks that we need to understand this to grasp the Greek notion of
the pseudos, the false. “This is the fundamental meaning of the Greek pseudos: to so twist
something that one does not see how it genuinely is. Pseudos is that which distorts and
twists” (GA 36/37:227).
which the author hides, an alias that covers him up” (GA 36/37:227). It’s
not false in the sense that there is nobody who answers to it. To the con-
trary, the pseudonym directs one to the author of the book, but does it in
such a way that “one does not see how he or she genuinely is.” Likewise, in
a perceptual error, we do see something (it is not a hallucination). But we
see it in such a way that it doesn’t show itself as it genuinely is. “Pseudos is
a showing that passes something off as something; thus it is more than a
mere covering up without passing it off as other than it is” (GA 17:32).
The discussion of the structural conditions of perception, moreover,
lets us recognize that the possibility of perceptual deception is built into
the very structure of our world. It is a “basic fact, in the sphere of dealing
with the world,” Heidegger insists, that error and deception “are inter-
woven in a completely fundamental way, and do not merely occur as
defective properties that one must overcome” (GA 17:39). Heidegger
thus offers a more equitable division of labor, attributing the blame for
the deception to the world and to the things in the world as much as to
our way of comporting ourselves in the world. It could be the case, of
course, that we are primarily responsible for the error, insofar as we might
respond wrongly to the solicitation. We might, for instance, lack the skills
to respond appropriately to what the situation calls on us to do. It might
be that I would be more susceptible to being deceived by the bush than a
deer hunter would—he probably has much better skills for distinguishing
deer from other things that might suggest a deer. At least, given that he
goes looking for deer with a loaded gun, I hope he has better skills. But,
even in this case, my deception is motivated to a considerable degree by
the skills I have and use effectively in coping with this sort of context.
As Heidegger puts it in the 1923–1924 lecture course, “the possibility
of deception lies . . . in an erroneous seeing, which is not motivated by
a careless consideration, but rather in the manner in which the existing
[human] being lives and encounters the world itself” (GA 17:36).
Thus we are not solely responsible for being deceived. It is also the case
that, at least sometimes, things in the world conspire to lead us to per-
ceive them wrongly: “there are entities that in their specific being have the
characteristic that they pass themselves of as something that they are not,
or as so characterized as they are not—where the possibility of deception
thus does not lie primarily in a wrong way of taking them up, but rather
in the entity itself ” (GA 17:32). He goes on to explain:
the things can elude us, and that is not to say they disappear. The elusiveness
of things comes to life by virtue of the fact that we encounter them circum-
stantially. We do not see a thing as . . . an object of scientific investigation.
This existence of things is much richer and offers much more fluctuating
possibilities than are thematically prepared. Because the world in its rich-
ness is only there in the particular concreteness of living, the elusiveness
is also much more encompassing and, with it, the possibility of deception is
there. The more concretely I am in the world, the more genuine is the exis-
tence of deception. (GA 17:37)
I think that Heidegger is pointing here to the way real entities and
contexts necessarily orient us toward more than can possibly be present
to us at any given moment. Merleau-Ponty makes a similar point in noting
that vision is “an operation which fulfills more than it promises” (PP 377).
For instance, when I see the facade of a house, I am oriented already to
the back and sides of the house. My vision of the front “promises” an
experience of the other sides. But the experience of seeing the other sides
is always much richer than what the promise prepared me for. So, like
Heidegger, Merleau-Ponty sees in the present experience an orientation
toward much more than can be presently experienced. Thus, perception
throws me open to a world, but can do so only by outrunning both me
and itself. Thus the perceptual “synthesis” has to be incomplete; it cannot
present me with a “reality” otherwise than by running the risk of error. It is
absolutely necessarily the case that the thing, if it is to be a thing, should
have sides of itself hidden from me. (PP 377)
We are thus always both open to the world and lacking certainty about it.
The lack of certainty is not a negative but a positive—it is that in virtue
of which we can understand and intend more than is present to us at any
given moment.
RESISTING TEMPTATION
Unfortunately, Heidegger doesn’t develop his view of perceptual decep-

tion much further. But in this final section, I would like to look at the
implications of acknowledging the positive character of deception, and
to hazard some preliminary suggestions about the lessons existential phe-
nomenology has drawn from the experience of perceptual deception.
I will focus, in particular, on Merleau-Ponty, to see how his account stands
with respect to the temptations I discussed at the outset. But the sum-
mary of Merleau-Ponty’s views that I offer here will be very tentative.
I will present his view as a loose collection of theses about the lessons to
be drawn from the experience of being deceived, cognizant that much
work remains to be done in order to provide a coherent theory of percep-
tual deception.
With a suitable description of the experience of deception in place,
we can begin to ask: how must we, the world, and our relationship to the
world be if we are to experience deception in this way? As I see it, the
key features of the description are the following. When we are deceived,
it’s because the thing really looks like what we take it as. At the same
time, things will look differently once the deception is uncovered. And
the deception is uncovered in the course of further perception/action/
exploration of the world.
Let’s look at each of these features of the description in turn, and see
what lessons are to be drawn from them.
For one thing, in many ordinary cases of perceptual deception, we are

deceived because the thing we mistakenly perceive really does look like
or sound like or taste like or feel like something else. The bush in the for-
est does, from such and such a vantage point, and in such and such light,
look like a deer. The strawberry schmear does look in many respects like
the salmon schmear. This is in direct contrast to some traditional modes
of thinking about deception—modes Merleau-Ponty calls “sketchy rea-
soning.” If we start not with an appreciation of the positive character of
deception, but instead with an assumption that deception is a kind of
negation, a departure from the objective world as it determinately pres-
ents itself to us, then the tendency is to see deception as the result of our
erroneous contribution to what is truly given in experience. There is not,
in fact, a deer on the path. And thus, the “sketchy reasoning” goes, we
must associate what is there with some memory of or past experience of
a deer. So the deception, on this account, is the result of the contributions
of memory to what is actually experienced.
But, Merleau-Ponty points out, this way of thinking about deception
in fact fails to accomplish what it sets out to, because the present expe-
rience must already have “form and meaning,” it must already look like
something, in order to call forth just these memories as opposed to others
(see PP 20–21). But that means that, in order to call forth the memory
of a deer to make the bush seem like a deer, for example, the bush must
already look like a deer. Otherwise, there is no reason why we would see
it as a deer as opposed to a gorilla or the shah of Iran, or anything else.
Indeed, it is this looking like a deer that makes the deception deceptive—
it “passes itself off as genuine perception precisely in those cases where
the meaning originates in the source of sensation and nowhere else.” If
that is so, then the supplement of memories comes too late to explain the
deception (PP 20).
The phenomenology of deception, then, points us to the inherently
meaningful structure of the perceptual world; indeed it expands our
understanding of it. It shows up as unmotivated the belief in a meaning-
less stratum of sensations, to which meanings subsequently are attached.
Merleau-Ponty illustrates this through a discussion of Zöllner’s illusion, an
optical illusion in which parallel lines are made to seem to be converging
(fig. 4.1).
For Merleau-Ponty, it is wrongheaded to start from the assumption that
the lines must actually be given in perception as parallel, and then to try
to explain how the lines end up being experienced as converging. Instead,
the interesting question to ask about this illusion is
How does it come about that it is so difficult . . . to compare in isolation the
very lines that have to be compared according to the task set? Why do they
thus refuse to be separated from the auxiliary lines? It should be recognized
that acquiring auxiliary lines, the main lines have ceased to be parallel, that
they have lost that meaning and acquired another, that the auxiliary lines
introduce into the figure a new meaning which henceforth clings to it and
Figure 4.1.
cannot be shifted. It is this meaning inseparable from the figure, this trans-
formation of the phenomenon, which motivates the false judgment and
which is so to speak behind it. (PP 35)
Illusions and other such deceptions point up the fact that we first of
all encounter meaningful structures, and we encounter particular enti-
ties in terms of their meanings. Something is meaningful when it leads
one who grasps the meaning beyond what is presented. To say that per-
ception is meaningful through and through is to say that there is noth-
ing experienced in perception that is absolutely and fully given in the
present; everything we perceive directs us beyond itself, attunes us to
anticipate further experiences. A color leads us to anticipate a modula-
tion of color as lighting conditions change. A shape or form leads us to
anticipate further adumbrations of the form as it moves relative to us.
Thus, what everything is is experienced in perception in virtue of what
Merleau-Ponty calls “the mode of existence and coexistence of perceived
objects . . . the life which steals across the visual field and secretly binds
its parts together” (PP 35).
Given the inherently meaningful structure of perception, it follows that
there is no particular thing about which we might not be deceived. There
is no bedrock component of our experience about which we couldn’t get
it wrong, because everything has the meaning that it has only in virtue
of the structure of meanings that solicit us to further exploration.8 The
8. This, incidentally, suggests the incompleteness of Husserl’s account of the

experience of perceptual deception as an “explosion” of the perceptual noema as new
“perceptual data” are experienced that fail to fit with preceding noema. What this story
doesn’t account for is the way the character of the perceptual data themselves changes
along with the “noema.”
schmear example illustrates this—the perceived color of the schmear var-

ies along with my expectations about the taste. Or consider the example
Merleau-Ponty introduces when making this point—the light patch on
the path that is mistaken for a stone. “Every sensation is already pregnant
with meaning,” he observes, “and there is no sense-datum which remains
unchanged when I pass from the illusory stone to the real patch of sun-
light” (PP 297). What before looked to be a broad, flat stone with a dif-
ferent color from the surrounding earth showed itself to be a differently
lighted patch of dirt of the same color. Perhaps what seemed to be a
shadow cast by the stone might now be seen as a darker gravelly patch.
Such experiences call into question the idea that there is an objec-
tive, stable, determinate perceptual world. If we suppose that there are
an indefinite number of meanings to which we could be attuned, and we
recognize that different attunements will result in different experiences
of the perceptual field, then we will have to conclude that there is no
final, objective fact of the matter about what is given to us in perception.
And, indeed, Merleau-Ponty argues that this kind of indeterminacy in the
perceptual world is a condition of our being deceived perceptually. Only
if the world has room for and accommodates deceptive as well as correct
perceptions, only then is it be possible to be deceived, since the deception
presents itself as accurately opening us up to the world. This means that
the world must be something more than all that is the case; it must be
rather a setting: “the world is not a sum of things which might always be
called into question, but the inexhaustible reservoir from which things
are taken” (PP 344):
In the very moment of illusion this possibility of correction was presented
to me, because illusion too makes use of this belief in the world and is
dependent upon it while contracting into a solid appearance, and because in
this way, always being open upon a horizon of possible verifications, it does
not cut me off from truth. But, for the same reason, I am not immune from
error, since the world which I seek to achieve through each appearance, and
which endows that appearance, rightly or wrongly, with the weight of truth,
never necessarily requires this particular appearance. (PP 297)
But this is not to say that anything goes. How are we to preserve the
distinction between deceptive and genuine perceptions, once we grant
that there are an indefinite number of different ways to perceive any given
perceptual field? We start from the notion of the inherent meaningfulness
of perception. This means that to perceive is to be drawn into or pointed
toward paths of further perceptual exploration and action. The distinc-
tion between genuine and deceptive perceptions is found in the degree to
which they lead us well, in the sense that they allow us to keep our grip
on the world around us. As Merleau-Ponty puts it,
my perception brings into coexistence an indefinite number of perceptual
chains which, if followed up, would confirm it in all respects and accord
with it. My eyes and my hand know that any actual change of place would
produce a sensible response entirely according to my anticipation, and I can

feel swarming beneath my gaze the countless mass of more detailed per-
ceptions that I anticipate, and upon which I already have a hold. (PP 338,
translation modified)
In the genuine perception, then, the perception is followed up with and

confirmed by further perceptions that were already anticipated in terms
of the meaning of the genuine perception. With a deceptive perception,
by contrast, what I am led to anticipate by the perception is not encoun-
tered in the perceptual field: “my body has no grip on it, and . . . I cannot
unfold it before me by any exploratory action” (PP 295).
It is thus further perceptions—perceptions that restore our grip on the
world—that annul the deceptive perception, and show it for the decep-
tion it was.
I place my confidence in the world. Perceiving is pinning one’s faith, at a
stroke, in a whole future of experiences, and doing so in a present which
never strictly guarantees the future; it is placing one’s belief in a world.
It is this opening upon a world which makes possible perceptual truth
and . . . thus enabling us to “cross out” the previous illusion and regard it as
null and void. (PP 297)
But now the fact that we can be deceived in perception, and yet nev-
ertheless correct our being deceived through further perception, shows
something important about the relationship in which we stand to our
perceptual experiences—namely, that “the percept is and remains, despite
all critical education, on the hither side of doubt and demonstration”
(PP 344). It’s important to attend to the nuances of this claim: Merleau-
Ponty is not claiming that I’m always correct about what I perceive.
Rather, that in the act of perceiving, my perception is not in the game of
being true or false. I can’t be mistaken in my perception in the sense that
what I perceive is false. But my perception is nevertheless correctible in
the sense that a prior perception can be “cancelled” or “crossed out”—we
come to recognize that the way we were seeing the world was not opti-
mal, given the practical aims implicit in our mode of engagement with the
world. “I say that I perceive correctly when my body has a precise hold on
the spectacle, but that does not mean that my hold is ever all-embracing”
(PP 297)—that is, for any given perceptual hold on the world, we could
recognize that other holds are possible, that this way of getting to grips
with the world has not come to terms with everything in the world, that
other ways of engaging the world might be more or less successful, or
guided by different concerns.
This view of perception will seem paradoxical as long as we think of
the success conditions of perception in the same way we think of the
success conditions of belief. But the paradox dissolves when we see per-
ception instead in terms of action, practical engagement with the world.
If I am pouring water into a glass, we don’t say that my way of gripping
the pitcher and holding the glass is “false.” It might be a mistaken way of
pouring the water in the sense that it will lead me to spill the water. And
there are undoubtedly better and worse ways of holding the pitcher and
the glass. But success here is not a matter of our grip conforming to an
ideal grip—it’s a matter of the action unfolding itself in such a way that it
allows me to achieve my goals in the world.
And this, in turn, suggests that it is wrong to think of perception in
terms of the possession of propositional contents. To see that there is a
stone on the path is not necessarily to have a particular attitude toward
the propositional content: there is a stone on the path. “I see the illu-
sory stone,” Merleau-Ponty argues instead, “in the sense that my whole
perceptual and motor field endows the bright spot with the significance
‘stone on the path’. And already I ready myself to feel under my foot this
smooth, firm surface” (PP 297, translation modified). I am, correspond-
ingly, deceived in seeing the stone if, for example, the resulting bodily
attitude causes me to stumble, or to change directions into a less optimal
path.
ACKNOWLEDGMENTS
A number of people have posed challenging questions and offered helpful com-
ments in response to earlier drafts of this chapter. I am particularly indebted
and grateful to Bert Dreyfus, Charles Siewert, Wayne Martin, Sean Kelly, Taylor
Carman, Iain Thomson, Bill Blattner, and Stefan Käufer for the fascinating discus-
sions this article occasioned.
Self-Deception, Deception, and the Way

of the World
David Sherman
A good deal of progress has recently been made toward resolving some
of the more troubling of the so-called paradoxes of self-deception. Two
interrelated paradigmatic shifts, in particular, are worth mentioning in
this regard. As an initial matter, the paradox that self-deception requires
one simultaneously to believe and not believe the truth about which
one is self-deceived is in the process of unraveling, for the assump-
tion on which it is based, that (intrapersonal) self-deception should
be understood on the model of (interpersonal) deception, is increas-
ingly being called into question. Arguments against this terminologi-
cally driven view, characteristically referred to as “the lexical approach,”
have rightly rejected the notion that there is something inherent in the
phenomenon of self-deception that requires the self-deceived person
simultaneously to believe and not believe the truth, 1 even if the (rather
misleading) term “self-deception” itself seems to presuppose this duali-
ty.2 Moreover, and, perhaps, more fundamentally, given that it seems
to be presupposed by the lexical approach, the paradox that self-de-
ception undercuts itself by virtue of the very condition of its possibil-
ity (i.e., how can I deceive myself if I deliberately set myself to the
task of doing so) is in the process of unraveling, for the assumption on
which it is based, that self-deception should be understood as an inten-
tional activity, is also increasingly being called into question. Arguments
against “the intentionalist approach” have adopted the notion that self-
deception should be understood as “motivated irrational belief,” and that
1. See Alfred R. Mele, Self-Deception Unmasked (Princeton: Princeton University Press,

2001).
2. Richard Moran justly says that there are “basic differences in the logic and
consequences of first-person and third-person attitudes,” and that with many self-reflexive
phenomena, including self-deception, there are “limits on modeling some of the relations
to oneself on the possibilities of relations to others.” See Authority and Estrangement: An
Essay on Self-Knowledge (Princeton: Princeton University Press, 2001), xxix–xxxiii.
82
Self-Deception, Deception, and the Way of the World 83
the motivating desires that skew the process of belief formation can go
all the way down.3
Despite this progress, problems remain. The social component of self-
deception, in particular, has been badly neglected, as much of the
literature on self-deception implicitly bears the mark of a monadic pre-
dilection, and this neglect has resulted in a failure to fully appreciate
the nature of the phenomenon. Many of our most basic beliefs, the ones
that fundamentally orient us toward the world, are socially constructed
and supported, and these basic beliefs are what often motivate “moti-
vated irrational belief.” Even Leibniz’s “windowless” (rational) monads,
which might be understood to form irrational beliefs on the basis of first-
order motivations caused by the internal interplay between (dynami-
cal) appetition and their distinctive underlying states (“perceptions”), are
ultimately “mirrors of the universe,” and all of their actions unswervingly
reflect the second-order motivation that is God’s “preestablished har-
mony.” Unhappily, this metaphysical account might not be the worst one
for making sense of self-deception in contemporary society. One need
not probe particularly deeply to come to the conclusion that beneath a
nominal individualism, which would understand all of our belief-skew-
ing motivations as internally generated, there is an all but ubiquitous
social totality that externally constructs these belief-skewing motivations
and passes them off as the individual’s own: second-order motivations
thus come off as first-order ones, and discerning what, if anything, truly
remains of the latter becomes exceedingly problematical. As a result, we
might be inclined to careen from pillar to post, from an unremittingly
individualistic account of self-deception to an unremittingly social con-
structivist one, but this inclination, too, should be resisted. Viewing “the
self,” with its desires and beliefs, as nothing more than a fiction produced
by overarching relations of discourse or power, which is how the social
constructivist views matters, makes no more sense of the phenomenon
of self-deception than individualistic approaches do. Depending on the
social constructivist, either self-deception is understood wholly in terms
of the larger society, regardless of that society’s own internal coher-
ence, or it is understood as an all-pervading phenomenon, given that the
3. It is with this last turn of the screw that Ariela Lazar distinguishes her view from
Mele’s in “Deceiving Oneself or Self-Deceived? On the Formation of Beliefs ‘under
the Influence,’ ” Mind 108 (April 1999), 265–90. Two points must be made here. First,
although Lazar does not entirely rule out the applicability of the intentionalist approach in
all cases of self-deception, which I take to be a mistake (if we understand “intentionalist”
in as robust a sense as she does), she is right to contend that if the intentionalist approach
does not apply “the requirement that the process of self-deception originates in the
subject’s holding the rational belief becomes obsolete” (270 n. 10). Second, in seizing on
the notion of “motivated irrational belief,” Lazar draws on the work of David Pears, one
of the first to make this distinction, but she goes beyond Pears (and Mele) by largely (if
not entirely) rejecting the intentionalist approach, which is not necessarily inconsistent
with the notion of “motivated irrational belief” (268 n. 7). See David Pears, Motivated
Irrationality (Oxford: Clarendon Press, 1984).
societal conventions that overrun “the self” are also intelligibility’s very
condition of possibility.
The individualist and social constructivist models of self-deception
each manifest a one-sided view of self-deception, and, to better under-
stand the phenomenon, it is necessary to view it as a dialectical product
of both viewpoints, which are its constituting poles. To pose the question
of self-deception as an “either/or”—that is to say, in either individual-
istic or socially constructed terms—is to falsify it from the very begin-
ning. Indeed, to pose the question in this way is to lose the phenomenon
itself: on the individualistic approach, self-deception becomes impossible,
and, on the socially constructed approach, it becomes either relativistic
or ubiquitous.4 (To oscillate between individualistic and social ways of
viewing oneself and the world, conversely, is a rather common strategy of
self-deception.) Yet self-deception, in its complex and multifarious forms,
which are more or less egregious, persists, and it persists in ways that are
readily identified by what Husserl would call “the natural attitude.”
Although, ultimately, neither the individual nor social poles is to be
privileged in making sense of self-deception, I shall begin by considering
its sociohistorical component. This is because the sociohistorical context
establishes the motivations for our behaviors, if not, in some less than uni-
versal sense, what is to count as truth itself. Nietzsche’s claim that “there
are no facts, only interpretations” is wrong if interpreted literally (which
is by no means the default position that one should take when interpret-
ing Nietzsche), but the element of truth in this claim, and what, at a
minimum, he plainly did believe, is that “the facts” are much thinner than
they are generally taken to be. Now, crucially, the types of truths that are
involved in self-deception are themselves nowhere near this thin, as they
go well beyond bare truths of the “it is snowing iff it is snowing” variety:
everything else being equal, to deny that it is snowing while one stands
4. Although Foucault does not speak in these terms, a relativistic account of self-
deception is implicit in his work, which I briefly consider later. Conversely, an account
of self-deception that would see the phenomenon as ubiquitous is implicit in Derrida’s
work. Thus, according to Derrida, Saussure rightly saw that words (written and spoken)
are arbitrary signs (“horse” might have just as well been used as “elephant” to signify
an elephant), and that signs get their meaning by virtue of their relation to other signs
in a “signifying chain,” but he continued to believe that the sign was comprised by not
only a signifier but also a signified, which means that Saussure’s sign still presupposes
some referent outside of language’s “signifying chain.” Derrida then purportedly brings
Saussure’s insight to fruition by claiming that the signified to which signifiers refer should
be lopped off, essentially suggesting that there is nothing to which language truly refers,
although the fiction of a referent is a necessary condition of intelligibility. As a result,
Derrida claims, all writing (not to mention speech, which, unlike writing, operated under
the illusion that its referent could actually be presenced) must be put “under erasure.”
What this suggests is that all writing (speech) is a deception of sorts, and thus all who
engage in writing (speech)—with the possible exception of the Derridean, who knows
that it’s all “under erasure”—engage in self-deception. All such cases of self-deception
are equally egregious, as the ineluctable presupposition of a nonexistent referent is not
amenable to any further gradations.
in the midst of a snowstorm is not self-deceptive but, rather, delusional.

A self-deceptive belief, in other words, must be plausibly true (and, in
fact, it might even be true, though it has been formed irrationally), and
to be plausibly true it cannot violate perceptually palpable, more or less
readily apparent, truths.5 Still, even if interpretation does not go all the
way down, it does go a long way, and it is the sociohistorical context that,
in one fell swoop, generates the sorts of interpretations that shape the
facts and affords the interpretive space within which self-deception is
able to maneuver.
SOCIOHISTORICAL TRUTH/SOCIOHISTORICAL DECEPTION
The concept of self-deception analytically presupposes the concepts

of deception and, ultimately, truth, and to make philosophical sense
of self-deception in sociohistorical terms it is necessary to first make
philosophical sense of truth in sociohistorical terms. To be scrupulously
avoided here are the two extreme positions, a hard teleological account,
which continues to have its adherents (although not all that many), 6
and social constructivism, which largely took off in the late 1960s in
reaction to the rightly perceived shortcomings of the hard teleological
account. Setting itself up as the only alternative to the hard teleological
account, however, social constructivism has, in effect, made the hard
teleological account synonymous with any sociohistorical account of
truth that purports to skirt social constructivism (just as it has made
“Cartesianism” synonymous with any philosophical account that does
not categorically reject the concept of “the subject”). In fact, the hard
teleological account and social constructivism are both deceptions of a
sort, and making sense of how these sociohistorical “truth-makers” can
themselves be deceptions will help clarify how self-deception can be
made sense of in sociohistorical terms.
With his claim that transcendental idealism is empirical realism’s con-
dition of possibility, it was Kant who opened the door to making sense
of truth in sociohistorical terms. Kant’s “Copernican revolution” holds,
5. This claim needs to be qualified somewhat. A self-deceptive belief that is based

on some state of affairs in the world must be “plausibly true,” but it would seem that a
self-deceptive belief concerning one’s own intentions need not be. Thus, a person with
a vicious habit might believe that he will kick it today despite similarly believing this
on a thousand previous occasions, and we might not deem this belief “plausibly true.”
Nevertheless, insofar as the person could make this belief good, it would be wrong to
say that such a belief is not “plausibly true.” Such a view, if held by the person, would
constitute what Sartre calls “bad faith,” which, I shall argue, stands in an intimate (although
by no mean identic) relationship to self-deception.
6. In the wake of the collapse of the Soviet Union and the nominal establishment
of “a new world order,” there was a resurgence of the hard teleological account, as was
exemplified by Francis Fukuyama’s End of History and the Last Man.
in essence, that we constitute the conditions of possibility with respect

to our experience (theoretical or practical) but that our judgments with
respect to the world still have an objective basis by virtue of an apriori
conceptuality (the categories or the categorical imperative). This permits
us to continue to speak of “the true” and “the right,” but now in terms that
make sense of us as rational animals for whom experience must initially
cohere if it is to rise to the level of a possible experience. Crucially, then,
we are responsible for “the true” and “the right,” the background against
which our (theoretical) judgments and (practical) actions are measured.
Of course, although Kant opened the door to making sense of truth in
sociohistorical terms by privileging apriori conceptuality and the unifying,
self-identical, self-reflexive self-consciousness that is its presupposition
(the transcendental unity of apperception), he refused to step through,
maintaining, in essence, that the categories through which we cognize the
world are invariant and that a sociohistorically tainted understanding of
ourselves as moral beings would be a rejection of our defining rational-
ity. For Kant, this itself would have been self-deception of the very worst
kind. Yet, by remaining every bit as formal as the categories, for which
the noumenon remains wholly out of bounds, the categorical impera-
tive presupposes some concept of noumenal agency, and this gives rise to
the well-known problem of reconciling a causally determined empirical
self and an entirely free intelligible one that is its transcendental ground.
What’s more, even if the complexities of this problem are circumvented
by viewing this transcendental self exclusively in regulative terms, a trou-
bling problem remains: as Hegel will argue, Kant is unable to juggle out all
sociohistorical considerations from his apriori morality, for in legislating
the moral law by way of the categorical imperative, they covertly manifest
themselves through the formulation of one’s maxim (principle of action).
Through the very process that would expel it, then, Kant’s apriori moral-
ity tends to morally sanitize the profane sociohistorical reality exactly as
it finds it.
The dynamics underlying Kant’s moral machinery thus open the space
for a particularly virulent form of self-deception, and it is particularly viru-
lent precisely because of its individualistic bias. In principle, to use the
terminology of self-deception indigenous to the individualistic approach,
self-deception can be overcome by the individual on Kant’s account
because the self that would deceive and be deceived by itself can always
be bracketed by the “real” self, the transcendental self, which is purport-
edly beyond the sorts of pathological determinations that would motivate
a person to engage in self-deception. For Kant, it is always within the pur-
view of our subjective intentions to ascend to the ground of the good will,
which, beyond the causal chains of natural laws, is the ground of a wholly
objective intention, the laws of freedom, as they issue from the categorical
imperative. If these laws are themselves just the stuff of their sociohis-
torical context, as Hegel’s criticism of Kant suggests, the objectivity that
Kant’s categorical imperative makes good is still an objective intention of
sorts, but not in the way Kant thinks: it is an objective intention to the
extent that it is the expression of the laws of a particular sociohistorical
context acting through the heteronomous individual, not the expression
of an autonomous individual who transcends the particular sociohistorical
context by acting on the categorical imperative so as to legislate universal
law. This is a particularly virulent form of self-deception because, like the
Christian notion of the soul, it enables certain of its practitioners to believe
that they are above the possibility of self-deception, and, moreover, that
they have a lock on “the Truth,” which, as Nietzsche would later argue,
is the most virulent form of self-deception. As Kant himself made abun-
dantly clear, his aim was to make good the moral and spiritual imperatives
of a Christian other-worldliness on the basis of pure reason, and, conse-
quently, it was not especially unfair of Nietzsche to claim that Kant only
sought to make good “popular prejudice,” that “Kant’s joke” was “to prove
in a way that would dumbfound the common man, that the common man
was right.”7 Kant’s own bit of self-deception, indeed, might well lie in his
belief that by absolutizing the universal moment that is necessary for any
genuinely moral comportment to make good his heteronomously deter-
mined Christian commitments, he was actually offering something more.
His deception, wittingly or not, was that those who act under the rubric
of an utterly autonomous individuality validate themselves rather than
that which stands over and against them, and that they do this so as to
overcome the sort of self-deceptiveness that Kant calls our “self-incurred
tutelage,” which this approach actually tends to perpetuate.
It was Hegel’s aim to bring Kant’s Copernican revolution to fruition,
and he seized on both the necessity of apriori conceptuality and the self-
reflexivity of a unifying self-consciousness as the conditions of possibility
for any experience, but he argued, in essence, that these transcendental
conditions must be understood in social terms. 8 Although, in some sense,
Hegel ultimately seeks to make good “the individual,” or at least reconcile
the individual to the social institutions of its world, he twists the individu-
al’s self-understanding away from Kant’s individualistic account. In place
of Kant’s categories and the transcendental unity of apperception, Hegel
offers “forms of consciousness” and “Spirit,” which refer, respectively, to
the socioculturally generated categories out of which a particular society
makes sense of its world (the Zeitgeist or “spirit of the times”) and to a
collective subject (a particular society or, more universally, all of human-
ity). The basic point is that whatever the contours of the society, the col-
lective is, epistemologically, prior to the individual, and the sociocultural
categories out of which a society makes sense of its world are, at a mini-
mum, the starting point for the individuals who constitute it.
7. Friedrich Nietzsche, The Gay Science, trans. Walter Kaufmann (New York: Random
House, 1974), 205–6 (sec. 193).
8. See, e.g., Robert B. Pippin, Hegel’s Idealism: The Satisfactions of Self-Consciousness
(Cambridge: Cambridge University Press, 1989).
By arguing that the apriori is social, and, therefore, aposteriori, as the

condition of experience’s possibility is nothing other than experience
itself, Hegel appears to open the door to the social constructivist account
of self-deception, and in a certain sense this is true, but it is surely not his
own position, and, indeed, he unambiguously undercuts it. To begin with,
a shift from one “form of consciousness” to another for Hegel results from
individuals who, by virtue of their particular experiences of the world,
come to revise their own beliefs. Hegel’s individuals, in other words, are
in a position to twist away from their (first-order) “motivated irrational
beliefs” by virtue of coming to discern the irrationality of the (second-
order) sociocultural motivations that motivate them, and when a large
enough number of individuals are able to come to this recognition, a new
set of (second-order) sociocultural beliefs, and thus motivations, is gen-
erated. This presupposes a kind of epistemic openness for which social
constructivist approaches cannot (and do not want to) account. When
a social constructivist such as Foucault, for example, offers up a histori-
cal account featuring successive epistemes (his archaelogical period) or
“regimes of truth” (his genealogical period), he cannot account for shifts
from one of these apriori structures of possible knowledge to another,
for there are no subjects (if by “subject” we mean an efficacious agent) to
bring about the change. There is no meaningful space for self-deception in
this social constructivist account (although there may be a goodly amount
of self-deception on the part of the social constructivist who wants to
hold it), and for two interrelated reasons: to the extent that “the self” is
thoroughly outer constructed, there is no self-determining subject to be
deceived, much less self-deceived, and one of the fundamental assump-
tions of the social constructivist account, that truth is relative to an
episteme or “regime of truth,” would preclude the notion that these apriori
structures of possible knowledge themselves can be understood as more
or less truthful. To the extent that an individual would be characterized as
self-deceived, therefore, the characterization would only be based on the
fact that his beliefs do not line up with his culture’s own “normalizing”
network of (power-) knowledge, which wholly produces the individual’s
beliefs nevertheless. A function of the culture, the “self-deceived” indi-
vidual may not believe in the truth of a set of contradictory propositions,
as the culture maintains that he ought, and he does not will this “false”
belief but, rather, it “wills” him by virtue of how matters genuinely appear
from his socially constructed perspective. Hegel, to be sure, does not have
any of these problems.
Crucially, however, Hegel’s historical account of how we come to “the
truth” might only be able to circumvent these problems by resorting to
teleological assumptions, and, even more troubling, he thinks that “the
truth” has been actualized in his own time. Properly qualified, however,
these commitments do not invalidate the basic structure of his histori-
cal account. As an initial matter, because they perform vital explanatory
functions without invariably careening into metaphysics, teleological
assumptions are not necessarily a bad thing. Human beings quite clearly
act in ways that evidence more or less persistent tendencies to strive for
particular ends, whether in terms of closed systems, such as markets, or,
indeed, in terms of the reproduction of life itself. Moreover, while Hegel
purportedly claims to speak from “the end of history” in the Phenomenol-
ogy of Spirit and, even worse, from a form of consciousness that he pro-
vocatively calls “Absolute Knowing,” I take his point to be a modest one. 9
What it means to achieve philosophical truth, or the truth about truth, is
to achieve the recognition that we constitute the world of our experience
by way of an apriori (socioculturally engendered) conceptuality, which
delineates the nature of the kinds of empirical truths we can achieve.
If we bear in mind that Hegel is speaking of second-order (instead of first-
order) belief formation, this achievement, no mean feat in itself, is not
inconsistent with Bernard Williams’s far more modest-sounding claim
that “the project of seriously pursuing the truth is one of controlling the
formation of belief.”10 In any case, Hegel’s construal of history in universal-
istic terms, which is based on the notion that the kernel of reason inherent
within our belief structures gears them toward “the truth,” sharply distin-
guishes his form of consciousness from Foucault’s relativistic episteme or
“regime of truth.”
For Hegel, this pursuit of truth does not stem from an idle philosophi-
cal commitment but, rather, from our deepest personal commitments (or,
as he puts it in the introduction to the Philosophy of History, from our pas-
sions), which themselves are but a manifestation of “the cunning of rea-
son” and, ultimately, truth working itself out in history. Hegel’s confidence
in the Phenomenology, purportedly made good from his “end of history”
perspective, is that collective belief reformation had, in fact, moved along
the right path: our knowledge, motored by a reason inherently geared
toward truth, and our interests, motored by our passions, had coincided.
Yet Hegel’s confidence, as exemplified by his own standpoint at the end
of the Phenomenology, was evidently less than warranted, as his own privi-
leging of the Prussian state less than twenty years later in the Philosophy
of Right amply attests. To clarify the problem here and what it means for
truth: it would be a mistake to categorically reject Hegel’s teleological
account, either launching forward into a social constructivism that would
see itself as bringing to fruition Hegel’s attempt to bring Kant’s revolu-
tion to fruition or falling back into a bare bones positivism that would see
truth as restricted to claims of the “it is snowing iff it is snowing” variety.
These two moves do violence to truth, and, while the underlying mechan-
ics are different, they do so in much the same way: in essence, the social
9. For a similarly modest account of “Absolute Knowing,” which has influenced my

own thinking on the matter, see Robert C. Solomon, In the Spirit of Hegel (Oxford: Oxford
University Press, 1983).
10. Bernard Williams, Truth and Truthfulness (Princeton: Princeton University Press,
2002), 133.
constructivist directly privileges the existing sociohistorical truths (if for

no other reason than that there is no basis for saying that they should not
be), while the positivist indirectly privileges the existing sociohistorical
truths by privileging what are nominally baseline facts that are actually
constructed by them. This is not to say that baseline truths are fictions
that ought to be rejected, for these are the things to which sociohistori-
cal truths must be tethered, but it is to say that any account of truth that
claims not to go beyond them already has. Ultimately, both accounts of
truth are sociohistorically generated deceptions.
Williams rightly says that if “the project of seriously pursuing the truth
is one of controlling the formation of belief,” we are already assuming that
people “actually want to find out the truth,”11 and this might not be the
case. As self-deception makes clear (in some as yet unspecified way), there
are reasons that people might not want to find out the truth, and, as decep-
tion makes even clearer, there are reasons that people might not want
others to find out the truth. At the level of first-order belief formation,
the order beyond which accounts of deception and self-deception do not
usually go, the mechanics of deception are as obvious as the mechanics of
self-deception are inscrutable. At the level of second-order belief forma-
tion, however, these phenomena take on a different look. If some variant
of Hegel’s sociohistorically conditioned apriori conceptuality is taken seri-
ously, which includes any account of truth that recognizes the inextricably
contextual nature of all knowledge, then what counts as truth or deception
is basically circumscribed by the sociohistorical categories out of which one
comes to know the world. Viewed in this way, the project of “controlling
the formation of belief” becomes, in Hegel’s terminology, the project of
controlling the formation of the form of consciousness, and, thereby, essen-
tially controlling what is to count as a truth. But unlike the account in the
Phenomenology, which, Hegel contends, is about the unfolding of philosophi-
cal truth, the real movement of history manifests the unfolding of particular
interests, and it does so in ways that are not necessarily compatible with the
truth, philosophical or otherwise. That is to say, at the level of second-order
belief formation, the form of consciousness, we are not necessarily talking
about beliefs that are oriented toward the truth, and, therefore, at least at
first blush, we might actually be talking about a different kind of belief.
First-order beliefs are inherently geared toward the truth, and, there-
fore, we do not intentionally bring them about. Rather, our substantive
beliefs about the world arise because we believe that they are true, or at
least truer than the beliefs that they have supplanted, for if this were not
the case, we would not believe in them and they would no longer qual-
ify as beliefs.12 At first blush, this account of first-order belief formation
11. Bernard Williams, Truth and Truthfulness (Princeton: Princeton University Press,
2002), 133.
12. This is the fundamental point of Williams’s “Deciding to Believe,” in Problems of the
Self (New York: Cambridge University Press, 1973).
dovetails nicely with Hegel’s account of second-order belief reformation.

New forms of consciousness arise by virtue of breakdowns in prior ones,
and while a form of consciousness does not logically entail the one that
supplants it, the ensuing form of consciousness arises because it seems
truer than the one it supplants. In reality, however, this is not the way sec-
ond-order belief (re)formation transpires, and I take it to be the case that
Hegel’s account falls short for two interrelated reasons: First, with the
exception of the master-slave relationship, every form of consciousness
in the Phenomenology is homogenous,13 but in the world second-order
belief reformation is a more or less contested terrain, and (at least before
a particular “form of consciousness” is able to get sociohistorical traction,
and generally even after) does not reflect the beliefs of all. Societies have
never been as homogenous as they are in the stories they like to tell about
themselves. Second, to say that second-order beliefs do not reflect the
beliefs of all is a bit of a misnomer, for at this level of belief (re)forma-
tion, “beliefs” are internally related to interests rather than truth. Forms of
consciousness are crafted by particular social formations who aim to “con-
trol the formation of belief” so as to either attain or maintain the social
upper hand (economically, politically, and/or culturally). Of course, these
groups have reasons for seeking to construct the kinds of second-order
beliefs that they do, and these reasons are surely, in no small part, socio-
historically determined, but if there is a “cunning of reason” involved, it is
internally related to interests rather than the truth, and interests, in turn,
are not internally related to the truth,14 as Hegel’s model presupposes. Put
more concisely, at the level of second-order belief (re)formation, reason
and truth are no longer internally related.
Second-order belief (re)formation is, therefore, of a qualitatively differ-
ent nature than first-order belief formation. First-order belief formation
deals with beliefs that arise from particular factual states in the world, and
the (epistemic) question is whether these beliefs are true (i.e., whether
they accurately track the particular factual states), while second-order
belief (re)formation deals with beliefs that are crafted in the service of
particular interests, and, ultimately, particular ways of living in the world,
13. Thus, the form of consciousness that immediately supplants the master-slave
relationship is stoicism, and when Hegel declares that “whether on the throne or in
chains” the aim of this form of consciousness “is to be free,” he is tacitly suggesting that the
differences between Marcus Aurelius and Epictetus (and the social classes of which they
are a part) are only a secondary matter, which is way too reductive in terms of capturing
the sociohistorical truth of the matter. See G. W. F. Hegel, Phenomenology of Spirit, trans.
A. V. Miller (Oxford: Oxford University Press, 1977), 121.
14. In the opening paragraph of “Deciding to Believe,” Williams says that his claim that
we do not decide to believe goes only so far as “straightforward factual belief,” and that he
is explicitly not dealing with “religious and moral beliefs, belief in the sense of conviction
of an ideological or practical character.” Problems of the Self, 136 (emphasis added). I will be
touching on ideology momentarily, but what I am suggesting here is that the sorts of
beliefs Williams is exempting from his discussion are those that make up second-order
beliefs and, therefore, can be intentionally chosen.
and the question (both ethical and existential) is whether a particular

way of living in the world hangs together. Crucially, however, second-
order beliefs are, in large part, the truth-makers in terms of first-order
beliefs, and this points to the possibility of an intentional construction of
an untruthful truth-maker, a second-order form of consciousness or struc-
ture of beliefs that, if not utterly indifferent to the truth, is not, for the
most part, motivated by it. Now, Hegel would reject this account, for he
contends that the criteria of truth are internally generated (i.e., conscious-
ness itself establishes the standard by which to measure the truth of what
it knows with respect to the world), and thus any knowledge of an object
necessarily presupposes a set of second-order beliefs that overdetermine
it. Even by Hegel’s lights this does not seem to be right, however, for
to the extent that Hegel’s account would hold that second-order beliefs
manufacture the object of experience from the bottom up, it would be
similar to social constructivism and, therefore, would not be in a position
to account for second-order belief reformation, which is at the heart of
Hegel’s dialectic. If ways of living do not cohere, it is not only because we
have not gotten our recognitive (or subject-subject) relationships right,
though this is surely part of it, but also because we have not gotten our
relationships to the objects of our experience right, and one way this hap-
pens is if we idealistically take it to be the case that our second-order
beliefs overdetermine the objects of our experience. It is one thing to
acknowledge that phenomena are necessarily mediated by our (sociohis-
torically engendered) second-order beliefs (or, in Hegel’s terms, our form
of consciousness), which I take to be right, but it is something else to say
that, by virtue of the ineluctability of this mediated relation, there is no
fact of the matter independent of the way the object is constituted by our
second-order beliefs.
To say that there is no fact of the matter independent of the way sec-
ond-order beliefs constitute the objects of our experience is tantamount
to saying, in the end, not only that second-order beliefs are the ground
of “ideology,” which is surely true, but also that second-order beliefs are
ineluctably ideological. Of course, the question is ultimately how “ideol-
ogy” is defined, and, indeed, in one sense second-order beliefs are ineluc-
tably ideological (in a rather innocuous way), since this would seem to be
bound up with the claim that a sociohistorically conditioned apriori con-
ceptuality is experience’s condition of possibility. Yet, in another sense,
the one that has held sway historically, “ideology” connotes something
far more pejorative, as it has been linked with an idea discussed in con-
nection with the formation of second-order beliefs, namely, that they are
crafted by particular social formations who aim to “control the formation
of belief” so as to either attain or maintain the social upper hand (in the
service of their own interests, if not an unqualified will to power). More
recent philosophers have been hostile to this sense of the term, and (gen-
erally under the rubric of rejecting it as a misguided paternalism) no lon-
ger see ideology as a deception that fosters a “false consciousness” cutting
against one’s genuine interests, but, instead, see it as part of an organic and
relational whole that is embodied in our social institutions. 15 For the most
part, however, by altogether rejecting the second sense of ideology, these
philosophers have, in effect, rejected the notion that there is anything
left of truth with respect to the first sense of the term, thus normalizing
the partisan nature of second-order belief (re)formation. Put somewhat
differently, the space for the second, more pejorative, sense of ideology
disappears only if there is no basis for judging ideology in the first sense
to be more or less true, and this effectively does away with the standpoint
that might critique the prevailing ideology, which is in accordance with
the tastes of social constructivism. Ideology, in the first sense, might well
be inevitable, but this does not mean that particular ideologies are all
equally true, which is effectively the case when truth itself disappears, and
this view, in itself, is not just in the service of ideology (in the second sense
of the term) but is also a manifestation of it.
What space, then, is left for truth with respect to second-order belief
(re)formation, particularly given the fact that this order of belief is geared
toward interests rather than truth? Positivism, as was indicated earlier, is
certainly not the answer, for its account of truth is too deflationary and
tends to recapitulate the sociohistorical status quo. Conversely, hard tele-
ological accounts such as Hegel’s (and I am thinking here of the Philoso-
phy of Right more than the Phenomenology of Spirit) are too inflationary, for
they are instilled with both an unwarranted necessity and an unwarranted
universality. What’s more, like positivism, they tend to recapitulate the
status quo, although from the other extreme, by making their sociohis-
torical standpoint into the culmination of truth. A middle position, one
that was staked out by Adorno, is to hang on to a teleological account but
to hang on to it in a highly qualified way. When Adorno says that “univer-
sal history must be construed and denied,” for example, he is implicitly
rejecting both the hard teleological account and social constructivism by
advocating, as J. M. Bernstein puts it, “teleology without a telos.”16 The
“truth” toward which history aims on this account is the overcoming of
deception, and, therefore, what truth means at any particular time would
be the overcoming of the deceptive sociohistorically conditioned (second-
order) beliefs of that particular time. This entails far more than can pos-
sibly be broached here, but for present purposes two things, in particular,
are worth noting. First, by “construing” universal history, a critical, truth-
oriented basis for analyzing the present state of affairs (and, therefore, our
second-order, sociohistorically conditioned beliefs) is retained, as we have
an expansive enough framework for making sense of the ways our exist-
ing form of consciousness (or second-order beliefs) falls short. (In other
15. See, e.g., Ernesto Laclau and Chantal Mouffe, Hegemony and Socialist Strategy:
Towards a Radical Democratic Politics (London: Verso Books, 1985).
16. J. M. Bernstein, Adorno: Disenchantment and Ethics (Cambridge: Cambridge
University Press, 2001), 336.
words, if universal history is categorically denied, our standpoint is coter-

minous with what we would analyze, and this telescoping would reduce
us to the role of painting “grey in grey,” as Hegel unhappily puts it in the
Philosophy of Right.) Second, by “denying” universal history, the unappeal-
ing aspects of universal history, its universality and necessity, go by the
wayside. There is the recognition that the truth does not reside with the
particular group that was able to get the upper social hand, which is to
recognize that at any historical time there are many groups whose inter-
ests (whether they are perceived or not) do not find expression in our
second-order beliefs. Moreover, there is the recognition that history is
characterized by contingency. One source of this contingency is the bare
facts of the world themselves, which means that although our second-
order beliefs go a long way down in terms of interpreting the facts on the
ground, thus providing the context in which we frame the truth as an
initial matter, they do not overrun the bare facts. Crucially, this account
opens up spaces for individual experience, and, as a result of this, opens
up spaces for a somewhat more robust account of self-deception, which
I will offer shortly. In the meantime, however, to claim that second-order
beliefs do not go all the way down because they do not overrun the bare
facts is not to claim that there is some immediate relation to the bare facts,
as what the facts mean is, invariably, sociohistorically mediated, but it is to
claim that there are facts and that these facts, though underdetermining,
circumscribe the range of sociohistorical narratives that can be truthfully
articulated with respect to them. (Conceptually, this is one way in which
one might distinguish between ideology in its pejorative and its nonpejo-
rative senses.) Moreover, although ideology (in the pejorative sense) must
be understood in terms of deception rather than self-deception, and thus,
in no small part, the claim “that ideology operates through self-deception
is itself a piece of ideology” (in the pejorative sense), as Allen Wood has
argued,17 at some point the readiness to buy into a particular ideology
does, in fact, become self-deception. (And this point, I would parentheti-
cally suggest, is the point at which one is more or less clearly confronted
with bare facts that contradict the ideology, an epistemic standpoint that
is becoming increasingly difficult to achieve in our burgeoning, overmedi-
ated, postmodern societies.)18 The readiness to believe in what is plainly
17. Allen Wood, “Ideology, False Consciousness, and Social Illusion,” in Perspectives on
Self-Deception, ed. Brian P. McLaughlin and Amelie Oksenberg Rorty (Berkeley: University
of California Press, 1988), 360.
18. The aim of manipulating the media is ultimately to manipulate what I have
been calling our second-order beliefs. And although all American administrations
try to manipulate the media, it may well be that the Bush administration's heavy-
handedness in this regard was unparalleled in American history. This project became more
auspicious with the horrors of September 11, 2001, when a new form of consciousness
was engendered, one overwhelmingly disposed to believe the Bush administration’s
propaganda, which, in large part, was designed to reinforce this disposition. Americans
have never had a particularly strong historical sense (which serves the goals of ideology
not the case attests to a motivated irrationality, a motivated irrationality

that is spurred by interests rather than the truth, but it is not infrequently
the case that such interests are inextricably intertwined with our deep-
est existential commitments, and, when it comes to these commitments,
even the most cynical individuals generally seek to bolster them with the
imprimatur of truth. The reasons for this are not only ideological: this
inclination is the homage that interest-driven duplicity pays to personal
integrity and, ultimately, to the truth.
SOCIOHISTORICAL DECEPTION, BAD FAITH,

AND SELF-DECEPTION
Although Kierkegaard and Nietzsche struggle with the problem of self-

deception, and, in many respects, accurately discern the ways in which
our self-understanding is influenced by the social world, Heidegger and
Sartre raise the ante by claiming that the particulars of the social world
are fundamentally constitutive of the self, which inextricably entwines
the self, self-deception, and the social world. Indeed, Heidegger’s well-
known antipathy toward the notion of interior mental states goes so far
that he does not even refer to human beings as selves or persons but,
instead, as Dasein (“being-there”), and the identity of any particular
Dasein wholly derives from the social world into which he or she has
been (contingently) “thrown.” One is as one does, and what one does is
delineated by das Man (“the they”), the social framework that sets out
the roles that any particular Dasein can assume and the practices (col-
lective or individual) in which any particular Dasein can engage. Like
Hegel’s form of consciousness, Heidegger’s das Man is epistemically prior
to the individual (Dasein), and it functions as the “truth-maker.” However,
unlike Hegel, whose broader dialectic culminates in philosophical truth
and social reconciliation, Heidegger contends that absorption in the world
of “leveling” das Man bespeaks a “fallenness” (marked by idle talk, curios-
ity, and ambiguity), and, following Kierkegaard, he seeks to carve out a
in the pejorative sense and effectively makes good social constructivism), but historical
amnesia, coupled with a committed distractedness, became the order of the day as
activities associated with such names as the Patriot Act, Abu Ghraib, and Guantanamo
Bay, which are antithetical to basic American principles, were met with relative
equanimity. Recognizing that the bare facts, even ones that were not incompatible with
the ideological justifications for its greatly expanded power, might tend to undermine
it, the Bush administration kept tight control on a host of images. And, in the end, the
hard facts, as evinced by such images, did undermine the Bush administration’s veneer
of competence. Social critics have argued that Hurricane Katrina (rather than the
Iraq War) was what precipitated this administration’s downward political spiral (and
emboldened an obsequious media), as there was no effective way for the administration
to spin its complicity in this disaster in the face of well-documented facts, which included
unambiguous images contradicting their claims.
space for authenticity. Heidegger evidences a strong ambivalence here,

and this ambivalence ultimately testifies to the belief that truth’s condi-
tion of possibility might be, simultaneously, its condition of impossibility,
or, put in the terms I have used, that sociohistorical truth is sociohistorical
deception. Nevertheless, by virtue of his virulently antisubjectivistic bent,
Heidegger does not speak of self-deception here, as the deception that is
our social world entirely overruns Dasein, unless, that is, Dasein is able to
recover its “ownness” by facing up to its “ownmost possibility” (death),
which engenders a certain resoluteness. What authenticity can mean on
an inauthentic landscape is far from clear, however, and by virtue of his
privileging of a more primordial truth, manifested in a more primordial
being-in-the-world inexorably covered up by das Man (which, again, is
the condition of truth), self-deception is both impossible and ubiquitous.
(Nevertheless, although Heidegger’s fundamental ontology leaves no
meaningful space for making sense of the phenomenon of self-deception,
it collapses into self-deception in a way that is not unlike Kant’s phi-
losophy: the more Heidegger’s authentic Dasein thinks it is getting to a
more primordial truth about its world, the more it unreflectively mistakes
sociohistorically mediated phenomena for primordial truth.)
I began this essay with the claim that to put the question of self-
deception in either individualistic or social constructivist terms is to lose
the phenomenon itself, as it becomes either impossible or ubiquitous, and
it is not simply a coincidence that Heidegger’s philosophy exhibits both
of these moments. In the first moment, Heidegger rejects subjectivity with
his account of Dasein and das Man, thus laying the foundation for what
will become social constructivism, while in the second moment he draws
on the self-styled “champion of subjectivity,” Kierkegaard, to make good
the possibility of an authentic comportment. For this reason, Heidegger’s
fundamental ontology is fundamentally antinomical. As I have suggested,
however, to make sense of the phenomenon of self-deception, it is neces-
sary to mediate these two moments, given that self-deception arises in
the interregnum, and, from the existential phenomenological standpoint,
Sartre makes the best go of it with his well-known account of “bad faith.”
Although there are a variety of problems with Sartre’s account, some of
which arise from his failure to properly delineate the relation between bad
faith and self-deception, it has the resources not only to skirt this either/or
but also to unravel the two “paradoxes of self-deception,” which are mani-
fested in the “lexical” and “intentionalist” approaches to the phenomenon.
What may well be the chief ambiguity in Sartre’s account of bad faith
in Being and Nothingness is a lingering commitment to a conception of
self-deception based on the interpersonal model of deception, even as his
account of bad faith has basically gone beyond this. Thus, Sartre initially
informs us, to be in bad faith is to “lie to oneself,”19 but the problem with
19. Jean-Paul Sartre, Being and Nothingness, trans. Hazel Barnes (New York:
Washington Square Press, 1956), 89.
this formulation is that the duality that exists between the deceiver and
the one who is deceived transpires within a single consciousness. This, of
course, is standard fare, a classical statement of the problem that defines
the lexical approach, but it is not one that Sartre sticks with for very long.
After he rules out the Freudian unconscious as a satisfactory explanation
of this phenomenon, given that it only pushes the duality problem back
a step (because there must then be some censor between consciousness
and unconsciousness that both believes and does not believe x), Sartre
formulates the problem of bad faith anew. Bad faith, he says, is grounded
in “the double property of the human being, who is at once a facticity and
a transcendence,” and derives from a failure to either “coordinate” or “syn-
thesize” these two properties of the human being. 20 According to Wood,
both of these formulations are troubling, and he asks: “Does Sartre really
expect us to believe that every case of self-deception involves attributing a
contradictory concept to something? And does he think that all contradic-
tory concepts derive from the facticity-transcendence relation? Neither
claim has much plausibility, and neither receives any real defense from
Sartre.”21 Wood’s questions, which are partially motivated by his view that
Sartre wrongly rejects the Freudian unconscious, go to the heart of Sartre’s
account of self-deception, and they must be adequately addressed.
As an initial matter, the best way to address these questions is to turn
them around, for I take the dynamics of the facticity-transcendence rela-
tion to be primary, and once this relation is clarified, the “contradictory
concept” problem, which I have already suggested is a misguided formu-
lation of self-deception, can be clarified. To begin, then, for Sartre, the
facticity-transcendence relation points to the fact that there are brute
facts about all human beings (including our bodies, our pasts, our envi-
ronments, and, crucially, the collective practices of the sociohistorically
informed contexts in which we live) but that we are always also “beyond”
or “transcend” these facts to the limited extent that we are free to inter-
pret them as we choose. Structurally, this relation is not unlike the rela-
tion I discussed earlier in the context of truth: there are facts of the matter
that our ability to interpret will not change, and, indeed, better and worse
interpretations depend on our relative faithfulness toward these bare facts,
while what these facts mean, and more substantial questions concerning
what is to count as truth, are chiefly interpretive in nature. (It should be
emphasized here that although many critics take Sartre to be offering a
hyperbolic notion of freedom, part of what constitutes facticity for him
incorporates the limiting sociohistorical truths of one’s world.) According
to Sartre, bad faith arises from overemphasizing one of these properties at
the expense of the other (usually, although by no means exclusively, factic-
ity at the expense of freedom), and he offers various examples to illustrate
20. Ibid., 98.

21. Allen Wood, “Self-Deception and Bad Faith,” in McLaughlin and Rorty, Perspectives
on Self-Deception, 215.
the two ways we can fall off the wagon. 22 Now, although Sartre gestures
in the direction of good faith, initially suggesting that it would involve
either coordinating or synthesizing these two properties, and then sug-
gesting that, contrary to bad faith (“not believing what one believes”), the
ideal of good faith is to actually “believe what one believes” and, thereby,
find “refuge in being,”23 the fact is that good faith is off limits on Sartre’s
account. To validly coordinate or synthesize facticity and transcendence
presupposes that they are functionally discrete, but, in fact, they are dia-
lectically intertwined, as transcendent freedom only exists in a factical
situation and a factical situation only arises from transcendent freedom.
Put simply, there is no (transcendent) Archimedean point from which to
coordinate or synthesize these properties, a fact that will become clearer
when considering the sociohistorical component of the factical self.
This lack of an Archimedean point, the existential unavoidability of
bad faith, and the ultimately sociohistorical nature of bad faith converge
in the context of what Sartre calls “the fundamental project.” According
to Sartre, a person’s possibilities arise within the framework of a hierar-
chy of projects and behaviors, and these projects and behaviors testify,
in turn, to the existence of a more basic project, “the initial project,”
which reflects one’s choice of oneself in the world. The content of this
self-constituting choice of oneself obviously varies from one person to
another, but the existence of the initial project itself testifies to the exis-
tence of an even more basic project, “the fundamental project,” a univer-
sal project Sartre describes as the project of being God. For Sartre, this
project is not religious but, instead, existential, essentially indicating that
human beings strive to be absolutely self-identical and yet absolutely
free. Without getting bogged down in Sartre’s terminology or in the sec-
ondary literature that this idea has generated, I take Sartre’s point to be
crucial: although the project of being God (absolutizing our facticity and
our transcendence) is itself a project in bad faith, the (empirical) pursuit
of this project by way of the initial project is, existentially, unavoidable,
as we must maintain a dialectical tension between these connected but
conflicting sides of our nature. Indeed, this is Sartre’s point when he
asserts that “freedom, which manifests itself through anguish, is charac-
terized by a constantly renewed obligation to remake the Self which des-
ignates my free being . . . [and that] this self with its a priori and historical
22. It has been pointed out that the examples are flawed in a number of ways, one of
which is that Sartre’s phenomenological descriptions are necessarily limited to his own
point of view, and that he has no basis for saying whether these people are in bad faith, as
he does not have a clue about what might actually be motivating them. I think that this is
right, and, indeed, I think that this points to the fact that attributing bad faith to another is
no simple matter, but I also think that it does not bear on the basics of his analysis of bad
faith as such. See, for example, Kathleen Marie Higgins, “Bad Faith and Kitsch as Models
for Self-Deception,” in Self and Deception: A Cross-cultural Philosophical Enquiry, ed. Roger
T. Ames and Wimal Dissanayake (Albany: State University of New York Press, 1996).
23. Sartre, Being and Nothingness, 115.
content is the essence of man.”24 For Sartre, then, the content of our
initial project or choice of ourselves is the content of our sociohistorical
context, and, by constituting the self, this choice gives rise, in one fell
swoop, to both the possibility of a situation and to the particular situa-
tion, whose meaning derives from the content of the choice.
If bad faith goes all the way down, it is because our world-constituting
initial choice of ourselves goes all the way down, for by virtue of this
ineluctable choice we are always already oriented to the world in a highly
particularized way. This distinguishes bad faith from self-deception, and,
in fact, makes clear that bad faith is logically prior to self-deception. Thus,
according to Robert Solomon, “bad faith is not just self-deception because
it is not primarily about belief. Of course, beliefs often follow, but bad
faith is about the very nature of our engagements in the world, and then,
perhaps, we form beliefs about our engagements in the world.” This is
basically right, but for my purposes here it can be profitably reformu-
lated. Bad faith is about belief (as all faith is), if by belief what is meant is
second-order belief, and what I would like to suggest here is that Sartre’s
account of bad faith mediates sociohistorical categoriality (i.e., the pre-
vailing “form of consciousness,” to use Hegel’s phrase) and the individual.
This accounts for the obvious fact that our self-identities often substan-
tially differ even if we are born into the same sociohistorical context, and,
indeed, into similar conditions within it: every person appropriates his
or her world in a particular way, but the content of our social world is
still, ultimately, the content of our selves (and, as discussed, this certainly
seems to be Sartre’s position). Understood in this way, our social cat-
egories might be viewed as third-order beliefs, and second-order beliefs,
which are basically constitutive of the self, are an amalgam of third-order
beliefs and the individual’s particular experiences. As a result, the greater
the ability of a particular subset of social interests to shape and control
these now third-order beliefs, the more individual experiences will be
schematized in accordance with them, and the more our self-identities
will appear to be epiphenomenal with respect to the social totality. This is
the kernel of truth in social constructivism, which functions ideologically
(in the pejorative sense of the term) by making a philosophical apriori of
a sociohistorical tendency that seems to be coming to fruition in these
postmodern times.
When Sartre’s account of bad faith is properly augmented by his
account of the self-constituting initial project, the claim that bad faith
inexorably arises from the relation between transcendence and facticity,
and, moreover, that bad faith is self-deception’s condition of possibility,
becomes more compelling. Still, this does not yet directly address Wood’s
questions, whether every case of self-deception involves the attribution
of a contradictory concept and whether all contradictory concepts arise
from the facticity-transcendence relation, although it does provide the
24. Ibid., 72.

grounds for doing so. Even if Sartre is less than clear on this himself, his
account of bad faith suggests that self-deception need not and should
not be understood as a variation on the model of interpersonal decep-
tion, which is what produces the “contradictory concept” problem, the
paradox of believing and not believing the very same thing. And, once
the interpersonal model goes by the boards, it becomes clear not only
that contradictory concepts do not arise from the facticity-transcendence
relation but also that no case of self-deception involves the attribution of
a contradictory concept.
Although the bulk of the chapter on bad faith in Being and Nothing-
ness deals with bad faith as an existential phenomenon, in the final four
pages of the chapter, Sartre turns to the epistemological questions that are
now commonly connected with the problem of self-deception. Sartre says
there that “to believe is not to believe,” which seems to raise the “contra-
dictory concept” problem, but this is not actually his point, as he goes on
to say that “every belief is a belief that falls short,” for “one never wholly
believes what one believes.”25 As Wood rightly contends, “to believe is not
to believe” is really meant to convey the fact that all beliefs are “imper-
fect,” and, at one level, he is right to explain this imperfection by point-
ing to the fact that “the human condition” is marked by “complexities,
ambiguities, uncertainties, and tensions” that “often hit us right where
we live.”26 At a deeper level, however, I take Sartre to be saying, more
radically, that because our beliefs are bound up with our self-constituting
initial projects, which are entirely ungrounded, and therefore, in Sartre’s
words, “metastable” (i.e., internally unstable and thus subject to fluctua-
tion), then our beliefs are, in essence, only as good as we are. It should be
noticed that this is not unlike what I have begun to call our third-order
beliefs, our social categoriality or form of consciousness, with which our
individual second-order beliefs stand in a dialectical relation: a person’s
overall belief structure depends on the internal coherence of his or her
“truth-makers,” both at the second and third levels, and, in both cases, part
of what makes his or her belief structures more rather than less coherent,
and thus more rather than less stable, is their relative faithfulness to the
bare facts. When Sartre says that “to believe is not to believe,” therefore,
he is not saying that there are no bare facts of the matter (which is just
what his notion of facticity would reject), and, indeed, he is not merely
saying that the bare facts are never taken up by us as such, as they are
always already bound up with our self-constituting initial projects. He
is saying, rather, that by virtue of being oriented to the world in a par-
ticular way (which is the result of the dynamical interplay between our
second- and third-order beliefs), we are always already inclined to see
the world in a particular way, and this “selective seeing” prereflectively
sifts though the variegated phenomena that perpetually confront us and
25. Sartre, Being and Nothingness, 114–15.

26. Wood, “Self-Deception and Bad Faith,” 218–19.
largely delineates the range of phenomena that can rise to the level of
facts for us. With respect to the “contradictory concept” problem, what
this means is that we do not actually believe and not believe the very
same thing, but, instead, that by virtue of the nature of self-consciousness,
we necessarily “spell out” some things at the expense of others,27 and what
we “spell out” is ultimately driven by our initial projects. Rather than
believe and not believe the very same thing, then, we “spell out” or make
explicit those aspects of the thing that conform to our self-understanding,
and we do not “spell out” or make explicit those aspects of the thing that
do not conform to our self-understanding. Thus, although all beliefs are
inherently “imperfect,” bad faith does not ultimately trade on this fact, as
Wood contends, but rather on the fact that we cannot be explicitly aware
of all the phenomena we experience from moment to moment: we do not
believe and not believe the very same thing, but we are more or less aware
of conflicting aspects of the very same thing, depending on the way these
aspects cut with respect to the self’s constitution. Understood in a quali-
fied way, therefore, Sartre, like Freud, trades on the notion of a “divided
mind,” even as he rejects the notion that Freud’s unconscious is needed to
account for a (relative) lack of awareness with respect to certain aspects
of a phenomenon.28
With his notion of the initial project, Sartre is also able to account for
the second (and, arguably, more fundamental) paradox of self-deception,
the problem of intentionalism (i.e., the problem of explaining how we can
“deliberately” deceive ourselves), and he does so in a way that transcends
the motivated irrationality approach. Even when it is claimed that moti-
vated irrationality can go all the way down, which obviates the need to
explain self-deception as an intentional activity, it is still usually claimed
that this occurs only when a person is in an aberrant state, as the irrational
belief is formed “under the influence” of his or her emotions and desires. 29
This aberrant state, in turn, tacitly presupposes a normal state, one in
which a person is not “under the influence,” and when in this state a per-
son is supposedly able to form rational beliefs because he or she is either
unmotivated or, more likely, motivated in a way that is undistorted by his
or her underlying interests. Yet, even this second alternative is much too
rationalistic, as it fails to account for the fact that even in one’s most dis-
passionate moods one is always already highly motivated due to the lim-
ited range of possible perspectives that are made available by one’s time
27. Herbert Fingarette makes this argument in Self-Deception (London: Routledge and
Kegan Paul, 1969).
28. Wood is thus right to say that “some sort of ‘divided mind’ explanation” is required
to account for the “selective inattention” that is part and parcel of Sartre’s account of self-
deception, but as this explanation suggests, it need not be one that depends on the positing
of “the unconscious.” See ibid., 223.
29. See, e.g., Ariela Lazar, “Deceiving Oneself or Self-Deceived? On the Formation of
Beliefs ‘Under the Influence.’”
and place, which is only another way of saying that a sociohistorically

engendered apriori conceptuality is experience’s condition of possibil-
ity. Sartre’s notion of the initial project, though unnecessarily reductive,
points to the fact that social categoriality, the form of consciousness out
of which a society makes sense of both itself and its world, must, in turn,
be appropriated by the individual, and that it is at the transcendental
level of this self-constituting particularized appropriation, which is no less
a condition of possibility with respect to experience, that the so-called
intention in self-deception arises. Consequently, we are all always already
“under the influence” twice over, for we are all always already subject to
the motivations of third-order (sociohistorical) and second-order (self-
constituting) beliefs. If belief formation “under the influence” is irrational,
then we are all irrational, if by “irrational” we mean that we cannot get
beneath a mediating self-interpretation, and that, in any event, there is
nothing beneath any mediating self-interpretation that could rationally
ground it: there is, in other words, just a perspectival knowing, and there
is no deep truth to which it does not have access. In sum, therefore, self-
deception should not be understood as an intentional deception by and of
the self (i.e., the self deceiving itself), such that the self believes and does
not believe the same thing. It should, rather, be understood as a selective
seeing and not seeing with respect to different aspects of the same thing
(which then gives rise to the particular beliefs that we form with respect
to the thing), and this selective seeing and not seeing is simultaneously a
condition and consequence of the self.
Even though the self is condemned to bad faith and (derivatively) self-
deception, however, there are still more or less egregious forms that bad
faith can take with respect to what might be called the regulative ideal
of good faith. Although all beliefs are metastable, given that the struc-
ture of the self is itself metastable, there are beliefs that are more or less
metastable because the broader structure of beliefs, and, finally, the self
with which they are entwined, are more or less coherent. This means, in
turn, that there are (sociohistorically engendered) initial projects that are
more or less coherent, given that it is the initial project that establishes
the framework of a person’s belief structure in the first place. Under-
stood in this way, a modified notion of good faith would entail a relatively
high degree of openness, which is a condition of having a critical attitude
toward new evidence. Thus, although we all unavoidably miss what appear
to be readily apparent facts (from another person’s perspective) by virtue
of the self’s necessarily limited perspective, it would not be unreasonable
to say that the less likely it is that obtrusive and relevant facts in a situa-
tion will go unacknowledged, the less egregious the form of bad faith and,
therefore, the less the tendency toward self-deception. Put differently,
although there are limits to our epistemic openness, since, as the price
of intelligibility, an initial project precludes us from being infinitely fluid
in terms of our self-identities and, therefore, infinitely open in terms of
new evidence, the questions are how selective our selective seeing is, how
much evidence can be accommodated within the existential framework

established by our initial projects, and, ultimately, how much contradic-
tory evidence it takes for us to see the untenability of our current way of
being, which should lead us to another initial project, a new existential
comportment that, all things considered, would be somewhat less in bad
faith. This account of good faith, it should be observed, is structurally akin
to the account of sociohistorical truth that was previously provided: like
sociohistorical truth, good faith can be neither predicated nor actualized,
and just as the social truth toward which history should aim at any partic-
ular time is the overcoming of the deceptive sociohistorically conditioned
beliefs that exist at that particular time, the good faith toward which we
should each individually aim at any particular time is the overcoming of
the self-deceiving aspects of the initial project that exist at that particular
time. In both cases, deception and its transcendence are contextualized.
In the final analysis, however, there is a basic asymmetry here, for the
project of overcoming bad faith (and, derivatively, self-deception) runs up
against the limits of social categoriality, which is its relevant context. Even
Sartre, who is usually reproached for offering an overly individualistic
and voluntaristic account of self-consciousness in Being and Nothingness,
implicitly connects the project of overcoming bad faith with the proj-
ect of overcoming destructive social relations in two highly discussed but
cryptic footnotes, both of which he wisely refuses to enlarge on. At the
end of his discussion of bad faith, Sartre provocatively (if unconvincingly)
holds open the possibility that we might be able to “radically escape bad
faith” through a “self-recovery of being,” but as he goes through the book’s
social chapters, and, in particular, his discussion of sadomasochistic inter-
personal relations, he makes clear that this purportedly ontological recov-
ery is really a social one, declaring that nothing he has said “precludes
the possibility of an ethics of deliverance and salvation.” 30 This indirectly
suggests, as Adorno more directly declares, that “wrong life cannot be
lived rightly,”31 which means that the stuff of our self-identities (initial
projects) and, thereby, self-deceptions is the stuff of our social world, and
that there are no individual “self-recoveries of being.” Nevertheless, this
recognition itself is surely not for nothing, for it contributes to a less self-
deceptive understanding of our self-identities and self-deceptions. More-
over, it makes clear that remaking the social world is in our enlightened
self-interest instead of a misguided altruism, as the current ideologues
would have it.
30. Sartre, Being and Nothingness, 116 and 534, respectively.

31. Theodor W. Adorno, Minima Moralia: Reflections from a Damaged Life, trans. E. F.
N. Jephcott (London: New Left Books, 1974), 39.
Duplicity Makes the Man

Or, Can Animals Lie?
Kelly Oliver
Duplicity: The quality of being ‘double’ in action or conduct; the

character or practice of acting in two ways at different times, or
openly and secretly; deceitfulness, double-dealing.
—Oxford English Dictionary
But an animal does not feign feigning ..... Speech begins only with
the passage from the feint to the order of the signifier, and that
the signifier requires another locus ..... For the Speech borne by the
signifier to be able to lie, that is, to posit itself as Truth.
—Jacques Lacan, Écrits (2006 )
The figure of the animal comes to the surface therefore in this

difference between pretense and deception.
—Jacques Derrida, “And Say the Animal
Responded?” (2003)
Once psychoanalysis enters the scene, the distinction between truth and
deception becomes mired in the murky mess of the unconscious. Inso-
far as unconscious forces drive us beyond our control and even beyond
our knowledge, then we are all and always a bunch of liars. Our motives
remain opaque and beyond our grasp, and our words always say more
than we intend. Indeed, as Freud describes it, the human psyche revolves
around deception: unconscious desires and fears sneak into consciousness;
repression works to hide traumatic memories, while the repetition com-
pulsion tricks us into reliving those traumas in new forms; dreams disguise
the truth of the psyche through processes of condensation and displace-
ment to fool the ego’s censors; neuroses pass themselves off as physical
ailments; and the more we deny something in the course of analysis, the
more likely that it is true.
Lacan takes these Freudian insights further and insists that truth is
nothing more than the ability of speech to lie; in other words, speech
104
Duplicity Makes the Man 105
is nothing more than the process of making lies appear to be true. And
the capacity to make what is false appear as true is uniquely human.
While animals can pretend and can feign, according to Lacan, they cannot
make what is false appear true, which requires pretending to pretend or
feigning the feint. Conversely, for Lacan, the double deception of speech
constitute man as human. It is man’s duplicity, his double pretense, his
double-dealing, that folds the feint back onto itself, that makes man
unique among the animals.
What is this duplicity of speech that animals lack? And how can we
distinguish between pretending and pretending to pretend? What are the
implications of maintaining that only man can lie? In this essay, I take up
these questions in order to situate the role that animals play for Lacan,
and to signal the role they play for Western thought more generally. My
thesis is that Lacan uses animals to make his work appear scientific so
that he can more persuasively outline the dynamics of the human psyche.
In other words, animals appear in his work to add rhetorical force to his
descriptions of the distinctive qualities of man. We see logic familiar in
the history of philosophy whereby animals are used to shore up the bor-
ders of man; in other words, animals are called as witnesses to man’s supe-
riority.1 Within this logic, animals are more than the constitutive outside
of man. They also teach man how to be human; man is human by virtue
of animal pedagogy.
DOUBLE-DEALING ANIMALS
Tracking animals in Lacan’s work, one is struck by the various functions

they play. In some places, Lacan points to a continuation between animals
and man; in others, he insists on a radical distinction between animals and
man; and many times, his conclusions about the divide between man and
animal are premised on the ways animals exhibit certain characteristics
usually associated with humans. His work displays an ambivalence about
man’s relation to animals, which I will link to his ambivalent relation to
science. On the one hand, his disdain for the empirical methods of behav-
ioral psychology and the science envy of logical positivism in philosophy
is explicit throughout this work. On the other hand, he frequently cites
studies involving animals to substantiate claims about human psychol-
ogy, and he devises complicated graphs to illustrate his theories. For my
purposes, I will attend to some of his many remarks about animals to
1. I have analyzed this logic in Rousseau, Herder, and Derrida in several articles.
See my “Animal Pedagogy: The Origin of ‘Man’ in Rousseau and Herder,” in Culture,
Theory and Critique 47, 2 (2006), 107–31; “Tropho-ethics: Derrida’s Homeopathic Purity,”
in Harvard Review (2007); and “Sexual Difference, Animal Difference: Derrida and
Difference ‘Worthy of Its Name,’ ” forthcoming. My forthcoming book Animal Pedagogy
will continue this line of thought by engaging the writings of Heidegger, Agamben, Merleau-
Ponty, Freud, and Kristeva.
diagnose Lacan’s ambivalence toward science evidence by his symptom-

atic ambivalence toward animals.
In a lecture on the state of psychoanalysis in 1956—a lecture filled
with irony and biting humor like most—Lacan says:
I have always been struck, while taking my little dog for a walk so he could
attend to his needs, by what we could glean from his activities that would
help us analyze the capacities that make for man’s success in society, as well
as the virtues that Antiquity’s thinkers meditated upon under the heading
of Means-to-an-End [Moyen-de-Parvenir]. I hope that this digression will,
at the very least, dispel the misunderstanding of attributing to me the doc-
trine of a discontinuity between animal psychology and human psychology,
which is truly foreign to my way of thinking. (2006, 404)
Note that Lacan’s insistence that his position allows for a continuity
between animal and human psychology is formulated in the negative,
with words such as “digression,” “dispel,” “misunderstanding,” “disconti-
nuity,” and “foreign.” He does not assert a continuity but rather denies
a discontinuity, and not between animals and humans, but between one
type of psychology and another. Even his style might be read as signaling
an ambivalence about animals in relation to men. Indeed, whatever he
learns from his dog has less to do with his dog than with himself (given
that for Lacan, relations between human subjects always comes back to a
self-relation, this should be no surprise).
The tongue-in-cheek quality of the passage is signaled in the beginning
with his allusion to a little doggy-do, which perhaps reflects his opinion
of behaviorist observations. In the lecture, this passage follows a sugges-
tion that behaviorists who find the measure of man in the animal might
improve their studies by considering “trace behavior” in both. It seems that
behaviorists discover something about man by observing animal behavior,
but the continuities may not be what they think; instead they may lie in
what Lacan calls “trace behaviors” that cannot be observed using their
crude experimental techniques (which are not even on par with Lacan’s
observations while walking his dog). These trace behaviors are likely
linked to imaginary formations consistent in both animals and humans
that constituteperception as a sort of “residue of the real,” which behav-
iorists ignore (see Lacan 1988b, 257). As such, this passage may indicate
ambivalence about behaviorism as much as about animals. Indeed, at the
same time Lacan invokes their findings and cites other studies in biology,
ethnology, and zoology, he jeers at their methods. And while he endorses
studies that demonstrate the existence of the imaginary or the function
of the imago or image in animals, he discounts conclusions that implicate
what he calls the symbolic, which he reserves for man alone.
Consider another negatively worded passage noteworthy because in
this rare instance, Lacan conjures continuity between the symbolic in ani-
mals and in humans: “It can be seen that I do not shrink from seeking the
origins of symbolic behavior outside the human sphere” (2006, 225). This
remark is sandwiched between criticisms of logical positivist attempts to
catalogue and thereby exhaust the meaning of speech and behaviorist

attempts to excise superfluous speculations by turning to the supposed
certainty of empirical experimentation on animals. Lacan goes on to
describe the symbolic order as one of exchange, as in the exchange of gifts
that seals a pact between individuals or tribes (see 225). While exchange
is necessary, it is not sufficient for symbolic language. Lacan imagines that
if it were, a group of sea swallows might possess a rudimentary language:
Were this the case, one would find a first approximation of language among
sea swallows, for instance, during display, materialized in the fish they pass
each other from beak to beak; ethnologists—if we must agree with them
in seeing in this the instrument of a stirring into action of the group that is
tantamount to a party—would then be altogether justified in recognizing a
symbol in this activity. (225)
But Lacan insists that they are not because, as he says,

For even if there appeared among the sea swallows some kaid of the colony
who, by gulping down the symbolic fish from the others’ gaping beaks, were
to inaugurate the exploitation of swallow by swallow—a fanciful notion
I enjoyed developing one day—this would not in any way suffice to repro-
duce among them the fabulous history, the image of our own, whose winged
epic kept us captive on Penguin Island; something else would still be needed
to create a “swallowized” universe. (228)
If Lacan’s flights of fancy regarding swallow-slavery are indicative of his

seeking the origins of the symbolic outside of the human, then perhaps he
is having us on a bit when he says that he doesn’t shrink from doing so.
Lacan suggests that symbolization requires exchange, exploitation, and
even enslavement (he insists that only humans are capable of enslaving
others), but something more is also necessary, something animals lack.
And it is this “something more” that turns our world into a humanized
world, something the swallows can never do with their “world” such as
it is. “This ‘something else’ completes the symbol, making language of
it” (Lacan 2006, 228). It does so by freeing the sound or word from the
here and now of experience and making it a concept that not only erases
the thing to which it refers but also makes it permanent, only now in the
world of symbols. This duplicitous operation of substitution of concept
for thing is supposedly unique to humans, in that it inherently links the
imaginary and symbolic realms in ways that, at least in principle, forever
cut them off from any immediate access to the real. Moreover, this link
means that in man and only in man the imaginary exists at the mercy, or
we could say pleasure, of the symbolic. Presumably this is why human
subjects are susceptible to analysis—the “talking cure”—while animal
subjects are not.
While Lacan dispels discontinuity and denies “shrinking” in the face of
the animal, at the same time, he insists that only man is capable of speech.
Sometimes it seems as if he is setting out criteria for speech, and thereby
for human subjectivity, and one by one compromising them in relation to
animals; and yet always maintaining a shifting “something more” reserved

for man alone. This something more, this trace, is what Lacan sees and
behaviorists miss when they look at animals. For example, Lacan discusses
experiments where dogs are tied to tables and trained to expect meat on
the signaling of a bell but are given apples instead, and the raccoon who
is “taught, by a judicious conditioning of his reflexes, to go to his food box
when he is presented with a card on which the meal is to be served is
printed” and who, “if the service disappoints him . . . comes back and tears
up the card that promised too much, just as a furious woman might do
with the letters of a faithless lover” (2006, 226). Lacan complains that
the behaviorists conducting these experiments tackle the problem of lan-
guage by trying to “grab it by the throat,” which altogether ignores its
essential imaginary dimension.
In his most famous essay, “The Mirror Stage,” and elsewhere, he invokes
animals in order to document the role of the image in subject formation
and ultimately in the symbolic exchange that gives way to language. In
“The Mirror Stage,” the pigeon and the locust are presented as evidence
for the essential role of the imaginary in subject formation. Ten years later,
Lacan again trots out the pigeon and locust as witnesses to the role of the
imaginary in man. Here, the role of the animal in contributing scientific
credence to Lacan’s theories is even more explicit:
I think, therefore, that I can designate the imago as the true object of psy-
chology, to the exact extent that Galileo’s notion of the inert mass point
served as the foundation of physics Those who do not wish to under-
stand me might object that I am begging the question and that I am gra-
tuitously positing that the phenomenon is irreducible merely in order to
foster a thoroughly metaphysical conception of man. I will thus address
the deaf by offering them facts which will, I think, pique their sense of the
visible, since these facts should not appear to be contaminated, in their eyes
at least, by either the mind or being: for I will seek them out in the animal
kingdom. (Lacan 2006, 153–54)
At the same time he bats around the scientific crowd, he is playing to

them by trying to make his theories interest them in particular; and he
does so by using studies of animals to prove his theses about man in a
more scientific and less metaphysical or speculative way. Given the irony
of his style, the duplicity of this turn to the empirical and observable in
order to prove the role of the image or invisible in psychic dynamics is
perhaps not lost on him.
If it is not the imaginary that separates man from animals, what is it? In
answering this question, to which he returns again and again, Lacan seems
to taunt the animal by continually giving something and then taking it
away. And, as we have seen, this fort-da with the animal psyche repeatedly
serves as rhetorical bolster for his theories of the human psyche. It seems
that language is the main barrier between man and animal. But what is
it about language that animals cannot muster? Lacan discusses various
aspects of language that might qualify: use of signs or codes, substitution
or displacement, the ability to conceptualize, the ability to respond, and

more specifically the ability to lie.
In “The Function and Field of Speech and Language in Psychoanalysis,”
Lacan makes a distinction between language and speech that turns on the
distinction between the use of codes by animals and the use of symbols by
man. He enlists the swallows and raccoon mentioned earlier, along with
bees, dogs, and other critters. He argues, again against behaviorists, that
although animals can be trained to react to verbal stimulus, their reactions
are never meaningful responses in terms of the symbolic order. Humans
can also be trained to react to stimuli in similar ways, but that does
not prove that they are like animals. Lacan lambasts Jules Masserman’s
studies of animals (including the dog and raccoon) and humans in terms
of their “responses” to bells and verbal commands. He discusses a study
in which humans were “trained” to contract their pupils on command
by exposing them to bright light in connection with the command and
then dispensing with the light. Against Masserman’s conclusion that these
contractions were “visceral reactions to the idea-symbol ‘contract’,” Lacan
muses.
I would have been curious to know whether subjects trained in this way
also react to the enunciation of the same term in the expressions ‘marriage
contract’, ‘contract bridge’, and ‘breach of contract’, and even when the
term is progressively shortened to the articulation of its first syllable alone:
contract, contrac, contra, contr For either the effects would no longer be
produced, thus revealing that they do not even conditionally depend on the
semanteme, or they would continue to be produced, raising the question
of the semanteme’s limits In other words, they would cause the distinc-
tion between the signifier and the signified, so blithely confounded by the
author in the English term “idea-symbol.” (2006, 226–27)
Lacan argues that words have meaning within symbolic systems by virtue
of their relationships rather than existing independently as idea-symbols.
This distinction between idea and symbol as sign again implicates the
animal.
Lacan insists that the logical positivists and behaviorists who reduce
language to nothing more than signs are mistaken. Looking for evidence,
as he says, like “bloodhounds on the scent,” he once more turns to “the
animal kingdom” to prove his point, and again with his typical irrever-
ent reverence for science: “I shall show the inadequacy of the conception
of language as signs by the very manifestation that best illustrates it in
the animal kingdom, a manifestation which, had it not recently been the
object of an authentic discovery, would have to have been invented for
this purpose” (2006, 245). Lacan gives the example of a certain bee’s
“wagging dance” that directs other bees to the location of food. Although
the bee employs a sign as a signal or even as a code, it is not language:
We can say that it is distinguished from language precisely by the fixed

correlation between its signs and the reality they signify. For, in a language,
signs take on their value from their relations to each other in the lexical
distribution of semantemes as much as in the positional, or even flectional,
use of morphemes—in sharp contrast to the fixity of the coding used by the
bees. The diversity of human languages takes on its full value viewed in this
light. (2006 245–46)
Note that it is the fixed nature of bee codes that teaches us the full value
the diversity of human languages—although we might wonder if all bees
use the same codes and, given that there are significantly many more
types of insects than any other class of animal, it seems likely that there
are many times more insect codes than there are human ones. In any case,
the bees provide a pedagogical lesson for the psychoanalyst in terms of
human languages.
In addition to a system of signs defined by their relations to each other,
what the bee lacks, it seems, is imagination. Indeed, it may be a lack of imag-
ination that prevents them from relating their signs to each other rather
than to the location of dinner. But, as noted, Lacan has already granted
imagination to animals, so it must be more than the too direct connection
between the sign and the real that is the “bee” in Lacan’s bonnet. We get a
sense of what this is when we return to the example of the swallows, who
lack the “something more” of human language. Recall that it is the lack of
concepts that prevent swallows from “swallowizing” the universe. Even if
animals have signs and imagination, they do not have “the permanence of
the concept” (see Lacan 2006). Here, Lacan concludes that animals are
incapable of the generalization necessary for conceptualization.
Elsewhere, however, Lacan says in passing (again complaining about
behaviorist methodology) “there are enough really rather tiresome labora-
tory experiments, which show that if one holds an octopus, or any other
animal, with sufficient doggedness in from of a triangle, they will rec-
ognize it in the end, that is to generalize it” (1988b, 322). So, if animals
are capable of signs, of imagination, and even of concepts, what is lack-
ing to make their forms of communication into language? Discussing the
swallows and the bees, Lacan suggests that it is the capacity to make the
concept stand in for the thing. In other words, animals are incapable of
substitution or displacement, operations necessary for meaningful speech.
It is the play between presence and absence that engenders language:
“Through what becomes embodied only by being the trace of a nothing-
ness and whose medium thus cannot be altered, concepts, in preserving
the duration of what passes away, engender things It is the world of
words that creates the world of things” (2006, 228–29). Other animals,
then, cannot quite muster the absence at the heart of representation that
makes man the speaking animal.
In another lecture, however, continuing his fort-da giving and reeling
back in, Lacan suggests that animals are indeed capable of displacement:
Let us say that, in the animal world, the entire cycle of sexual behaviour is
dominated by the imaginary. On the other hand, it is in sexual behaviour
that we find the greatest possibilities of displacement occurring, even in

animals ...... The possibility of displacement, the illusory, imaginary dimen-
sion, is essential to everything pertaining to the order of sexual behaviour.
(1988a, 138)2
This conclusion follows a discussion of the sexual behavior of sticklebacks,

who engage in a kind of zigzagging mating dance instigated by beautiful
colors on the belly and back of the male. 3 “We can,” says Lacan, “quite
easily make a cut-out which, even when poorly put together, will have
exactly the same effect of the female” (122). These fish are easily fooled
when it comes to sex; they see what they want to see, which provides
Lacan with further evidence of the function of the image in both animal
and human psyches, particularly when it comes to sex. He says, “what
does the development of instinctual functioning teach us in this respect?
The extraordinary importance of the image” (137). Like the pigeon and
the locust before them, these stickleback fish teach us the importance of
the image in the human psyche. In his discussion of the stickleback’s mat-
ing rituals, Lacan takes their lessons and applies them to neurosis: “sexual
behavior is quite especially prone to the lure. This teaches us something
which is important in working out the structure of the perversions and the
neuroses” (122). This raises the question of whether or not animals can be
perverts and neurotics. In any case, it is once again animal pedagogy that
provides Lacan with evidence and insight into the psyche of man.
If animals are capable of imagination, conceptualization, and displace-
ment, what does man possess that makes him singular among them?
What is this “something more”? Throughout his work, Lacan consistently
maintains that the human subject is constituted by a fundamental alien-
ation or gap that renders the imaginary necessary compensation for the
subject’scongenital faults, what he calls man’s “prematurity” at birth.
Because man’s infancy and dependence is much longer than that of other
animals, he is incapable of motor coordination that allows him to affect
his surroundings for the first several years of life. This incapacity instigates
imaginary, even hallucinatory, compensations for the lack of motor skills.
The infant’s reflection in the mirror serves as an image of wholeness and
coordination that conjures what Lacan calls a misrecognition of a uni-
fied experience and agency that are not yet or truly the infant’s own.
2. Lacan makes it clear that this imaginary function is the same in animals and man:
“Is this true for man, yes or no?...... This is nothing other than the imaginary phenomenon
which I just spelt out in detail for you in the animal” (Lacan 1988a, 138).
3. Reminiscent of Merleau-Ponty, Lacan says: “what comes to play in releasing
the complementary behavior of the male and female sticklebacks? Gestalten” (1988a,
137). Merleau-Ponty also uses the example of the stickleback to make the case for the
importance of mimesis in his Nature lectures (2003, 196). It is clear that Lacan was
influenced by Merleau-Ponty’s theories of perception (e.g., see 1988a, 58). For a discussion
of the relationship between Lacan and Merleau-Ponty on the question of the stickleback,
see my Animal Pedagogy, forthcoming.
This misrecognition ushers in human dependence on the image and on

the Other (both in terms of systems of meaning and in terms of human
others—like the caregiver who holds the infant in front of the mirror) to
mediate his experience. It is the gap between the infant’s mirror image
and the reality of his lack of coordination that make him unique among
the animals. This is why, according to Lacan, while the human infant
delights in his mirror image, the chimp finds his mirror image threatening.
Presumably, the chimp recognizes the image as a threat to his immedi-
ate hold on the world, while the human misrecognizes the image as the
source of his own agency. It is in the gap between the image and the real
that the symbolic can be inserted, which is why the gap itself makes man
distinctive (see Lacan 1988a, 141–42). Lacan concludes:
Living animals are sensitive to the image of their own kind. This is an abso-
lutely essential point, thanks to which the whole of living creation isn’t an
immense orgy. But the human being has a special relation with his own
image—a relation of gap, of alienating tension. That is where the possibility
of the order of presence and absence, that is of the symbolic order, comes
in. (1988b, 323)4
Lacan describes this gap as “a certain biological gap,” which “already

assumes the lack” (323); this biological, or we might say animal, lack is
what animals lack.
TRUE LIES
In his first posthumously published book, L’animal que donce je suis,

Jacques Derrida probes some of the appearances of animals in Lacan’s
Écrits; and he challenges Lacan’s desire for a fixed criterion to distinguish
man from animals. As Derrida reminds us throughout his book, Lacan
is not so much concerned with giving these ethereal characteristics to
animals as with questioning whether or not humans possess them in the
fixed way indicated by philosophers who put stock in some absolute limit
between man and animal. Derrida points out that on Lacan’s account,
paradoxically, man is superior to animals because of his imperfection
and lack, for which he must compensate; thus, man “received speech and
4. Lacan’s idea that species stick to their own images and that there is no “orgy”
in nature seems extremely conservative, given his theory of the role of the image and
the imaginary in both animals and humans. Moreover, it operates with an extremely
conservative notion of sex as reproductive sex and a conservative notion of reproductive
sex at that. What about the reproductive relation between bees and flowers? What about
nonreproductive sex acts of animals with inanimate objects, plants, or animals of other
species? What about the way sexual behavior in the very animals Lacan cites changes with
their changing environment? In “Sexual Difference, Animal Difference” (forthcoming), I
argue that attending to the hugely diverse sexualities of various animal species opens up
new ways of conceiving of sex and reproduction.
technics only inasmuch as he lacks something” (2003, 124). In the words

of Kalpana Seshadri-Crooks, for Lacan
the animal’s existence in the imaginary, where it merely reacts to “vital situ-
ations,” also means that it cannot be conditioned by its own word—as in
vouching for something or lying—which more or less expels it from time
and mortality. The implication here is that the animal lacks the lack that
constitutes the human subject. It is a wholly sufficient entity in that it lives
in a state that is anterior to good and evil. It is neither a subject of language,
nor subjected to language in the manner that the human necessarily must
be because of the biological fact of his or her premature birth. Thus, as Der-
rida points out, the human subject in Lacan is constituted by its lack, which
is what distinguished human from animal. (2003, 100)
Insofar as the animal lacks the lack constitutive of humanity, it appears

as a mythic figure of wholeness and unity; the animal comes to repre-
sent man’s lost archaic ancestor, which triggers both nostalgia for its
perfection and hostility toward it for man’s relative fragmentation and
imperfection.
Seshadri-Crooks supplements Derrida’s criticisms of Lacan, suggesting
that although Derrida is onto something, there are resources in Lacan’s
works that help us to challenge the fixity of the limit between man and
animal. She argues that in his reading of Freud’s Totem and Taboo, Lacan
exposes the myth of the animal ancestor and of animal perfection; and
that by so doing, he “leads us to ask: does the animal exist?” (2003, 104).
Obviously this is not to say that animals don’t exist. Rather, it is a chal-
lenge to the general concept of the animal, which as Derrida so forcefully
argues corrals nearly infinite numbers of diverse living beings into one
general category and assigns to them one name: “This agreement con-
cerning philosophical sense and commonsense that allows one to speak
blithely of the Animal in the general singular is perhaps one of the great-
est, and most symptomatic idiocies [betises] of those who call themselves
human” (Derrida 2002, 409). Humans define themselves against all other
living creature by denying vast differences between those beings (and
thereby differences among themselves). This move to divide man from
animal and then privilege man sets up more hierarchical binaries invoked
to justify more violence and “beastly” behavior. Derrida points out that
only humans can be called “beastly.” More to the point, the animal’s lack
of lack has been associated with a lack of evil or cruelty, which are seen as
the sole providence of men. Paradoxically, man is superior to the animal,
then, because he is capable of cruelty and evil while the animal is not.
Man claims a privilege over the animal because while it is innocent and
true, he is corrupt and duplicitous.
The centerpiece of Derrida’s engagement with Lacan is the following
passage from Écrits, which Derrida also quotes at length:
this Other, distinguished as the locus of Speech, nevertheless emerges as
Truth’s witness. Without the dimension it constitutes, the deceptiveness of
Speech would be indistinguishable from the feint, which, in fighting or sex-

ual display, is nevertheless quite different. Deployed in imaginary capture,
the feint is integrated into the play of approach and retreat that constituted
the first dance, in which these two vital situations find their scansion, and the
partners who fall into step with it find [that] which I will dare to write as
their “dancity.” Moreover, animals show that they are capable of such behav-
ior when they are being hunted down; they manage to throw their pursuers
off the scent by briefly going in one direction as a lure and then changing
direction. This can go so far as to suggest on the part of game animals the
nobility of honoring the parrying found in the hunt. But an animal does not
feign feigning. It does not make tracks whose deceptiveness lies in getting
them to be taken as false, when in fact they are true—that is, tracks that indi-
cate the right trail. No more than it effaces its tracks, which would already be
tantamount to making itself the subject of the signifier. (2006, 683)
In this passage, Lacan maintains that animals react to other animals (and to
humans) with a “dance” that indicates relationality with others. But their
very being in the world is not mediated by the dance, nor are they moved
in it by the Other in the way that constitutes humanity. The human world
is a world of signification through which subjects are constituted by virtue
of the Other of meaning that is inherited from others on whom the fledg-
ling subject is dependent for years. And although animals can pretend and
even lure, they can neither erase their tracks nor make false tracks appear
true. In other words, they cannot do what humans do when speaking,
namely, make a symbol stand in for the thing. Representation erases the
thing in favor of the concept that substitutes token for reality; animals are
incapable of this double operation of erasure and substitution, the inher-
ent duplicity of speech. Speech is the human activity of putting the feint
of erasure and substitution in the service of truth; we use words to tell the
truth about things by first erasing those very things. As Nietzsche might
say, we hide truth behind a bush and then praise ourselves when we find
it. The animal may deceive its predator or its prey, but it doesn’t deceive
itself. Animals may be capable of deception, but man is the self-deceptive
animal.
Lacan insists that speech, as distinct from language, opens the field of
signification to “noise” or superfluous meanings that “speak” unconscious
desire. He says:
the antinomy immanent in the relations between speech and language thus
becomes clear. The more functional language becomes, the less suited it is
to speech, and when it becomes overly characteristic of me alone, it loses its
function as language . . . what is redundant as far as information is concerned
is precisely what plays the part of resonance in speech. For the function
of language in speech is not to inform but to evoke. What I seek in speech
is a response from the other. (2006, 246–47)
This passage suggests that animal existence is purely utilitarian and that ani-
mals’ reactions always function to promote life; whereas human existence
is characterized by meaning in excess of use-value, the chatter with which
we continually and futilely try to fill the void or gap between us and the
world opened by our premature birth. This superfluous stuff is the stuff of
response and not of reaction; and, Lacan claims that animals are incapable
of any true response. Given Lacan’s repeated suggestion that there is never
any real communication between human subjects and that any appearance
of one is merely self-delusion, we might wonder what exactly he means
by “response.” Perhaps, for Lacan self-delusion is the essence of response,
which becomes something like an echo chamber or hall of mirrors.
Elsewhere, following Derrida, I have challenged whether, particu-
larly in light of psychoanalytic theory, we can maintain the distinction
between reaction and response in human beings. 5 Here, suffice it to say
that Lacan insists that “a reaction is not a response” and without evidence
from the “animal kingdom” to prove it, he is left with bald assertion and
math (see 2006, 247). His ambivalence toward both sciences, particularly
behaviorism, evidenced by his ambivalence toward animal studies, leads
him to math. It seems that as he moves closer to this “something more”
that cannot be perceived in the “animal kingdom,” he turns to mathemes
rather than to animal studies to make his work appear more rigorous; but
Lacan’s math fetish is beyond the scope of this essay, which prefers to
engage his animal fetish.
In L’animal, Derrida “deconstructs” the distinction between reaction
and response, along with the distinction between pretending and pre-
tending to pretend, or feint and feigning the feint. After all, how can we
tell the difference between pretending and pretending to pretend? How
can we distinguish first-order pretense from second-order pretense? Can
we so easily distinguish the two senses of duplicity: double actions and
double-dealing? How can we distinguish pretending from lying? Derrida
summarizes Lacan on pretense versus lie:
According to Lacan, the animal would be incapable of this type of lie, of this
deceit, of this pretense in the second degree, whereas the “subject of the sig-
nifier,” within the human order, would possess such a power and, better still,
would emerge as a subject, instituting itself and coming to itself as a subject
by virtue of this power, a second-degree reflexive power, a power that is
conscious of being able to deceive by pretending to pretend. (2002, 130)
The animal, then, is supposedly incapable of a second-order true lie

because although it can pretend, it is not conscious of its pretense. If
Lacan has once again reduced the divide between man and animal to
consciousness, then he is not only begging the question of the border
between the two but also treading a precarious line, given the limits of
consciousness in the psychoanalytic account of the human psyche. What
does it mean, post-Freud and post-Lacan, to say that man is conscious?
And, post-Freud and post-Lacan, how can we distinguish the second-
5. See my “Tropho-ethics: Derrida’s Homeopathic Purity” (2007), and “Sexual

Difference, Animal Difference: Derrida and Difference ‘Worthy of Its Name’ ”
(forthcoming).
order reflexivity of consciousness from its roots in unconscious, unreflex-

ive operations “behind the scenes”? Once consciousness becomes a sleight
of hand, it becomes difficult to maintain any distinction between first and
second orders or conscious response and “unconscious” or instinctual reac-
tion. How can we tell the difference between an animal that operates in
relation to the world without the mediation of the Other and the human
who operates by virtue of a mediation that he disavows?
Contra Lacan, Derrida argues that every seemingly simple pretense
could be a pretense of pretense and every pretense of pretense could in
actuality be simply pretense. And “as a result, the distinction between
lie and pretense becomes precarious” (Derrida 2002, 136). If the animal
is capable of pretense, then it has already taken the other into account.
And once it does this, then it is possible for its pretense to become a
pretense of pretense, which as Lacan describes it is always dependent on
the other. Once a creature starts thinking about how to deceive another,
and once the other creature can be deceived, it is difficult to discern the
levels of deception possible, since it (deception) happens in the relation-
ship itself. On the other hand, once deception is taken to be the essence
of consciousness and of speech, then it becomes problematic to base any
absolute border between man and animal on the truth of consciousness
or speech. In addition, once Lacan takes us to the order of the “something
more” beyond any observable behavior, it becomes impossible to perceive
the difference between first-order and second-order pretense, since the
difference is really only all in the mind.6
Derrida also challenges Lacan’s claim that animals cannot erase their
tracks whereas man can. Derrida reminds us that throughout his (Derrida’s)
work he has shown that “the structure of the trace presupposes that to trace
amounts to erasing a trace as much as to imprinting it” (2002, 137). And he
maintains that even a simple pretense involves some kind of erasure insofar
as it moves in a fictional direction. Even the pigeons, locusts, and sticklebacks
falling in love with cardboard cutouts or their reflections in mirrors have
gone some distance in changing, suspending, or erasing their relation to “real-
ity.” Even more so, the animals who fake out their predators and their prey
by playing dead or wounded or imitating someone else render “impercep-
tible a sensible trace” (see 137). When they are successful, these duplicitous
beasts pass for what they are not and thereby erase their “truth.”
Again following Derrida, we might wonder whether humans can ever
erase their tracks, not only in the metaphorical or metaphysical senses
6. Lacan describes the shift from the function of the image in the animal to the
function of the image in man as “imperceptible”: “but what is new in man is that
something is already sufficiently open, imperceptibly shifted within the imaginary
coaptation, for the symbolic use of the image to be inserted into it” (1988b, 322–23).
And it is this imperceptible shift that opens the way to speech and language, a path
closed to the animal. The difference between man and animal, then, is not something that
can be perceived on the level of the imaginary, precisely the level on which Lacan fixes
the animal.
evoked by Derrida but also in the physical sense, or we could say, in the
animal sense. As a thought experiment—and some environmental groups
have tested this out—try to imagine living on the earth without leaving a
trace, even for one day, for one hour, for one minute. It is as impossible for
man to erase his tracks as for any animal. But unlike most animal tracks
and traces, human tracks leave irreversible damage. And although some
animals also wreak havoc on the environment, none has done as much
as man to destroy that which sustains him. Can humanity learn to tread
lightly on the earth? Given the growing concerns about the environment
and the effect of human waste on it, and the dawn of the consequences
of global warming caused by man’s lack of concern for erasing his tracks,
wemight wonder once again what we might learn from the animals. Per-
haps they could teach us a few new steps in the dancity through which all
creatures live (and die if not lie) together.
References
Derrida, Jacques. 2006. L’animal que donc je suis. Paris: Galilée.
——— . 2005. Rogues. Trans. Pascale-Anne Brault and Michael Naas. Stanford:
Stanford University Press.
——— . 2004. “The Animal That Therefore I Am (More to Follow).” In Animal
Philosophy, ed. Peter Atterton and Matthew Calarco. London: Continuum
Press, 113–28.
——— . 2003. “And Say the Animal Responds?” In Zoontologies: The Question of the
Animal. Trans. David Wills. Ed. Cary Wolfe. Minneapolis: University of Min-
nesota Press, 121–46.
——— . 2002. “The Animal that Therefore I am (More to Follow). Trans. David
Wills. In Critical Inquiry 28 (2) (Winter), 369–418.
Lacan, Jacques. 1988a. The Seminar of Jacques Lacan. Bk. 1. Freud’s Papers of Tech-
nique 1953–1954. Trans. John Forrester. Ed. Jacques-Alain Miller. New York:
Norton.
——— . 1988ab. The Seminar of Jacques Lacan. Bk. 2. The Ego in Freud’s Theory
and in the Technique of Psychoanalysis 1954–1955. Trans. Sylvana Tomaselli. Ed.
Jacques-Alain Miller. New York: Norton.
——— . 2006. Écrits: The First Complete Edition in English. Trans. Bruce Fink.
New York: Norton.
Oliver, Kelly. 2006. “Animal Pedagogy: The Origin of ‘Man’ in Rousseau and
Herder.” Culture, Theory and Critique 47 (2), 107–31.
——— . 2007. “Tropho-ethics: Derrida’s Homeopathic Purity.” Harvard Review,
37–57.
——— . Forthcoming in Hypatia, 24:2 2009.“Sexual Difference, Animal Differ-
ence: Derrida and Difference ‘Worthy of Its Name.’ ”
——— . Forthcoming.Biting the Hand that Feeds You: Animal Pedagogy and the Sci-
ence of Kinship (Columbia 2009).
Merleau-Ponty, Maurice. 2003. Nature: Course Notes from the Collège de France.
Trans. Robert Vallier. Comp. Dominique Séglard, Chicago: Northwestern Uni-
versity Press.
Seshadri-Crooks, Kalpana. 2003. “Being Human: Bestiality, Anthropophagy, and
Law. Umbr(a), 97–114.
Lie Catching and Microexpressions
Paul Ekman
In this chapter I consider two separate but related issues. The first is why
most people are unable to accurately judge from demeanor whether
someone is lying or truthful (Ekman and O’Sullivan 1991; Ekman, Frank,
and O’Sullivan 1999; Frank and Ekman 1997). Most liars make mistakes
that could be detected but usually are missed. Both perpetrating a lie
and detecting a lie, in most people, seem to be poorly developed skills.
I will provide six explanations for why most of us do not catch liars from
demeanor.
The second topic I consider is microfacial expressions (micros) that
reveal concealed emotions. While most people do not report seeing micros,
in an hour most people can learn to do so. This raises the question, related
to the first topic, why, if it is so easy to learn, have people failed to do so?
Before addressing these issues, I will first explain how I distinguish lying
from other forms of deceit, and then discuss the evidence that suggests
that people are very poor lie catchers.
The intent of the liar is one of the two criteria I use to distinguish lies
from other kinds of deception (Ekman [1985] 2000). The liar deliberately
chooses to mislead the target. Liars may actually tell the truth, but that
is not their intent. And truthful people may provide false information—
say, bad advice from a stockbroker—but that is not their intent. A liar
has a choice; a liar could chose not to lie. We are all tempted to lie, but
we do not always do so. Lying is not irresistible; it is, by my definition, a
conscious, considered choice. I do recognize that lying can become a habit
and then be performed I don’t think you need be, it makes it awkward
with little consideration, but at least initially, all such habits began as con-
sidered choices about whether or not to do so. Presumably, a pathological
liar is compelled to lie and by my definition, therefore, is not a liar.
The second criterion for distinguishing lies from other deceptions is
that the target is not notified of the liar’s intention to mislead. A magician
is not a liar by this criterion, but Uri Geller is a liar, since Geller claimed
that his tricks were not magic. An actor is not a liar, but an impostor is.
“Let the buyer beware” is one example of an explicit warning that prod-
ucts or services may not be what they are presented to be. (Of course,
118
that warning does not appear in advertisements, nearly all of which are
designed to convey the opposite message.) Poker is still another situation
in which the rules of the game sanction and notify the players that decep-
tion will occur, and, therefore, one cannot consider bluffing to be a lie.
Sometimes notification of an intention to mislead is implicit in the
framing, to use Goffman’s (1974) term, of the situation. In real estate
transactions, the potential buyer is implicitly notified that the seller’s ask-
ing price is not the actual price the seller would accept. Various forms of
politeness are other instances in which the nature of the situation notifies
the target that the truth may not be spoken. It is not proper for the host to
scrutinize the dinner guest to determine whether the guest’s claim to have
enjoyed the evening is true any more than for the aunt to worry whether
the nephew is lying when he says that he appreciated the tie she gave him
for Christmas. In such situations, deception is expected; even if the target
might suspect that the truth is not being told, it is improper to question it.
Only certain types of deception may be allowable: The poker player can
deceive by bluffing but not use marked cards; the home seller can deceive
about the true selling price but can not conceal a known defect.
In courtship, it is ambiguous whether the parties should expect truth-
fulness. The saying “All’s fair in love and war” warns lovers not to believe all
they are told. Recent public opinion polls suggest that lies that downplay
the number of previous sexual partners are common among college-aged
adults. Yet I expect that lovers generally want to believe in the truthful-
ness of their lovers. Many popular songs testify to the betrayal felt when
lies are discovered (although some do warn that lies may be expected).
Romantic love requires that partners make collusive efforts to develop
and maintain myths about each other and the nature of the relationship.
When a police officer interviews a person suspected of a crime, the
policeman asks but typically does not expect to be told the truth, at least
not initially, if the suspect is guilty of the crime. The officer often stresses
the importance of and benefits of being truthful, though it may not actu-
ally be in the best interests of the guilty person. It is not uncommon
for the police to tell the suspect information that is not true in order to
induce a confession, and the U.S. Supreme Court has ruled that confes-
sions obtained by means of such lies are admissible. In the United King-
dom, they are not.
I differ from Bok (1982), who only considers false statements to be lies.
I argue that one can falsify without words, and one need not falsify, ver-
bally or nonverbally, to lie (Ekman [1985] 2001). Concealment is just as
much a lie as falsification, if there is an expectation that information will
be revealed. When filling out a job application that asks for a listing of all
previous employment, omitting the one from which one was fired would
be a concealment lie, for there is an obligation to reveal. In personal rela-
tionships, it is not always so clear-cut, and the liar, once discovered, and
the target may disagree about whether or not an obligation to reveal the
concealed information was in force.
Lie Catching and Microexpressions 119
Concealment and falsification are different techniques for accom-

plishing the same objective. The issue is the motive, not the technique
employed to accomplish it. If the motive is to mislead, then the choice
between falsifying or concealing is simply a matter of which technique
will work better in a given instance. Elsewhere, I have explained why
most liars would prefer to conceal rather than falsify if the situation will
allow it, and I have described some other techniques for implementing a
lie (Ekman [1985] 2001).
Now let us consider what we know about how well people can detect
lies from demeanor. The evidence that most people do poorly at catching
lies comes from the following type of experiment. Students are recruited
to lie or tell the truth about something that usually does not matter much
to them. It has no relevance to their pasts or their expected future lives.
Sometimes in a weak (in my judgment) attempt to motivate them, they
are told it is important to be able to lie, or that smart or successful peo-
ple succeed in this task. Videotapes of their behavior are shown to other
students, who are asked to identify who is lying and who is telling the
truth. Whether whether the lie catcher succeeds or fails does not matter
much to the lie catacher. Typically, most of those trying to catch the liars
perform at chance or just slightly better than chance. A major drawback
of the metaanalyses reported by Bond and DePaulo (2006) on clues to
deceit is that they included all studies, most of which contain the limita-
tions I have described, making their conclusions of little relevance to lies
in which the stakes are high, the context is relevant to the subjects’ past
and future histories, and punishment awaits liars who are caught as well
as truthful people who are misjudged as lying.
My colleagues and I (Ekman and Friesen 1974; Ekman, Frank, and
O’Sullivan, 1999) have incorporated these features in a number of ways
in the experiments we designed. We have tried to make the lies relevant
to the life of the liar and to set the stakes for success or failure as high
as possible. We have attempted this for two reasons. Only in high-stakes
lies are emotions about telling the lie (fear, guilt, excitement, or what
I have called duping delight) likely to be aroused and betray the lie. It
is not just the leakage of these strong emotions that provide behavioral
clues to deceit; these emotions also disrupt the liar’s cognitive process-
ing and result in evasive, implausible, and stumbling accounts. (Friesen
and I coined [1969] the term leakage to refer to both a liar’s uninten-
tional betrayal of the truth through demeanor and deception clues—ways
a liar’s demeanor indicates that the truth is not being told but without
revealing the truth.) Leakage is a result of a high emotional and cognitive
load. A second reason for studying high-stakes lies is that these are the lies
with which society is most concerned.
In one of our experimental scenarios, we examined how well nurses
could conceal the negative emotions they felt when witnessing films show-
ing amputations and burns. They were highly motivated to succeed in this
lie, because they thought our experiment offered them the opportunity
to develop a skill they would need to use when confronting just such
upsetting scenes in their future work. In another of our scenarios, the
subjects had a chance to take and keep $50 if they could convince the
interrogator they had not taken it, whereas the subjects who did not take
the $50 could only earn $10 if the interrogator believed them when they
said they had not taken the $50. In another scenario, we first identified
the social issues the subjects felt most strongly about, and then asked each
subject to describe that opinion honestly (and earn $10 if believed) or to
claim to have the opposite of his or her true opinion (and earn $50 if
believed). In our most recent work, we have selected members of politi-
cally extreme action groups; they are given the opportunity to deprive an
opposing group of funds and to divert the same funds to their own group,
if they can fool the police officer who interviews them.
In most of our work, we give subjects the choice whether to lie or tell
the truth, as people have in real life. People may choose not to lie for
many reasons; one is their own knowledge, based on past experience, that
they are almost always caught. Including in the sample of liars such ter-
rible liars—people who would not choose to lie unless forced to do so by
the experimenter—could inflate the detection rate. In virtually all previ-
ous research, on either interpersonal deception or polygraph lie detection,
subjects were not given the choice whether to lie or be truthful (another
reason for discounting the metaanalyses by Bond and DePaulo). One
exception is the study of the polygraph by Ginton, Daie, Elaad, and Ben-
Shakhar (1982), in which the subjects were free to make a choice about
whether or not to cheat; they were not randomly assigned to cheat and
lie or not cheat and be truthful. Subjects in Stiff, Corman, Knizek, and
Snider (1994) were not told whether or not to cheat on a quiz. Bradley
(1988) also allowed subjects to choose whether to lie or tell the truth in
a polygraph study.
Another unique feature of our recent experiments is that we told the
subjects that they would be punished—and it was a considerable pun-
ishment—if the interrogator judged them to be lying. Both the truthful
person mistakenly judged to be lying and the liar who was detected would
receive the same punishment. Thus, for the first time in research on lying,
both the truthful person and the liar might be afraid—of being disbe-
lieved if telling the truth, of being caught if lying. If only the liar is afraid
of being accused of lying, the situation is too easy for the lie catcher and
not relevant to most of real life. And if neither the liar nor the truthful
person fear punishment, the situation has little relevance to the telling of
lies in the criminal justice world or in national security, let alone in marital
disputes, parent-child conflicts, and so on.
Although our recent experiments can claim to have more ecological
validity than most of the literature on either interpersonal deceit or poly-
graph lie detection, our findings about detectability were not much differ-
ent. Most of those who watched the videotapes and made their judgments
operated at a level of chance or only slightly better. Before exploring why
people do so poorly as lie catchers, I will discuss some limitations of our

research that could have led us to underestimate the ability to detect lies
from demeanor.
For the most part, the observers who judged who was lying and who
was telling the truth had no vital interest at stake in achieving accuracy.
They were not offered higher pay if they were more accurate. And catch-
ing liars was not intrinsically rewarding, because most of these people did
not make their livings catching liars. This limitation has been addressed
in our study (Ekman and O’Sullivan 1991) and in work by other research
groups (Kraut and Poe 1980; DePaulo and Pfeifer 1986) that did study
professionals concerned with catching liars. We have found that customs
officials, policemen, trial court judges, FBI, CIA, DEA, and Bureau of
Alcohol, Tobacco, and Firearms professionals, forensic psychiatrists, and
trial lawyers perform at a level not much better than chance.
Perhaps accuracy would be higher if those making the judgments had
been able to ask the questions, rather than being passive observers. I can-
not rule this out, but I doubt it would be so. The requirement to formulate
questions might well detract from the ability to process the information
provided by the person being judged. This is why, in many interrogations,
one person asks the questions while another sits passively considering the
suspect’s responses. It would be interesting to have professional inter-
rogators ask the questions in our experiments and then see if those who
watch the answerers on the videotapes are more accurate than has so far
been found.
Our observers were not familiar with those whom they judged, and
it might be argued that such familiarity would benefit accuracy. There
are, of course, many situations in which judgments about lying are made
without any prior familiarity with the person being evaluated, and our
experiments at least are relevant to those instances. But I doubt that
familiarity always benefits lie detection. While it should provide the basis
for discounting idiosyncratic behaviors, it may do so at a cost. We tend to
become invested in our friendships and work relationships, and the wish
to preserve them may lead us to develop blindness to behaviors that could
disrupt them. Trust makes one vulnerable to being misled, as usual levels
of wariness are reduced and the benefit of the doubt is routinely given.
Involvement in a relationship can lead also to confidence in one’s abil-
ity to detect deception (Sillars and Scott 1983), and such confidence may
itself make one more vulnerable (Levine and McCornack 1992). Familiar-
ity should be an unmitigated benefit only when it is with a person one has
had reason to distrust, and about whom one has acquired knowledge of
how and when they betray the relationship.
In our experiments, the observers were only shown a few minutes of
each interview before being required to make their judgment. But longer
samples may not necessarily benefit lie detection. In one study we did, the
samples shown were twice as long, and accuracy did not improve. And we
know from the behavioral measurements we have made that there are
clues to deceit in these shorter samples. Nevertheless, we cannot rule out

this limitation. If people were given much longer samples to judge, of an
hour or two, accuracy might improve.
A critic might also wonder whether accuracy was so poor because
there were few behavioral clues to deceit, but, as I have just mentioned,
that is not the case in our experiments. Measurements we have made
of facial movements, voice, and speech show that high levels of accu-
racy are possible—over 80 percent correct classifications of who is lying
and who is telling the truth. While making those measurements required
slow-motion replays, we also know that accurate judgments are possible
just by viewing the videotapes at real time. A small percent of those we
have studied have reached 80 percent or better accuracy, and they have
done so in more than one scenario, so it is unlikely that their accuracy was
a fluke (O’Sullivan and Ekman 2005). And we have found a few occupa-
tional groups that as a group were highly accurate. United States Secret
Service professionals were highly accurate on the emotion lie (a lie about
the liar’s emotions); none of them scored at or below chance, and a third
were above 80 percent accurate. Interrogators specially selected for their
known skills and given a week of training showed similar accuracy on the
opinion lie (a lie about the liar’s opinion), as did forensic psychologists.
Although in our experiments the stakes in the lies were much higher
than in other research on lying, certainly they were not as high as they are
in many criminal or national security cases. Perhaps if the stakes had been
a great deal higher, the videotapes would have contained many obvious
signs of deceit, resulting in much higher accuracy of detection. I cannot
argue against that possibility, but as I just described, some occupational
groups were accurate when judging our videotapes. The question remains
why all the other groups were not.
The information is there, and it can be detected by some but not by
most. Before considering why the overwhelming majority of people do
poorly, consider one more feature of our experiments that probably ben-
efited accuracy and may have led us to overestimate rather than under-
estimate accuracy. In all of our recent studies, we have told our observers
that between 40 and 60 percent of the people they will see will be lying.
Initially we did not give this instruction, and we found that a group of
policemen judged everyone they saw on the videotape as lying, later
explaining that everyone lies, especially to the police. Knowing the base
rate of lies is an advantage people do not always have, and should enhance
lie detection.
Granting that our evidence is not conclusive, our videotapes do contain
behavioral clues to deceit, which some people can recognize accurately
but most do not. For the purpose of this discussion, let us consider this
evidence as suggesting that in actual life, most people—the overwhelm-
ing majority—do not detect high-stake lies from demeanor. The question
I pose is why not? Why can we not all do better at this? It is not that we
do not care. Public opinion polls time and again show that honesty is
among the top five characteristics people want in a leader, friend, or lover.
And worldwide there are no end of stories, films, and songs that describe
the tragic consequences of betrayal.
My first explanation is that we are not prepared by our evolutionary
history to be either very good lie catchers or lie perpetrators. 1 I suspect
that our ancestral environment offered very few opportunities to lie and
get away with it, and the costs for being caught in a lie might have been
severe. If this speculation is correct, there would not have been any selec-
tion benefit for people who were unusually adept in catching or perpe-
trating lies. The fossil record does not tell us much about social life, so
one can only speculate about what life as hunter-gatherers might have
been like. I add to that my experience forty years ago working in what
was then a Stone Age preliterate culture in what is now called Papua
New Guinea.
There were no rooms with doors and little privacy in the small vil-
lage where I was living, in which everyone knew and saw everyone else
every day. A lie would most often be betrayed by its target or someone
else observing actions that contradicted the lie or by other physical evi-
dence. Adultery was an activity that persons often attempted to conceal
by means of lies. Such lies were uncovered not by reading the betrayer’s
demeanor when proclaiming fidelity but by accidentally catching him or
her in the act in the bush.
While lies about beliefs, emotions, and plans might be more likely to
go undetected in such an environment,2 some of those lies will eventu-
ally lead to one or another action, and then the difficulty of concealing or
falsifying actions in a setting in which there is no privacy applies.
In a society in which one’s survival depends on cooperative efforts with
other members of one’s village, the reputational loss from being caught in
a high-stakes lie might well be deadly. No one might cooperate with you if
you are known to have engaged in serious lies. You cannot change spouses,
jobs, or villages with any ease.
Cheney and Seyfarth make very similar points. An important constraint
against lying
arises from a species’ social structure. Animals that live in stable social
groups face special problems in any attempt at deceptive communica-
tion. Among socially living animals deceptive signals will probably have
to be more subtle and occur at lower frequencies if they are to go unde-
tected. Equally important, if animals live in social groups in which some
degree of cooperation is essential for survival, the need for cooperation can
reduce the rate at which unreliable signals are given. (1990, 189)
1. I am grateful to Helena Cronin at the London School of Economics for asking

me why evolution did not prepare us to be better lie catchers; also to Mark Frank at
Rutgers University and Richard Schuster at the University of Haifa for their many helpful
comments on this manuscript.
2. Helena Cronin raised this possibility.
To have some special skill in detecting (or perpetrating for that matter)
lies would not have much adaptive value in such circumstances. Seri-
ous, high-stakes lies probably did not occur that often because of limited
opportunity and high costs. When lies were suspected or uncovered, it
was probably not by judgments of demeanor. I have focused just on intra-
group lies; certainly lies between groups, and their costs and detection,
could be quite different.3
While there are altruistic lies, my discussion has dealt with less friendly
lies, lies that occur when one person gains an advantage, often at the cost
of the target. When the advantage is gained by violating a rule or expec-
tation, we call that cheating. Lies sometimes may be required to accom-
plish the cheating activity, and lies are always required to conceal having
cheated. Those cheated do not typically appreciate having been cheated
and are motivated to uncover any lies involved. But my bet is that cheat-
ing was not likely to have occurred often enough in our ancestral environ-
ment to confer an advantage on those who might have been unusually
adept at spotting when it did occur. And as I argued earlier, there was
probably so little privacy that cheats would be caught by means other
than discernment of their demeanor. The biologist Alan Grafen wrote:
The incidence of cheating must be low enough that signaling remains on
average honest. As signalers maximize their fitness, this implies that the
occasions on which cheating is advantageous must be limited. Perhaps the
signalers for whom cheating is advantageous are in a minority, or . . . only
on a minority of occasions does it pay a signaler to cheat. Cheating is
expected in evolutionarily stable signal systems, but the system can be
stable only if there is some reason why on most occasions cheating does
not pay. Cheats impose a kind of tax on the meaning of the signal. The
central fact about stable signaling systems is honesty, and the debasement
of the meaning of the signal by cheats must be limited if stability is to be
maintained. (1990, 533)
By this reasoning, signals that cheat, which I would call lies, would
have a low incidence. Cosmides and Tooby’s (1992) findings suggest that
we have evolved a sensitivity to rule infractions and do not reward cheat-
ers, and this may explain why cheating does not occur often. However,
our findings suggest that people are not likely to catch cheaters on the
basis of an ability to spot their lies from their demeanor.
To summarize my argument, our ancestral environment did not pre-
pare us to be astute lie catchers. Those who might have been most adept
in identifying a liar from demeanor would have had minimal advantage.
Serious lies probably did not occur often, because lack of privacy would
have made the chances of being caught high. Lack of privacy would also
mean that lies would typically be discovered by direct observation or
other physical evidence, making it unnecessary to rely on judgments of
3. I am grateful to Leda Cosmides and John Tooby at the University of California,

Berkeley, and to Richard Schuster at the University of Haifa for pointing this out.
demeanor. Finally, in such cooperative, closed, small societies, when lies

were uncovered, the reputational costs to the individual would have been
high and inescapable.
In modern industrial societies, the situation is nearly the reverse.
Opportunities for lying are plentiful; privacy is easy to achieve, and there
are many closed doors. When you are caught, the social consequences
need not be disastrous, for you can change jobs, change spouses, change
villages. A damaged reputation need not follow you. By this reasoning, we
live now in circumstances that encourage rather than discourage lying;
evidence and activity are more easily concealed, and the need to rely on
demeanor as an indicator of a person’s truthfulness is greater. And our
evolutionary history has not prepared us to be very sensitive to the behav-
ioral clues relevant to lying.
That granted, one might ask why we do not learn this detection skill
in the course of growing up? One possible explanation is that our parents
teach us not to identify their lies. Their privacy may often require that
they mislead their children about just what they are doing, when they are
doing it, and why they are doing it. While sexual activity is one obvious
focus of such lies, there might well be other activities that parents want
to conceal from their children.4
A third explanation is that we generally prefer not to catch liars, because
a trusting rather than a suspicious stance enriches life, despite the possible
costs. To always doubt, to make false accusations, is not only unpleasant
for the doubter, but undermines much chance of establishing intimacy in
mating, friendships, or ongoing work relationships. One cannot afford to
disbelieve one’s friend, child, or spouse when he or she is actually telling
the truth, and so one errs on the side of believing the liar. Trusting others
is not only required, but it makes life easier to live. What matter if the
cost is not detecting some who take advantage of that trust?—and one
might not even need to know about it. Only the paranoid who foregoes
such peace of mind—and those whose lives are actually at some risk if
they are not constantly alert to betrayal. Consistent with this formulation,
we obtained evidence that abused children living in an institutional set-
ting were more accurate than other children not living in institutions in
detecting lies from demeanor (Bugental et al. 2000).
My fourth explanation is that we often want to be misled; we collude
in a lie unwittingly because we have a stake in not knowing the truth. 5
Consider two examples from spousal relationships. It may not be in the
interest of a mother with a number of very young children to catch her
mate’s lie that conceals his infidelity, particularly if he is having a fling in
which he is not diverting resources that would otherwise go to her and her
4. I am grateful to Alison Gopnik at the University of California, Berkeley, for

suggesting this explanation.
5. For evidence consistent with my reasoning see Tooby and Cosmides 1989.
children. The philanderer does not want to be caught, so they both have
an interest in the lie not being uncovered. A similar logic is at work in the
following, more altruistic lie and collusive belief. A wife asks her husband
“Was there any other woman at the party whom you thought was more
attractive than me?” He lies by claiming she was the most attractive when
she was not. He does not want to make her jealous, and he does not want
to deal with her having feelings of insecuirty, and she may have a strong
desire to believe she was the most attractive.
In some kinds of collusion, the target who wants to believe the liar may
not benefit from the lie or may benefit only in the short run. Consider
what could perhaps be the most infamous example of a target believing
a liar who meant him harm. I refer to the famous meeting between the
British prime minister Neville Chamberlain and Adolph Hitler, the chan-
cellor of Germany, on September 15, 1938 (the following account is from
my [1985] 2001, 15, 16):
The world watched, aware that this might be the last hope of avoiding
another world war. (Just six months earlier, Hitler’s troops had marched
into Austria, annexing it to Germany. England and France had protested
but done nothing further.) On September 12, three days before he was to
meet with Chamberlain, Hitler demanded to have part of Czechoslovakia
annexed to Germany—inciting rioting in Czechoslovakia. Hitler had
already secretly mobilized the German army to attack Czechoslovakia, but
his army would not be ready until the end of September. If he could keep
the Czechs from mobilizing their army for a few more weeks, he would
have the advantage of a surprise attack. Stalling for time, he concealed his
war plans from Chamberlain, giving his word that peace could be pre-
served if the Czechs would meet his demands. Chamberlain was fooled;
he tried to persuade the Czechs not to mobilize their army while there
was still a chance to negotiate with Hitler. After his meeting with Hitler,
Chamberlain wrote to his sister: “in spite of the hardness and ruthlessness
I thought I saw in his face, I got the impression that here was a man who
could be relied upon when he had given his word.” Defending his poli-
cies against those who doubted Hitler’s word, Chamberlain five days later
in a speech to Parliament explained that his personal contact with Hitler
allowed him to say that Hitler “means what he says.”(Ekman, [1985] 2001,
pp. 15, 16)
Hitler reportedly wrote:“The victor will not be asked afterward whether

he told the truth. In starting and waging war it is not justice that matters
but victory” (Ekman [1985] 2001). Why did Chamberlain believe Hitler?
Not everyone did; there were many in the opposition party in Britain
and elsewhere who recognized that Hitler was not a man of his word.
Chamberlain unwittingly, I believe, colluded in Hitler’s lie because he had
a need to believe Hitler. If Chamberlain had recognized Hitler’s lie, he
would have had to confront the fact that his policy of appeasement had
put his country at grave risk. Since he had to face that fact just a few weeks
later, why did he not recognize it during this meeting with Hitler? That
question arises from a rational but not a psychological understanding of

the situation. Most of us operate on the unwritten principle of postponing
having to confront anything that is very unpleasant. And we may do so by
collusively overlooking the evidence of a liar’s untruthfulness.
Chamberlain was not unique. The targets of lies, often unwittingly,
collusively want to believe the liar. The same motive—not wanting to
recognize impending disaster—explains why the businessman who mis-
takenly hired an embezzler continues to miss the signs of the embezzle-
ment. Rationally speaking, the sooner he discovers the embezzlement the
better, but psychologically that discovery will mean he must face not only
his company’s losses but also his own mistake in having hired such a ras-
cal. In a similar fashion, it may appear that everyone but a cuckolded
husband knows what is happening. Or the preadolescent using hard drugs
may be convinced that her parents surely must know what she is doing,
while they unwittingly strive to avoid spotting the lies that would force
them to deal with the possibility that they have failed as parents and now
have a terrible struggle on their hands. One is nearly always better off in
the short run to cooperate with the lie, even if that means that the conse-
quences tomorrow will be even worse.
A fifth explanation, based on Erving Goffman’s writings (1974), is that
we are brought up to be polite in our interactions, and not to steal infor-
mation that is not given to us. A rather remarkable example of this is how
one unwittingly averts one’s gaze when the person one is talking to cleans
his or her ears or picks his or her nose. Goffman would also have argued
that the false message sometimes may be the more socially important
message than the truth. It is the acknowledged information, the informa-
tion for which the person who presents it is willing to take responsibil-
ity. When the secretary who is miserable about a fight with her husband
the previous night answers “Just fine,” when her boss asks “How are you
this morning?” that false message may be the one relevant to the boss’
interactions with her. It tells him that she is going to do her job. The true
message—that she is miserable—he may not care to know about at all, as
long as she does not intend to let it impair her job performance.
None of the explanations I have offered so far can explain why most
members of the criminal justice and intelligence communities do so
poorly in identifying liars from demeanor. Police and counterintelligence
interrogators are not taking a trusting stance with their suspects, they are
not colluding in being misled, and they are willing to steal information
not given to them. Why don’t they do better in identifying a lie from the
liar’s demeanor? I believe they are handicapped by a high base-rate and
inadequate feedback. Most of the people they deal with probably are
lying to them. Those with whom I have spoken estimate the base rate of
lying as more than three-fourths. Such a high base rate is not optimal for
learning to be alert to the subtle behavioral clues to deceit. Their orienta-
tion all too often is not how to spot the liar, but how to get the evidence
to nail the liar. And when they make a mistake and learn that someone
was wrongfully punished, that feedback comes too late, too far removed
from the mistaken judgment, to be corrective.
This suggests that if you expose people to a lower base rate of lying,
around 50 percent, and give them corrective feedback after each judg-
ment they make, they might well learn how to accurately identify lies from
demeanor. One element in this explanation of the failure to detect liars
from demeanor is that most people do not get feedback about whether
their judgments are correct or not. Not just the police lack such informa-
tion. If you were to decide that your friend had lied to you, how would you
determine if you were correct? Most often, I expect, you would simply stop
seeing the friend. The failure to definitively check your judgment against
an independent criterion of truthfulness is probably quite common and
responsible for people generally not learning from mistaken judgments.
A last factor to consider is that most people do not know what to
look or listen for in order to make accurate judgments of truthfulness.
An abundance of misinformation appears in the popular press and in the
courses offered by organizations such as Neuro-Linguistic Programming.
Even if they know what the clues to deceit are, it appears from our recent
but still unpublished research that without special training, they will miss
one of the most valuable clues: micro facial expressions, or micros.
MICRO FACIAL EXPRESSIONS
I discovered the existence of micro facial expressions (micros) when I was

examining a film I had taken of a psychiatric patient. I learned from her
doctor that she had confessed to lying in a prior session to conceal her plan
to kill herself if she was granted a one-day pass from the hospital, and I had
filmed that interview. The first time I watched the film in which she had
been lying, I saw no evidence of that; she smiled a lot, spoke optimistically,
and seemed cheerful. I would have believed her; the doctor was about to
grant her request when she confessed her lie. She had fooled him, too.
I and my colleague Wally Friesen then used an elaborate slow-motion
projection to examine each and every facial expression and gesture, frame
by frame, then in very slow motion, then a bit more rapidly, and so forth.
It took more than one hundred hours for us to go through the twelve-
minute film, but it was worth it.
In a moment’s pause before answering the doctor’s question about her
plans for the future, we saw a look of intense anguish—it was only two
frames out of twenty four, one/twelfth of a second—which was quickly
covered by a smile. We watched it again and again; there was no doubt
what it revealed. In freeze frame, her true emotion was extremely clear:
her deliberately concealed anguish! Alerted then about what to look
for, we found three other such very brief expressions of anguish in the
twelve-minute film. Once we had seen them in slow motion, we found
that we could spot them when looking at the film in real time.
We called them micro facial expressions in our first publication on non-

verbal leakage (Ekman and Friesen 1969). Later, I found that Haggard
and Isaacs (1966) had discovered micros three years before us, but their
report differed from ours in two ways: they proposed that micros are signs
of repressed, not deliberately suppressed, emotion, and they claimed that
micros are not visible at real time. We found that it was possible to see
micros without slowed motion if you knew what to look for. We did not
yet know how easy it would be to teach people to spot micros.
Putting our two discoveries together (supported by further evidence
we had obtained on both deliberate concealment and repressed emotions
[Horowitz 19XX]), it is now clear that micros can occur in cases of both
deliberate concealment and, as Haggard and Isaacs found, repression. It is
important to note that the micro looks the same in both cases. Which of
the two is in play must be determined by the context in which the micro
occurs and by further questioning.
This is a good place to explain what I mean by context, for that term
typically refers to four quite different matters, and each of them is impor-
tant in evaluating the information contained in a micro. The broadest
meaning of context is the nature of the conversational exchange; is it a first
meeting, a casual conversation, a formal interview, an interrogation in
which the other person knows that he or she is under suspicion of wrong-
doing? The very same micro might have a different significance in each of
those conversational contexts.
The second contextual issue is the history of the relationship; what has
transpired before in this conversation and in previous contacts between
the person being evaluated and the evaluator. And what does each expect
and want their future relationship to be?
A third contextual issue is what has been called speaker turn: does the
micro show up when the person being evaluated is speaking or listen-
ing? The fourth contextual issue that has to be considered in evaluating
a micro is congruence: whether the emotion shown in the micro fits or
contradicts the speaker’s simultaneous speech content, sound of voice,
gestures, and posture. If the micro is shown when the person is listening,
then the focus should be on how well it fits with what the evaluator is
saying, and with what the person being evaluated says next.
All four of these contextual matters must be considered, not only when
evaluating the emotion revealed in a micro but also when evaluating the
information shown in a, normal, macrofacial expression of emotion. All
four must also be considered when evaluating signs of emotion in the
voice, signs of emotion in the posture, and other cognitively based clues
to deceit (not discussed in this chapter).
Most people do not see micros when they occur during a conversation,
competing for attention with the speaker’s words, the sound of the speak-
er’s voice, and the speaker’s gestures. Another reason people miss micros
is that they are often distracted by thinking about what to say next rather
than watching the speaker closely. Even when we have shown people
micro expressions out of context—with the sound turned off, and no need
to think of replies—most people do not report seeing many of them.
When I first tried to teach people how to spot micros, I was surprised
to discover how quickly they learned. After a few years, I put the training
materials, which I called the Micro Expression Training Tool (METT), on
a self-instructional CD. It shows fifty-six different people; each appears
only once. Each exposure first shows the person in a neutral expression,
and then an emotion flashes briefly (for one-fifteenth of a second), and
then the expression returns to neutral. The respondent must choose one
of seven emotions (anger, fear, sadness, disgust, contempt, surprise, fear,
happiness). There is a pretest of fourteen items, a training section, and a
review section, each of which shows expressions that are commonly con-
fused (e.g. anger versus disgust), with voiceover commentary. A practice
segment provides feedback after each judgment and the opportunity to
see each practice item repeatedly. The posttest provides information on
the extent of improvement.
It takes less than forty minutes to complete the CD, half of which
consists of training. On average, people improve 30 to 40 percent from
pretest to posttest, while those who do not receive training, practice, and
review show little change from pretest to posttest. Importantly, those who
have been trained, compared to a no-training control group, also report
discerning the emotions shown in micros contained in sound videos of
spontaneous interviews.
My next METT CD, a substantial improvement over the first, includes
more practice items, training at one-thirtieth of a second, more than one
version of anger and disgust, and a wider variety of cultural backgrounds
of the people shown. A third version provides three times as much prac-
tice as the second, and is intended to bring respondents up to a very high
accuracy level in detecting micros. A fourth version provides training in
the recognition of profile views of micros. (The CDs are available online
at http://www.paulekman.com.)
The METT is now used in the training of law enforcement, national
security, and health professionals, visa interviewers, salespersons, negotia-
tors, job recruiters, and mental health professionals. Everyone who uses the
METT is enthusiastic about what he or she has learned to see; the improve-
ment is obvious to people, even without the posttest score to document it.
It is curious that this skill can be learned so quickly by nearly everyone, and
yet most people, even highly motivated people, are not able to see micros
prior to training. If it is so useful, and if people can learn it so quickly, why
do they need our training tool to learn it? The answer, I believe, is that in
the real world we do not get the necessary experience; we do not know
that these expressions exist, and even if we did, we would not get practice
in trying to spot them and getting immediate feedback about whether
our judgments are correct. On the other hand, curiously, there are people,
about 1 percent of those we have tested, who do see micros without train-
ing, and I cannot explain why they have this ability.
I do not want to give the impression that most lies are detected because
of micros or any other sign of emotional behavior. Sometimes detection
has nothing to do with the liar’s demeanor—a lie is betrayed by incon-
trovertible evidence from another source, such as a reliable eyewitness or
physical evidence. Sometimes a liar cannot resist bragging, revealing his
secret to an untrustworthy source who turns him in. The notorious spy
John Walker sold to the Soviet Union the secret of how the United States
made the propellers on its nuclear submarines silent—very important
information, because before that the Soviets couldn’t tell where our subs
were hiding, while noisy propellers revealed the location of theirs. Walker
was not caught by a polygraph or by an astute interrogator; he bragged
to his wife about how much he was being paid, heedless of the fact that
she was now his ex-wife, and he was behind in alimony payments! She
turned him in.
There are occasions when a liar does not deliberately or unintentionally
spill the beans and there is no incontrovertible contradictory evidence.
Most criminal prosecutions that go to trial (rather than being settled by
a plea bargain) are like that. The decision about truthfulness is made
by a jury, in some part at least by evaluating demeanor, trying to decide
whether witnesses are telling the truth on the basis of what they say and
how they say it. In such cases, the lie is not usually about what emotion is
being felt at the moment of the lie but is about an action—usually, though
not always, an action that has already occurred. But even on the witness
stand, emotions can become involved in lying.
Micros are only one type of clue that must be followed in evaluating
truthfulness (even if so far our research on deception has suggested they
are the most important). The accuracy achieved by using micros alone,
while statistically significant, is not high enough to have practical use.
Macroexpressions, symbolic gestures (emblems), voice, and speech con-
tent must also be considered to obtain very high accuracy in evaluating
truthfulness. My book Telling Lies (2001; 2009) describes these and other
important issues relating to why people lie, and when it is most and least
possible to evaluate truthfulness.
Basically, it is only changes in behavior that matter, changes in behavior
that occur when there is a change in the topic under discussion. If I am
always hesitant when I talk, it should be disregarded unless my hesita-
tions increase markedly or decrease markedly (suggesting rehearsal). Sec-
ond, even a marked hesitation in a response to a question to which there
should be an easy answer might be due to a totally innocent process. Even
if I am not duplicitous, I might be hesitant if questioned by my wife about
where I was when she was unable to reach me if, say, I am thinking about
why my wife seems suspicious of me or if I am trying to decide whether
I should call her on it. That is why I call all such behavioral signs hot spots,
not signs of lying; they mark a place where you need to find out more by
questioning, background checks, and so on. Alternative explanations of
why the behavior occurred have to be ruled out before concluding that a
change in behavior is evidence of lying.
Only Pinocchio had an obvious sign that would only show up when he
was lying. The rest of us at best have hot spots. Even if my face registers
a microexpression of fear when my spouse asks me about being parked
outside the St. Regis hotel in the middle of the afternoon when I usually
would be at work, it is only be a hot spot. I might be afraid my wife is not
going to believe me, or I might be worried about the state of our marriage
if she has such doubts about me. I might be concealing my fear because
I do not want her to know I am having such questions about her and our
marriage, even though I have done nothing she would consider wrong.
Of course, the fear might be showing up because am I about to deny an
actual infidelity. But since it is only a hot spot, more information is needed
to clarify what generated the emotion being concealed.
Emotions do not tell us their sources. When we see signs of a concealed
emotion in a micro, or we see a normal facial expression that contradicts
a speaker’s words, voice, or gesture, that is a hot spot. Hot spots are not
proof of lying; they merely mark places where we need to find out more
to make an accurate evaluation of truthfulness.
A final caution: not everyone who suppresses or represses an emotion
shows a micro. We have found micros in about half of the people who
are deliberately lying in our research studies. The presence of a micro
means something (an emotion exists and it is being concealed), but its
absence does not tell us whether concealment is or is not occurring. We
still do not know why only some people generate micros when they con-
ceal emotions.
More generally, we have not found any behavioral change that always
occurs in every person who is lying; that is why lie catchers must learn
to be alert to every aspect of demeanor, for it is never possible to know
ahead of time where the important information will show up. This news
always disheartens television interviewers and print media writers, who
are disappointed I cannot tell them the surefire behavioral clues to deceit.
They do not exist. Anyone who says there is an absolutely reliable sign of
lying that is always present when someone lies and never present when
someone is truthful is either misguided or a charlatan.
ACKNOWLEDGMENTS
Large portions of the discussion here on why we do not catch liars were previ-
ously published in Social Research 63, 3 (Fall 1996), 801–17; part of the discussion
of micros appeared in my essay “Emotions and Lying,” in Emotions Revealed, 2nd
ed. (New York: Holt, 2007).
I am grateful to Mardi J. Horowitz, M.D., for providing the opportunity to
examine interviews with patients who had repressed specific emotions.
References
Bok, S. 1982. Secrets. New York: Pantheon.
Bond, C. F., and B. M. DePaulo. 2006. “Accuracy of Deception Judgments.” Person-
ality and Social Psychology Review 10, 214–34.
Bradley, M. T. 1988. “Choice and the Detection of Deception.” Perceptual and
Motor Skills 66, 43–48.
Bugental, D. B., W. Shennum, M. Frank, and P. Ekman. 2000. “‘True Lies’: Chil-
dren’s Abuse History and Power Attributions as Influences on Deception
Detection.” In Attribution, Communication Behavior, and Close Relationships,
ed. V. Manusov and J. H. Harvey. Cambridge: Cambridge University Press,
248–65.
Cheney, D. L., and M. Seyfarth. 1990. How Monkeys See the World. Chicago: Uni-
versity of Chicago Press.
Cosmides, L., and J. Tooby. 1992. “Cognitive Adaptations for Social Exchange.” In
The Adapted Mind, ed. J. Barkow, L. Cosmides, and J. Tooby. New York: Oxford
University Press.
DePaulo, B. M., and R. L. Pfeifer. 1986. “On-the-job Experience and Skill at
Detecting Deception.” Journal of Applied Social Psychologv 16, 249–67.
Ekman, P., 2001.Telling Lies: Clues to Deceit in the Marketplace, Marriage and Poli-
tics (1985). 3rd ed. New York: Norton.
Ekman, P., and W. V. Friesen. 1969. “Nonverbal Leakage and Cues to Deception.”
Psychiatry 32, 1, 88–105.
Ekman, P., and W. V. Friesen. 1974. “Detecting Deception from Body or Face.”
Journal of Personality and Social Psvchology 29, 288–98.
Ekman, P., and M. N. O’Sullivan. 1991. “Who Can Catch a Liar?” American Psy-
chologist 46, 913–20.
Ekman, P., M. O’Sullivan, and M. Frank. 1990. “A Few Can Catch a Liar.” Psycho-
logical Science 10, 3.
Frank, M., and P. Ekman. 1997. “The Ability to Detect Deceit Generalizes across
Different Types of High Stake Lies.” Journal of Personality and Social Psychology
72, 1429–39.
Ginton, A., N. Daie, E. Elaad, and G. Ben-Shakhar. 1982. “A Method for Evaluat-
ing the Use of the Polygraph in a Real-life Situation.” Journal of Applied Psychol-
ogy 67, 131–37.
Goffman, E. 1974. Frame Analysis. New York: Harper and Row.
Grafen, A. 1990. “Biological Signals as Handicaps.” Journal of Theoretical Biology
144, 517–46.
Griesil, Dorothee, and J. C. Yuille. 2005. “Credibility Assessment in Eyewitness
Memory.” In Handbook of Eyewitness Psychology, vol. 1, Memory for Events, ed.
M. Toglia, R. Lindsay, D. Ross, and D. Read. Mahwah , N.J.: Erlbaum.
Haggard, Ernest A., and Kenneth S. Isaacs.1966. “Micro-momentary Facial Expres-
sions as Indicators of Ego Mechanisms in Psychotherapy.” In Methods of Research
in Psychotherapy, ed. Louis A. Gottschalk and Arthur H. Auerbach. New York:
Appleton Century Crofts.
Kraut, R. E., and D. Poe. 1980. “On the Line: The Deception Judgments of Cus-
toms Inspectors and Laymen.” Journal of Personality and Social Psychology 39,
784–98.
Levine, T. R., and, S. A. McCornack. 1992. “Linking Love and Lies: A Formal Test
of the McCornack and Parks Model of Deception Detection.” Journal of Social
and Personal Relationship 9, 143–54.
O’Sullivan, M., and P. Ekman. 2004. “The Wizards of Deception Detection.” In

O’Sullivan and Ekman, The Detection of Deception in Forensic Contexts. Cam-
bridge; Cambridge University Press.
Sillars, A. L., and M. D. Scott.1983. “Interpersonal Perception between Intimates:
An Integrative Review.” Human Communication Research 10, 153–56.
Stiff, J., S. Corman, B. Krizek, and E. Snider. 1994. “Individual Differences and
Changes in Nonverbal Behavior.” Communication Research 21, 555–81.
Toobv, J., and L. Cosmides. 1989. “The Logic of Threat.” Paper presented at the
annual meeting of the Human Behavior and Evolution Society, Evanston,
Illinois.

PART II
TRUTH, LIES, AND SELF-DECEPTION: THE
THEORY AND THE ETHICS
8
Deception and Trust
Alan Strudler
Nobody doubts that deception is sometimes wrong. But there is a peren-

nial question about when and why it is wrong. One of the most accom-
plished moral philosophers to write in the last few decades, Bernard
Williams, wrote a book addressing this question. Indeed, in the middle of
Truth and Truthfulness, he answers the question twice. 1 He argues that the
wrong in deception should be understood in terms of a deceiver breach-
ing a trust with the deceived. And he argues that it should be understood
in terms of the deceiver manipulating the deceived. Williams offers these
accounts contiguously within the space of a long book, but he does not
explain their relationship: whether there is any tension between them, or
whether instead the two accounts are consistent, complementary, perhaps
even stylistic variants on a single idea. This relationship seems puzzling in
a way worth study. I will suggest that the breach of trust and manipula-
tion accounts of deception are related directly: breach of trust is a kind of
manipulation. I will also maintain that not all manipulation in deception
involves a breach of trust, and that deception that involves a breach of
trust may involve a wrong that is distinguishable from that which occurs
in other deception. Deceiving is often best understood as a defensive act
governed by the norms of self-defense. Deceiving that involves breach of
trust and deceiving that involves other forms of manipulation are accept-
able under very different circumstances, I will conclude.
Because Williams does little to distinguish between the ideas of breach
of trust and manipulation, it will not suit my purposes to rely on him for
an account of the ideas. This presents a difficulty: there is no standard
or uncontroversial account of manipulation or trust, and no adequate
account of the ideas can be completed in this brief essay. I will therefore
develop accounts of manipulation and breach of trust to the extent neces-
sary to make my argument about deception. I will first present cases that
I hope elicit tensions in judgment about the acceptability of deception,
Note: This essay is a revised version of “Deception Unraveled,” Journal of Philosophy 102,
no. 9 (2005), 458–473.
1. Princeton: Princeton University Press, 2002, ch. 5.
139
140 Truth, Lies, and Self-Deception
and then explore how ideas of breach of trust and manipulation more
generally may help explain these tensions. I will also examine the possibil-
ity of a more extended application of my argument.
Many of the cases that I discuss will be taken from commercial life. There
are two related reasons for this. First, what makes deception often seem
acceptable are strategic factors, for example, the need to protect private
or proprietary information; these factors are comparatively accessible in
commercial contexts. Second, the leading philosophical discussions often
concern deception about bewildering problems of life and death, love and
betrayal. Yet morality matters not only in such matters of high stakes but
also in the mundane choices we face every day. Commercial life promises
cases of deception that are emotionally cool enough to limit the risk of con-
founding but rich enough to bring out the structure of deception, I suggest.
MECHANISMS FOR SECURING BELIEF
You are thinking about buying a used car from me. You live in the des-
ert, where you take long drives on lonely roads, so you want a car with
a cooling system that works well. You ask me about the condition of my
car’s radiator. I tell you that a mechanic inspected the car this morning,
and that he said that the radiator is fine. I know more, but do not say: the
radiator would leak if I had not poured Gunk in it to temporarily plug
a hole. I brought the car to this mechanic because he was incompetent,
would not notice the Gunk, and hence would provide good cover for me.
When I told you that the mechanic found no problem with the radia-
tor, did I wrong you? I did not lie. By repeating, without comment, the
words of the mechanic, I created an impression that there was no reason
to doubt what he said. I wrongly deceived you. I told a truth, but mis-
led you. What more can be said about the wrong that I committed? By
deceiving you, did I, as Williams might suggest, wrongly manipulate you?
Or breach your trust?
On the surface, the charge of wrongful manipulation seems plausible.
I take it as a sufficient even though not exhaustive characterization of
manipulation that one person manipulates another when he intentionally
causes that person to behave as he wishes through a chain of events that
has the desired effect only because the manipulated person is unaware
of that chain.2 This case, “Car Sale,” seems manipulative according to this
characterization, because in it I cause you to have a false belief in order
to get you to behave as I wish, and in doing so, I willfully hide from you
facts about the causal process, including the fact that the mechanic was
tricked. Also, I seem to breach a trust: by my behavior, I seem to vouch
2. I acknowledge that other kinds of manipulation, for example, that which operates
by preying on a person’s emotions, may be transparent to their victims. But these are not
my concern here.
Deception and Trust 141
for the mechanic’s assessment, inviting your trust and then breaching it.
There is perhaps a strong tie between breach of trust and manipulation in
“Car Sale.” After all, my success in manipulating you seems deeply bound
with my success in getting you to trust what I convey about my experi-
ence with my mechanic.
Soon I will argue that in assessing the wrongness of a particular decep-
tive act, it is important to distinguish between breach of trust and other
manipulation elements. In cases such as “Car Sale,” the distinction may
appear less than clear. But it does not take much changing of the case to
sharpen this appearance. One might do so, for example, by supposing that
I cultivate trust by solemnly explaining that for the last ten generations,
my family’s deepest commitment has been to Quaker values according to
which it is more important to be honest and forthcoming, and to act for
the sake of others, than to make money. My pronouncements—true but
misleading, because they represent my family’s past but not my present
intentions—are a solicitation of trust, and may play a distinctive causal
role in eliciting your confidence in what I say or suggest. By making these
pronouncements, I convey that you can rely on my goodwill, that is, my
intention to act for your sake and not simply for my advantage.3 I manipu-
late you, because in ways that I willfully hide from you, I cause you to
have a false belief about whether I have goodwill toward you, and then
exploit that belief in order to get you to behave as I wish. But because
I emphasize the goodwill inherent in my Quaker background, trust plays
a distinctive role in this manipulation.
Anyone who has glimpsed the recent large literature on trust will
notice that not all would accept my characterization of trust in terms
of goodwill. Sometimes, one might object, you trust another person to
be truthful not because of his goodwill, but because you know that it
is in his interest to advance your interest, and he has reason to think
that his being truthful will advance your interest. 4 For example, a repeat
buyer of fish may believe the fishmonger’s report of freshness because he
knows that it is in the fishmonger’s interest to retain him as a future cus-
tomer, and not because the fishmonger has goodwill in a stronger sense.
The fishmonger, as I imagine him, exhibits no goodwill, because he does
not act for the sake of his customer, but instead acts in order to get an
advantage for himself. No doubt the sort of reliability the fishmonger
demonstrates is socially important. At least for purposes of this essay, his
reliability is best characterized without using the idea of trust. I propose
that one may regard someone as reliably truthful because of his goodwill,
and one may be regard someone as reliably truthful because of personal
3. The idea that trusting a person requires viewing that person as possessing relevant
goodwill is ably advanced in Annette Baier, Moral Prejudices (Cambridge, Mass.: Harvard
University Press, 1995), ch. 6.
4. For representative examples of this approach, see Elinor Ostrom and James Walker,
eds., Trust and Reciprocity (New York: Sage, 2003).
advantage he gains from truthfulness; there is reason, I will suggest, for

reserving the term “trust” for the first kind of regard, but by doing so not
denying the importance of the second.
One need not demonstrate trustworthiness to secure credibility. One
can, for example, secure it externally, by providing evidence that one has
reason for being truthful even though the reason functions independently
of one’s goodwill. One can intelligibly say to another: “I know that you
don’t trust me, but you nonetheless have reason to believe what I say; if
what I say is false, I will suffer the consequences.” If a mafia don tells me
that I must be truthful to you and publicly threatens me if I am not, there
is reason to think that I will be truthful, and it has nothing to do with my
trustworthiness. Indeed, if it is known that were it not for the mafia don,
I would lie through my teeth, that need not undermine my reliability as
a truth teller, so long as the mafia don is known to remain consistent with
his threats. Thus there is good reason to distinguish between a trustwor-
thy truth teller and a reliable truth teller.
Despite the artificiality of “Mafia Don,” there are similarities between
it and common cases in which one secures credibility without relying on
trust. In “Car Sale,” for example, one bolsters one’s credibility by finding
reasons to believe that what one suggests is corroborated by an indepen-
dent, external source—a mechanic. Here, it seems that you believe what
I suggest (that there is no problem with the radiator) not because of trust,
but because you learn about a report by an external source. Perhaps you
trusted the mechanic, but he did not breach a trust with you. And you did
not trust me; you would never have believed me without the mechanic’s
report. So I manipulated you, and deceived you, but none of this involved
my breach of trust.
Both the mafia don and the auto mechanic, in different ways, enhance
one’s credibility without appeal to trust. One may find more mundane
ways in negotiation to establish credibility in the absence of trustworthi-
ness. For example, one might reveal evidence about the meagerness of
one’s assets or about the attractiveness of rival offers, hoping to show that
one has a good reason for having a particular settlement preference. The
existence of this evidence would not show that one is trustworthy, but it
might show that there is good reason to think that one is reliably truthful
about the stinginess of one’s settlement preference.
Just as one need not rely on the appearance of trustworthiness to
attain credibility for what one says or suggests, one need not rely on this
appearance in order to deceive and manipulate one’s audience. Suppose
that I create the illusion of a mafia don in order to bolster my credibility.
Because the don is my creation, I manipulate my audience in using his
story, just as I would if I had used another form of manipulation and
solicited the audience’s trust. In each case, I cause my audience to have
false belief through a process that succeeds only because I make it opaque
to that audience. In one case, the manipulative process is in gaining trust
about my truthfulness, and then breaching it; in the other, it is in causing
you to believe that there is an external basis for my truthfulness, and then
using that credibility to cause you to have a false belief. There is, thus,
substantial variation in manipulation. One can deceive a person in a way
that manipulates him but does not breach his trust.
So far, I have tried to sketch two distinguishable processes through
which deception occurs: manipulation that involves a deceiver breaching
trust and simple manipulation, which does not rely on the idea of trust
in this way. In the next section I will argue for the moral acceptability
of certain deception that does not rely on trust. I aim then to show how
adding an element of trust makes a difference for the moral assessment
of deceptive acts.
UNAVOIDABLE DECEPTION
However wrong my misleading statement seems in “Car Sale,” there are

other cases in which misleading statements seem less troubling, despite
the fact that they involve intentionally causing a false belief in the mind of
an innocent person. In this section, I will consider another example from
negotiation, arguing that certain deceptive acts that occur in negotiation
are morally acceptable because of their defensive function. To do this, I will
develop an analogy between the use of force to defend oneself against an
act of physical harm and the use of deception to defend oneself against
a query that risks economic harm. I will maintain that a straightforward
application of traditional principles of self-defense justifies the use of
deception in common negotiation.
Consider “House Sale.” You ask me whether I would pay as much as
$300,000 for your house. My voice dripping with doubt, I say, “I might
go that high; but it is an awful lot of money for me.” What I say is liter-
ally true, even though I thought that I would have to pay much more
than $300,000, and would in fact have been happy to pay $400,000. My
statement that I might go that high is misleading and deceptive, because
it falsely suggests something about my settlement preferences—that
I might not be happy, or even willing, to pay $300,000. But notice that
by misleading you in this way, I need not breach your trust. You regard
me as an adversary, someone trying to get the best of you. Hence, you
expect me to give you little more than the literal truth in our discussion.
In this respect, you can expect no more from me than our legal system
expects from a hostile witness being crossexamined by an attorney in a
trial: the witness may not lie, without committing perjury; but as long
as what the witness says is literally true, he is not responsible for false
suggestions arising from his statements. Similarly, I will maintain, in a
competitive negotiation, as long as one remains in the realm of settle-
ment preferences, one is not responsible for more than the literal truth
of what one says; one is permitted to defend oneself by making false
suggestions.
“House Sale” and “Car Sale” are not morally on par, but they involve
important similarities. In both cases, a party deceives in order to hide
information that his bargaining foe would find useful. Both cases there-
fore involve protecting oneself against possibly damaging invasive queries.
Even so, the two cases differ in a way that renders salient some limits on
the right to use self-defense. Hence I will argue for the moral acceptabil-
ity of deception about settlement preferences in “House Sale,” and then
distinguish it from the deception in “Car Sale.” My argument will draw on
traditional ideas about the norms of self-defense.
Consider a case of self-defense from outside the realm of deception.
I credibly threaten to slug you. May you use force to prevent my immi-
nent attack? It depends. One may use violence defensively only if it is
unavoidable, that is, only if one cannot otherwise, through less costly
means, prevent an attack. Call this the avoidability principle. If there is
a policeman nearby whom I do not see but you know could, through
nonviolent intimidation, secure the peace, then the avoidability principle
requires that you ask for his help rather than use force. The avoidability
principle, thus described, applies to the use of violence in self-defense.
We can easily extend it so that it applies to the use of deception in self-
defense, saying simply: “One may use violence or deception defensively
only if it is unavoidable, that is, only if one cannot otherwise, through less
costly means, prevent an attack.” In its defensive use, deception seems
often to satisfy this principle. Indeed, the idea of unavoidability is part of
the most discussed example—at least among philosophers—of acceptable
deception: a murderer asks you for the location of his intended victim,
who happens to be hiding in your cellar, and you contemplate deceiving
him. Few doubt that it is morally acceptable to deceive in “Murderer.” By
deceiving in this case, one defends an innocent person against a wrongful
threat, and uses the least objectionable means possible to do so.
“House Sale” is not “Murderer.” Nothing profound is at stake. Life goes
on if one does not deceive. Indeed, there are several ways to avoid deceiv-
ing: one might disclose one’s pricing preferences, or refuse to answer
queries aimed at eliciting those preferences, or simply walk away from
the deal. Because the deception in “House Sale” seems easy to avoid and
hence in some sense avoidable, the deception may appear not to satisfy
the conditions for legitimate self-defense, and thus appear morally dubi-
ous. These appearances, I will argue, are illusory: available paths of avoid-
ance have costs that one should not be expected to bear. There are limits
on what one must do to avoid defending oneself, even outside the realm
of deceptive defense. In the traditional jurisprudence of self-defense, these
limits may be understood in terms of the duty to retreat. Thus, if someone
threatens to attack a person on that person’s property, and the prospective
victim can safely avoid the attack by retreating into his home, then he has
no duty to retreat rather than to fend off the attack using reasonable
force; when he stands on the right to not retreat, he may use reasonable
force to protect himself. In this case, the use of force remains unavoidable,
because one has available no reasonable means to avoid it while standing

on one’s right. Deception in negotiation sometimes seems to exhibit a
similar character. Even though one might be able to avoid deception by
walking away from a deal, or taking the loss that comes from exposing
one’s reservation price, one has no duty to absorb these costs, and so one
may stand one’s ground by deceiving the other party about, for example,
one’s settlement preferences, I will maintain.
My argument about the unavoidability of defensive deception in nego-
tiation becomes easier if I make what I take to be a fairly uncontroversial
assumption: any party in a competitive negotiation has reason to learn
what his bargaining foe would regard as his worst acceptable price, that
is, his reservation price. This assumption is plausible because information
about your opponent’s reservation price is normally financially advanta-
geous in negotiation; it gives you a basis for knowing how hard to press
for concessions, and for knowing when pressing for concessions makes
no sense. Information about reservation price is particularly useful in a
case like “House Sale,” in which there is nothing but price about which to
bicker. In a competitive negotiation, every gain in price that your foe gets
is a loss for you. Thus, because your foe gains the prospect of a financial
advantage by learning your reservation price, you have reason to protect
the privacy of information about that price. Now suppose that your bar-
gaining foe directly asks about your reservation price. It is against your
interest to reveal the truth, and you have no obligation to incur the cost
that would come from doing so. What are your options? You might refuse
to answer. But then, your bargaining foe, seeking the only information
that might help him get a good price, has reason to not rest satisfied with
your refusal. He might thus prod for your price by sending trial offers
and asking you for explanations of your response to these offers, in the
hope of gaining the information to make an intelligent guess about your
position. If you try to stay clean of this process by maintaining silence in
the face of these forays, this may send its own message, for example, that
the truth is too embarrassing to speak; silence may seem like a strategic
ploy. No matter what you say or do, your opponent may use it as a basis
for making an inference about your reservation price. You have no reason
to acquiesce in his reading your behavior (and thus using your behavior)
in ways that may be unfavorable to your interests. It follows that the
prudent course in negotiation is to defend yourself against this prospect
of harmful curiosity by transmitting misleading messages about your res-
ervation price. Because of the normal structure of competitive negotia-
tions like “House Sale,” deception about reservation prices may thus be
generally unavoidable, if you want to stand your ground and protect your
bargaining position.
I have now tried to make plain how the norms of self-defense may
be used to vindicate deception that aims at fending off the prospect of
economic harm. My focus has been on answering skepticism that such
deception can satisfy a traditional requirement for the use of self-defense,
the requirement of unavoidability. If my argument thus far is correct, it

leaves open the question how much deception it licenses. This is a ques-
tion I take up in the next two sections.
LIMITS OF UNAVOIDABILITY
It may seem unsettling that, as I conclude in the previous section, one

may sometimes deceive to protect against mere economic loss. One
might worry that the standard assumed seems promiscuous. It might
seem to sanction, for example, the deception about my radiator in “Car
Sale,” because that deception may seem unavoidable if I am to protect
my price. Indeed, the standard might seem to sanction much of what the
law deems fraud. The worry is misplaced, however, because I never sug-
gested that deception is relevantly unavoidable whenever it is the most
effective means for protecting economic advantage. In fact, in the cases of
the sort I have so far considered, you may deceive only to protect propri-
etary information, information about which your audience has no right to
know. In general, you cannot plausibly claim unavoidability for defensive
action that serves the purpose of wronging a person, whether the defen-
sive action takes the form of deception or force. If I want to invade your
house to liberate a kidnap victim, and the only way you can fend off my
attack is by using force against me, then your use of force is not unavoid-
able. You have no right to deceive in order to do wrong. Undermining a
person’s choice about the fundamental nature of what he buys, merely to
gain an economic advantage over him, is wrong. So the deception in “Car
Sale” is wrong. But this may raise a new worry. If it matters, in determining
unavoidability and thus in vindicating deception, whether your audience
has a right to know, then perhaps it will suffice for all casuistic purposes
to focus simply on whether the audience has that right. What is the rel-
evance of self-defense?
I wish to show that self-defense has ineliminable relevance in assessing
deception. I will argue that the casuistry of self-defense constrains how
and when one may deceive in ways not reducible to considerations about
the audience’s right to know. To do this, we must find a case in which we
hold constant the moral situation of the deceived, but vary the situation of
the deceiver only in ways that directly alter the acceptability of deception.
Here, again, it will be useful to draw on traditional norms of self-defense,
in particular, the principle that if you are responsible for creating your own
need to use self-defense, then that limits or eliminates your right to use
self-defense. Suppose, for example, that one evening, I enter a dangerous
neighborhood, because I am lost. Hoodlums threaten me, and I run. My
pride wounded, I return the next night, carrying a gun, hoping that the
hoodlums again threaten me, so that I might shoot them in self-defense.
In this case, because I gratuitously create the need to use self-defense,
I undercut my claim to the right. I have a right to self-defense, but not a
right to use force to protect against trouble that I have sought out. More
generally, a person has no right to claim that self-defense is unavoidable
when he bears some responsibility for the attack he faces.
The relevance of this limit on the right to self-defense can be seen in
a recurring problem of deception in securities law. The problem involves
cases in which corporate officers are asked about arguably proprietary
information that would, if disclosed, adversely affect the value of their
firm’s stock. Suppose that a firm’s viability is staked to the successful
development of a particular product. And suppose that there has been
a setback in its development. You, a corporate officer, are optimistic that
the product will eventually be a success, but reasonable people might dis-
agree. If word gets out about the difficulties, the value of the firm’s stock
might be so adversely affected that the firm would be unable to develop
the product. What should you do when journalists and your shareholders
ask about progress on the product? If you disclose, the harm is clear. If you
say “No comment,” you encourage adverse inferences that will likely have
the same harmful effect because people will wonder what you are hiding.
When facing these problems, many firms find ways to say things that are
literally true but misleading; they deceive in order to protect what they
regard as proprietary information. Is this deception acceptable?
Courts disagree about how to handle this problem. Some tolerate mis-
leading statements; others do not. Judge Richard Posner takes the tolerant
approach in Eisenstadt v. Centel,5 an influential case in which the pro-
prietary information was not about product development but about an
unsuccessful auction. If word got out about the auction, it would depress
stock prices. So corporate officers faced a sensitive problem when jour-
nalists and members of the public asked about the success of the auction.
Centel’s most troubling comment, in response to these queries, was to say
that the bidding process went well.6 In fact, only a handful of parties bid,
and none came close to an acceptable price. The auction was a bust.
I find cogent the case that Centel wrongly deceived its stockhold-
ers. Posner does not. He maintains that Centel’s comment that the bid-
ding process went well was true, because “the auction process went as
smoothly as could be desired.”7 By that, he means merely that the auction
opened in an orderly way, bids were taken, assessed, and so forth. Sub-
stantively, Posner acknowledges, the auction “failed,” because it produced
very few bids and no remotely acceptable bid. Since the auction failed,
was Centel’s statement that the process went well misleading, if not false?
Posner thinks not, because he believes that it was plain that the intended
5. 113 F. 3d 738 (1997).

6. At 738. In his opinion, Posner quotes Centel officers as saying that the “bidding
process” went well. When Posner discusses what officers said, he instead speaks about the
“auction process” going well. I will follow Posner in ignoring any difference in the meaning
of these phrases.
7. At 746.
audience would not form a false belief on the basis of Centel’s statement
about the auction process. Posner’s assertion about audience response to
Centel’s statement is, of course, pure conjecture; he gives no evidence
for it. Worse, it relies on mystery. If it is as obvious as Posner maintains
that nobody would believe Centel’s statement and suggestions, then why
should corporate officers bother to make them? Why say or suggest some-
thing that you know your audience will not believe? Posner’s answer to
this question invokes an analogy with writing letters of recommendation
in an environment in which inflation, or exaggeration of praise, is the
norm. If it is the community norm to say that merely good candidates
are wonderful candidates, then if one says that one’s own merely good
candidate is (merely) good, one’s words, or what one suggests by uttering
those words, will be understood according to the community norm, and
one will send a false message about the candidate, Posner thinks.
By invoking the letter-writing analogy, Posner suggests that saying that
the auction process went well is merely a complex way of conveying
some minimal truth about the auction, perhaps that the process started
and ended smoothly. But the letter-writing analogy is inapt. In its state-
ments about bidding, Centel did more than exaggerate an evaluation. It
encouraged false belief about the occurrence of specific events. Contrary
to Posner’s benign view, there are plausible scenarios in which even a rea-
sonable person would form false beliefs on the basis of Centel’s statement
that the bidding process went well, and conclude that acceptable bids
were made. A reasonable person might not notice that Centel praised the
bidding process rather than the bids themselves. More important, even
if a reasonable person did notice the word “process,” he might still quite
plausibly believe that some acceptable bids were made, because he might
reasonably think that part of a successful process is that acceptable bids
are made. It is easy to imagine other statements about process that, like
Centel’s, are misleading. Suppose that a scientist says that the experimen-
tal process goes well; his justification is that he successfully collected and
tested all needed data; but he does not mention that the data disconfirm
his hypothesis. Or suppose that a lawyer claims that the trial process goes
well; his justification is that all witnesses show up, the proceeding begins
and ends in a timely way, and procedural formalities are respected; but
he does not mention that he is decisively losing. The scientist’s and the
lawyer’s attempts to hide behind a technical idea of pure process are ruses
aimed at hiding failure. So is Centel’s.
The best argument for Centel cannot rely on Posner’s idea that its offi-
cers said nothing false or misleading. Instead, that argument must estab-
lish something that a judge might understandably be reluctant to say:
that it was morally acceptable for Centel to make a false or misleading
statement. In trying to make that argument, the similarity between “Cen-
tel” and “House Sale” is crucial. In “House Sale,” it is morally acceptable
to make misleading statements for defensive reasons: in order to fend off
queries about matters that, if disclosed, might prove harmful, and about
which one’s audience has no right to know. Centel’s best argument must
make the case that it was similarly acceptable for Centel to make mis-
leading statements that would fend off queries about matters that, if dis-
closed, might prove harmful, and about which the audience had no right
to know. There is a problem for such an argument. “Centel” and “House
Sale” differ in a crucial respect that bears on the issue of the avoidability
of deception. Centel officers, unlike the house seller, could have ex ante
adopted a policy that would allow them to avoid the need to deceive.
Suppose that Centel announced a policy before the auction: we will hold
an auction, but we will make no comment about it until one month after
the auction ends. Then there would have been no need to make a mislead-
ing statement about the auction process because Centel’s silence would
not be subject to the adverse interpretation that the truth was too ugly
to speak. This was an option more readily available to Centel than to the
house seller because unlike Centel, the house seller is not an institution
with the capacity to make public commitments in advance of a transac-
tion. But Centel chose not to make that public commitment, knowing
that the intrigue of an auction might strengthen stock prices. By choosing
to stay silent, Centel created the conditions that made it necessary for it
to defend itself from queries about the auction. Because it was responsible
for its need to deceive, it forfeited the right to use deception defensively,
and wrongly deceived people about its auction.
In this section, I have used the example of Centel to argue that norms
of self-defense are important in assessing deception, and not reducible
to considerations such as an audience’s right to know. Some deception is
morally acceptable, I have maintained, because it provides an unavoidable
defense against invasive queries that threaten harm. I have also argued
that Centel’s deceptive statements do not satisfy the unavoidability con-
dition, and were therefore wrong. Centel’s predicament was not unique.
As I suggested, it is shared by many firms that face the prospect of disclos-
ing embarrassing news. One may, of course, disagree with me about how
a firm should respond to this predicament. If my argument so far is cor-
rect, then the best way of doing so acknowledges the relevance of norms
of self-defense, arguing that an analogy with “House Sale” is strong: that
because no feasible strategy of precommitment to silence is available, the
right of self-defense warrants making misleading statements.
DECEPTIVE WRONGS
No doubt some, taking a position I will call moral purism, would react
differently to the prospect of intrusive queries in competitive contexts
like “House Sale” and “Centel,” saying that because deception is inherently
wrong, no matter how much it might cost to forswear it, one should not
deceive. In important respects, moral purism resembles pacifism, which
finds wrong in all intentional violence. Something is true in pacifism.
Intentionally harming an unwilling human being, even in pursuit of the

social good, seems abhorrent. Despite this, pacifism seems an unreason-
ably rigid view that ultimately tolerates too much violence. Similarly,
there may be truth in purism, even though its scope is limited.
Consider another version of “House Sale,” “Quaker House Sale.” This
time, I, the buyer, employ the Quaker gambit. I invite you to trust me, sug-
gesting, but not stating, that because of my distinctive faith, I would not
deceive you about anything, including my settlement preferences. Then, by
uttering an effective half-truth, I deceive you about my settlement prefer-
ence. I cultivate your trust and breach it, just so I might gain an economic
advantage. Here my behavior, because it breaches trust, seems deeply and
unequivocally wrong. That is the truth purism sees: deception that breaches
a trust is typically a clear wrong. The problem in purism is that it overgen-
eralizes this truth, assuming that the wrong that occurs in deception by
breach of trust is matched by the magnitude and kind of wrong that occurs
when one deceives through simple manipulation. I will maintain instead
that deception that relies on breach of trust is comparatively problematic
deception. More specifically, I will maintain that deception that relies on
breach of trust is always wrong when used in pursuit of economic gain,
even though it may be acceptable when used against a substantial wrong.
As I have been considering cases in which deception might be rendered
acceptable by virtue of its role as a defensive device, it will be useful to
consider again more general ideas in the traditional ethics of defense. One
provides that any defensive measure one employs must be proportional
to the threat one faces. Hence it is wrong to defend oneself against the
threat that someone will steal one’s coffee by blowing off her head with
a bazooka, even if that is the only effective protective measure one has
available. In commercial interactions of the sort I have been considering,
I suggest, a breach of trust is also disproportionate to the threat it would
aim to thwart. A breach of trust is not an attack of the same magnitude
as a bazooka attack, but it is a more substantial invasion than that which
occurs when one deceives without breaching a trust.
How does deception by breach of trust compromise autonomy? “Quaker
House Sale” provides an example. Suppose that you are seduced by the
Quaker gambit, and so trust me on the matter of my settlement prefer-
ences. You cannot entertain skepticism about these preferences. Indeed, it
makes no sense to say that you trust what I say about my settlement pref-
erences, but that you at the same time entertain skepticism about what
I say. Trusting me about what I say regarding my settlement preferences
means nothing if not that you accept what I say as true, without skepticism.
Because you accept what I say as true, you have transferred the effective
locus of your decision-making on your beliefs about my settlement prefer-
ences to me. Your autonomy with respect to this matter is in my hands.
Thus if I deceive you and betray your trust, I compromise your autonomy.
There is a natural objection to the foregoing line of argument. One may
complain that autonomy cannot be compromised so easily. If you
choose to accept what I say about my settlement preferences in “Quaker

House Sale,” and what I say turns out to be false, that has no bearing
on the autonomy of your choice. Instead, by believing me, you autono-
mously took a chance, and lost your gamble, just as you might in coming
to believe any other uncertain proposition, one might argue. When you
believe me, you do not get the truth, hence do not get what you want.
But plainly, simply failing to give a person what he wants is not thwarting
his autonomy. Deceiving a person who trusts you about your settlement
preferences, one might thus think, may cause him to make a mistake, but
does not undermine his autonomy.
This objection misunderstands the constraints imposed by respect for
autonomy. It supposes that if you autonomously choose to hand over
decision-making authority to a person, as you do when you genuinely
trust him, and that person then exercises the authority, your autonomy
is thereby not abused or otherwise compromised. But the mere fact that
you autonomously choose to hand over your decision-making authority
does not show that your autonomy is respected when that authority is
exercised. Showing respect would require that the authority be exercised
in a way that is consistent with your reasons for handing it over. Still, the
objection may seem hard to assess, because it is hard to say in general
terms what constitutes compromising autonomy and what constitutes
respecting it. Perhaps it will be easier to start from the particular, a clear
case of undermining autonomy after authority has been handed over.
Consider this case. Hannah expects to be unconscious when she soon
undergoes surgery. It is possible that difficult life-and-death decisions will
have to be made at that time, but the options cannot be enumerated in
advance. She has discussed her values about such matters with Anna, and so
she writes a note designating Anna as a surrogate decision-maker. The surgery
does not go well, Anna is consulted, and knowingly chooses against Hanna’s
values. It seems clear that Anna compromises Hanna’s autonomy. One might
argue: because Hanna was able to exercise her choice that Anna acts as her
surrogate, her autonomy was respected. But that seems implausible because
Anna was entrusted to act consistently with Hanna’s values and she did not.
Anna breached a trust, acting in a way that is presumptively wrong, and that
might be excusable only in the most extraordinary circumstances. For morally
similar reasons, I suggest, deception through breach of trust seems presump-
tively wrong, nothing that can be justified or excused by the mere prospect
of the pecuniary gain that drives typical negotiation. Breaching trust is a grave
act, an assault on autonomy, something we might entertain in response to a
grave wrong, such as the threat of murder, not something we should enter-
tain merely as a way to gain advantage in a competitive negotiation.
If I am correct about the wrongness of deception based on a breach
of trust, it does not follow that all deception is wrong. Not all deception
involves a breach of trust. Recall “House Sale.”
When I complain that $300,000 would be a burden for me to pay, you
remain skeptical. You know that I am trying to make a plausible impression
that my reservation price is low. So you take my comment as one piece of

evidence, to be weighed with other pieces of evidence about my financial
position, my other buying opportunities, my apparent bargaining strategy. In
“House Sale,” unlike “Quaker House Sale,” you do not trust me. You retain
the full independence of your judgment, and do not give me sufficient con-
trol over matters that I might jeopardize your autonomous choice. Because
your autonomy need not be at risk and because I have a genuine need to
protect my position, I may deceive you without wronging you, it seems.
The position that I defend may have implications for an ancient problem
in ethics. It may help explain the moral relevance of the distinction between
two kinds of deceptive acts: (1) lies, that is, deception that occurs through
acts of false assertion, and (2) deception that occurs through making merely
misleading statements. As I have argued, the wrong in a speaker’s decep-
tive speech act must be understood in terms of the mechanism for secur-
ing belief that the speaker employs, whether that mechanism is trust or
something else. If my argument is correct, then we should expect to find a
moral difference between lying and mere misleading if these deceptive acts
connect to mechanisms for securing belief in different ways. In future work,
I will explore the hypothesis that in many cases lying involves a greater
wrong than misleading because lying relies on a deeper kind of trust than
does misleading, and that in other cases, lying is wrong when misleading is
not because of the different ways these deceptive acts connect to trust.
I do not expect everybody to agree with me about the possibility of
acceptable deception in cases like “House Sale.” I am instead concerned
with how trust matters in assessing deception, given that deception can
occur even in the absence of trust. My suggestion on this score is sim-
ple. It is always morally unacceptable to deceive a person in a way that
breaches his trust, unless that deception is necessary to defend against
a grave wrong. But it may be morally acceptable to defend a person in
the absence of trust if that deception is necessary to defend against an
action that may thwart one’s legitimate interests. One might reject my
suggestion and still see an important moral difference between deception
involving a breach of trust and simple deception. One might thus say that
other things being equal, deception involving a breach of trust is morally
worse than deception by other manipulative means. No doubt deception
by breach of trust is commonly worse than deception by more simple
manipulation. But that contrast does not track the complexity of the dif-
ference between these two kinds of deception. Indeed, there are cases
in which both kinds of deception would be morally acceptable—neither
being wrong, hence neither being worse than the other; for example, the
case in which one seeks to fend off the murderer at the door making que-
ries about the whereabouts of his prospective victim. The morally impor-
tant difference between deception that occurs through breach of trust
and deception that occurs through more simple manipulation, I have
argued, is that there are different rationales appropriate for employing
each kind of deception.
Lying, Deception, and Related Concepts
Thomas L. Carson
Conceptual questions about the nature of lying and deception are prior
to questions about the moral status of lying and deception. Any theory
about the moral status of lying (or deception) presupposes an account of
what lying (or deception) is.
This essay offers definitions of the concepts of lying and deception
and explains the distinctions between lying, deception, withholding infor-
mation, “keeping someone in the dark,” bullshit, spin, and half-truths.
In section 1, I propose a definition of lying. Roughly, a lie is a deliber-
ate false statement that the speaker warrants to be true. Three features
of my definition are noteworthy. First, I argue that standard dictionary
definitions overlook a necessary condition of lying, namely, that the liar
cannot believe that the statement she makes is true. Second, contrary to
most standard definitions, I argue that lying does not require that the liar
intends to deceive others. (I appeal to cases in which one is compelled or
enticed to make false statements, cases of lying in which one can benefit
by making false statements, even if they don’t deceive others, and cases
of bald-faced lies in which the liar knows that others know she is lying
and therefore has no hope or intention of deceiving them.) Third, I hold
that in order to tell a lie, one must make a statement that one warrants to
be true. According to my definition, any lie violates an implicit promise
or guarantee that what one says is true. My definition makes sense of the
common view that lying involves a breach of trust. To lie, on my view, is
to invite others to trust and rely on what one says by warranting its truth
and, at the same time, to betray that trust by making a false statement that
one does not believe to be true.
In section 2, I propose a definition of deception and discuss the rela-
tionship between deception, withholding information, and keeping some-
one in the dark. Roughly, deception is intentionally causing someone to
have false beliefs. This definition is close to the mark. However, it needs
to be qualified to deal with cases such as the following. I intentionally
cause you to believe statement X and X is false, but I neither believe
that X is true nor believe that X is false. (In this case I do not cause you
to believe a statement that I believe to be false.) There are two main
153
differences between lying and deception. First, unlike “lying,” “deception”

implies success. An act must actually cause someone to have false beliefs
in order to count as a case of deception. Intentional false statements need
not succeed in deceiving others in order to count as lies. A statement
that is not believed by others and does not mislead anyone could still be
a lie. Second, although a lie must be a false statement, deception needn’t
involve making a false statement; true statements can be deceptive, and
many forms of deception do not involve making statements of any sort.
Thus, many instances of deception are not cases of lying.
In section 3, I discuss the relationship between lying, deception, and
bullshit. Harry Frankfurt claims that bullshit requires the intention to
deceive and that bullshit is not lying (bullshit is a kind of misrepresenta-
tion “short of lying”). I argue that both claims are mistaken—bullshit need
not be intended to deceive others, and one can produce bullshit while
lying. I also argue that Frankfurt’s famous claim that bullshitters, unlike
liars, are unconcerned with the truth is mistaken; there are clear cases of
bullshit in which bullshitters are very concerned to say what’s true. Sec-
tion 3 also offers a brief account of the concepts of “spin” and “half-truths”
and the relationship between these concepts and lying and deception.
1 THE DEFINITION OF LYING
1.1 Lies and Falsehoods

In order to tell a lie, one must make a false statement. Showing that a
statement is true is always sufficient to counter the accusation that one
has told a lie. People often attempt to mislead others without saying any-
thing that is false. Often, people who are ordinarily very careless about
how they say things choose their words with great care in order to mislead
others without saying anything that is literally false. For example, sup-
pose that I know that a used car I am selling frequently overheats. You
are a prospective buyer and ask me whether the car overheats. If I say
“No,” I am presumably lying. If I answer by making the true statement
“I drove this car across the Mojave Desert on a very hot day and had no
problems,” I am not lying. Even though this statement is true, it might still
be very misleading—perhaps I drove the car across the desert four years
ago and have had lots of trouble with it overheating since then. Often,
people engage in this kind of verbal trickery because they take them-
selves to be avoiding lying. It is difficult to account for this widespread
phenomenon unless we concede that ordinary language does not count
such statements as lies. The recent controversy about whether or not
President Clinton lied to the Starr Grand Jury when he denied having a
“sexual relationship” with Monica Lewinsky illustrates this point. Clinton
claimed that his statement that he “did not have sexual relations with”
Monica Lewinsky was not a lie because it was true, given the definition
Lying, Deception, and Related Concepts 155
of “sexual relationship” specified in the questions he was asked by the

grand jury. Clinton went through verbal contortions attempting to avoid
acknowledging his relationship with Lewinsky without saying anything
false. Clinton claimed that he didn’t say anything false and, thus, didn’t lie
or perjure himself. Those who hold that Clinton’s statement was a lie are
claiming (among other things) that what he said was false. (Perjury laws
presuppose that a lie must be a false statement. Perjury is defined as lying
under oath in court. The law counts someone’s testimony as perjurious
only if what she says is false. Therefore, the law presupposes that a lie
must be a false statement.)
A more difficult case for my view that a lie must be a false statement
is the following:
I go fishing on a boat with a friend, John. He and I both catch a fish at the
same time. Although we don’t realize it, our lines are crossed. I have caught
a very big fish and John has caught a little one, but we mistakenly believe
that I caught the small fish and John caught the big one. We throw the two
fish back into the water. I go home thinking that I caught a small fish. When
I return, my father, an avid fisherman, asks me how I did. I say that I caught
a very large fish and threw it back into the water, thereby intending to
deceive him about the size of the fish that I caught.1
My linguistic intuitions tell me that a lie must be a false statement, and

that, therefore, what I say in this case is not a lie. I intend to lie in this
case, but I don’t. Others report conflicting intuitions about this case and
the question of whether a lie must be a false statement. (Fried, Isenberg,
Williams, Bok, and Chisholm and Feehan all defend definitions of lying
according to which a lie needn’t be a false statement.) To the extent that
it rests on disputed intuitions, my claim that a lie must be a false state-
ment is open to question. I won’t pursue this issue further here. (In note
28, I explain how my preferred definition of lying can be modified to
accommodate the view that a lie needn’t be a false statement.)
1.2 Standard Dictionary Definitions of Lying

All lies are false statements, but not all false statements are lies. I do not
lie if faulty memory causes me to state something false when I am trying
my best to be accurate and truthful. Sometimes people say things that are
false in order to make a joke. If I say something that is clearly false as a
joke that is not intended to be taken seriously, I am not lying.
What is the difference between a lie and (the broader notion of ) a
false statement? Standard dictionary definitions of lying say that a lie is
a false statement made with the intent to deceive others. The first defini-
tion of the word “lie” in the Oxford English Dictionary is the following:
1. I owe this example to Bruce Russell.

“a false statement made with the intent to deceive.”2 Webster’s Inter-

national Dictionary of the English Language (1929) gives the following
definition: “to utter a falsehood with the intent to deceive.” These two
definitions overlook an essential feature of lying. If a statement is a lie,
then the person who makes it cannot believe that it is true. Showing that
one believed what one said is always sufficient to rebut the claim that one
told a lie.3 The fact that I intend to deceive you by means of stating x does
not necessarily imply that I believe that x is false. I might try to deceive
you by saying something that I believe is true. For example, suppose that
I tell you that Joe is away from his home because I hope to deceive you
into thinking that you can easily break into his house and steal his paint-
ings. In fact, I know that he has a sophisticated burglar alarm with a video
camera that is likely to catch you in the act. I believe that Joe is away from
home, but contrary to what I believe, he is at home. My statement is false
and it is intended to deceive you, but it is not a lie, because I believe that
it is true.
The definition of lying needs a condition to rule out the possibil-
ity that one believes that what one says is true. However, it is unclear
exactly how this condition should be formulated. We might say that in
order to lie, one must make a false statement that one believes is false
(or believes is probably false). Alternatively, we might say that in order
to tell a lie one must make a false statement that one doesn’t believe to
be true. These two different ways of formulating the condition yield dif-
ferent results in the following sort of case. I make a false statement when
I don’t have the slightest idea whether or not it is true. Such statements
are characteristic of bullshit. According to the strong condition (given
earlier), this statement cannot possibly be a lie (no matter what other
conditions it satisfies) because I don’t believe that the statement is false
or probably false. The weaker condition allows for the possibility that
statements of this sort are lies. (In such cases, the person who makes the
statement does not believe that it is true.) I don’t know of any decisive
reason for preferring the stronger condition to the weaker condition
or vice versa. Rather than attempt to show that the correct definition
of lying must incorporate the stronger condition (or the weaker condi-
tion), I think we should simply say that there are broader and narrower
concepts of lying.
1.3 A Reformulation of the Dictionary Definitions

The foregoing objection to the dictionary definitions of lying and the
uncertainty about how to formulate the condition that the liar must
2. Oxford English Dictionary, 2nd ed. (Oxford: Clarendon Press, 1989).

3. D. S. Manison also criticizes standard dictionary definitions on these grounds; “Lying
and Lies,” Australasian Journal of Philosophy 47 (1969), 134.
believe that what he says is false (or not believe that it is true) suggest the
following definition of lying:
L1. A person S tells a lie iff: 1. S makes a false statement x, 2. S believes
that x is false or probably false (or, alternatively, S doesn’t believe that x is
true), and 3. S intends to deceive another person by means of stating x (by
stating x, S intends to cause another person to have false beliefs).4
The first condition of my definition needs explanation and clarifica-

tion. What does it mean to say that a person, S, makes a false state-
ment? This means roughly that (1) S produces (utters, writes, signs,
etc.) a linguistic token, t, that expresses a proposition, X, (2) X is
false, and (3)) S does this with the intention of communicating X to
someone or some group of people. The last clause is clearly necessary.
I might idly or absentmindedly produce linguistic tokens that express
propositions with no intention of communicating those propositions
to anyone. For example, while being bored by a committee meeting,
I might write a sentence expressing a proposition on a piece of paper.
Or, in an empty room, I might read aloud statements from a book
I am looking at. Doing this does not constitute stating anything unless
I somehow intend to communicate to others. (More on this later in
section 1.13.)
Consider the following revised version of L1:
L1'. A person S lies or tells a lie iff: 1. S makes a false statement x, 2. S
believes that x is false or probably false (or S doesn’t believe that x is
true), and 3. by stating x, S intends to cause others to believe x.
Definitions L1 and L1' differ only in their formulation of condition 3.

They differ in cases in which a person intends to deceive others by means
of stating something false (x) but does not intend to cause them to
believe x.
4. Linda Coleman and Paul Kay argue that these three conditions give us an account of
central or paradigm cases of lying. According to Coleman and Kay, anything that satisfies
all three conditions is clearly a lie. However, when a statement satisfies only two of the
three conditions, it is not clear whether or not it counts as a case of lying. Linda Coleman
and Paul Kay, “Prototype Semantics: The English Verb ‘Lie.’ ” Language 57 (1981), 26–44.
Arnold Isenberg defines a lie as a statement that one believes to be false and is intended
to deceive others, “Deontology and the Ethics of Lying,” Philosophy and Phenomenological
Research 24 (1964), 465–80. Isenberg’s definition is equivalent to L1 without the
stipulation of condition 1 that statement x is false. Bernard Williams proposes a definition
of lying very similar to Isenberg’s. Williams’s definition is as follows: “I take a lie to be an
assertion, the content of which the speaker believes to be false, which is made with the
intention to deceive the hearer with regard to its content”; Truth and Truthfulness: An Essay
in Genealogy (Princeton: Princeton University Press, 2002), 98. Unlike Isenberg, Williams
stipulates that the liar must intend to deceive others about the content of the statement
in question. Isenberg and Williams’s definitions are very similar to the following definition
that Sissela Bok attributes to Augustine: “having one thing in one’s heart and uttering
another with the intention to deceive”; Lying (New York: Pantheon Book, 1978), 33. Bok
herself defines a lie as “any intentionally deceptive message which is stated,” 13.
1.4 Lying and the Right to Know the Truth

Those who are attracted to the absolutist view that lying is always wrong
sometimes try to deal with objections by appealing to very narrow defi-
nitions of lying according to which the examples posed as cases of per-
missible lying are not genuine cases of lying.5 The best-known version
of this kind of definition is the view that a necessary condition of one’s
lying is that the person to whom one’s statement is directed has a right to
know the truth, so that speaking falsely to someone who has no right
to know the truth cannot be a lie. According to this definition, my making
a deliberate false statement to a thief who asks me where I hid my money
would not be a lie because the thief does not have a right to know the
truth about the matter in question.6 This kind of constraint can be incor-
porated into the definition of lying as follows:
that x is false or probably false (or, alternatively, S doesn’t believe that x
is true), 3. S intends to deceive another person by means of stating x (S
intends his statement to cause another person to have false beliefs), and
4. the person to whom he makes the statement has the right to know the
truth about the matter in question.
L2 is the result of adding condition 4 to L1. Adding something like con-

dition 4 to the definition of lying greatly narrows the concept of lying
and removes some of the most serious objections to the absolutist view
that lying is always wrong. Given such definitions, some of the strongest
examples commonly adduced against the claim that lying is always wrong
are not genuine instances of lying. Consider the following variation on
Kant’s infamous example from his essay “On a Supposed Right to Lie
from Philanthropic Concerns.” Suppose that an acquaintance of yours
comes to your door and asks you the whereabouts of a personal enemy.
He evidently wants to murder the other person. You deliberately say that
she (the man’s enemy) is at a certain location when you know that she
is not. You do this in the hope of misleading the man and saving her life.
I would call this a case of permissible lying. According to L2, this is not
a case of lying. I have no substantive moral disagreement with defend-
ers of L2 about this case. We both agree that it is morally permissible to
speak falsely to the man and mislead him. Our only disagreement is about
whether we should count this example as a case of lying.
The sort of definition under consideration is sharply at variance with
ordinary language. Ordinary language counts the example in question as a
case of lying. There is a strong presumption against any definition of lying
5. See Bok, Lying, 14–15.

6. Grotius holds that in order to count as a “falsehood” (lie), a statement must conflict
with the rights of the person(s) to whom it is addressed (the person’s right to know the
truth). On the Law of War and Peace, trans. F. W. Kelsey (Indianapolis: Bobbs Merrill,
1925), bk. 3, ch. 1. (This passage is reprinted in the appendix to Bok, Lying, 263–67.)
so much at odds with ordinary language. Using the term “lying” in accor-
dance with this definition is likely to engender confusion. Defenders of
L2 face a very strong burden of proof. Pragmatic considerations also seem
to weigh against this definition of lying. Definition L2 makes it impossible
for us to determine whether or not certain acts are lies until we have first
resolved difficult and controversial moral questions (about whether or
not someone has a right to know the truth). 7 It is a matter of consider-
able controversy exactly when other people do and do not have a right to
know the truth about particular matters. If we accept L2 (or any similar
definition), then we can’t call a statement a lie unless we have reason to
think that the audience has a right to know the truth. There are good
pragmatic reasons for us to use the concept of lying to help point out
and distinguish between salient features of actions and thereby assist us
in making moral judgments. In order to serve this purpose, the concept of
lying must be defined independently of controversial moral assumptions.
Some people endorse definitions such as L2 in order to defend the
absolutist view that lying is always wrong. (I don’t know of any other rea-
sons to accept L2). However, L2 does not rule out ostensibly clear cases
of morally permissible lying. Consider the following example. A man has
just had open heart surgery and is temporarily in a precarious state of
health. His surgeon says that he must be shielded from any emotional
distress for the next few days. Unbeknownst to the patient, his only child,
Bob, has been killed in an automobile accident. When the patient awakens
after the surgery, he is surprised that Bob isn’t there and he asks, “Where
is Bob?” You fear that in his condition, the shock of learning about Bob’s
death might cause the man to die. So you lie and say that his son has been
delayed, all the while firmly intending to tell him the truth and apologize
for lying when he is out of danger. Given appropriate qualifications, it is
morally permissible to lie to the patient. Definition L2 doesn’t exclude
this as a case of lying—surely the father has a right to know the truth
about his son’s death. This seems to be a case of morally permissible lying
that violates someone’s right to know the truth. Not every case of making
a false statement to save the life of an innocent person is a case of making a
false statement to someone who has no right to know the truth.
1.5 That the Intent to Deceive Is Not Necessary for Lying

Definition L1 seems to me to be fairly close to the mark. However, some
clear cases of lying in which one is compelled or enticed to make false
statements (and some cases of lying in which one can benefit by making
false statements) do not involve any intention to deceive others. These
cases are counterexamples to L1 and to most standard definitions of lying
7. See Bok, Lying, 14–15. Bernard Williams also argues against attempts to construe
the concept of lying very narrowly in order to make plausible the view that lying is always
wrong no matter what the consequences; see Truth and Truthfulness, 105.
(such as the dictionary definitions noted earlier) according to which the

intent to deceive others is necessary for lying. Suppose that I witness a
crime and clearly see that a particular individual committed the crime.
Later, the same person is accused of the crime and, as a witness in court,
I am asked whether or not I saw the defendant commit the crime. I make
the false statement that I did not see the defendant commit the crime, for
fear of being harmed or killed by him. It does not follow that I intend that
my false statements deceive anyone. (I might hope that no one believes
my testimony and that he is convicted in spite of it.) Deceiving the jury
is not a means to preserving my life. Giving false testimony is necessary to
save my life, but deceiving others is not; the deception is merely an unin-
tended “side effect.” I do not intend to deceive the jury in this case, but it
seems clear that my false testimony would be a lie.
Here, it might be objected that even though I don’t intend to deceive
the members of the jury about the defendant’s guilt, I still intend to
deceive them into falsely believing something else, namely, that I believe
that what I am saying is true. However, if we modify the example, we can
generate a clear case of lying in which the liar doesn’t intend to deceive
the members of the court about anything. Suppose that I know that the
crime and my presence at the scene of the crime were recorded on a video
camera so that there is almost no chance that the jury will believe that
I believe what I am saying. Further, suppose that (1) I am confident I will
not be charged with perjury, even if everyone believes that I am lying, and
(2) I am indifferent to other people’s opinion about my character. Given
all of this, I do not intend to deceive anyone into thinking that I believe
what I am saying.
In the example in which I am the witness, I lie by making a false
statement, even though I have no intention of deceiving anyone about
anything. This case demonstrates that, at least sometimes, there is a per-
formative act involved in making a statement. Sometimes we want to “go
on record” and claim the truth of a particular statement. When we go on
record to claim something, we warrant the truth of what we say. It is pos-
sible for me to go on record to claim the truth of something that I know
is false without intending to deceive anyone. In my witness example, I go
on record to claim that I did not see the defendant commit the crime, in
order to avoid being killed or harmed. (In a case of bribery, I might lie by
testifying and saying something false “on the record” without having any
intention to deceive others; I might simply want to receive my reward for
giving the testimony and hope that it doesn’t deceive anyone.)
We can also imagine cases in which someone “goes on record” and war-
rants the truth of something he knows to be false in order to avoid insti-
tutional punishment of one sort or another. Suppose that a college dean
is cowed whenever he fears that someone might threaten a lawsuit and
has a firm, but unofficial, policy of never upholding a professor’s charge
that a student cheated on an exam unless the student confesses in writing
to having cheated. The dean is very cynical about this and believes that
students are guilty whenever they are charged. A student is caught in the
act of cheating on an exam by copying from a crib sheet. The professor
fails the student for the course, and the student appeals the professor’s
decision to the dean, who has the ultimate authority to assign the grade.
The student is privy to information about the dean’s de facto policy and,
when called before the dean, he (the student) affirms that he didn’t cheat
on the exam. He claims that he was not copying from the crib sheet. He
claims that he inadvertently forgot to put his “review sheet” away when the
exam began and that he never looked at it during the exam. The student
says this on the record in an official proceeding and thereby warrants the
truth of statements he knows to be false. He intends to avoid punishment
by doing this. He may have no intention of deceiving the dean that he
didn’t cheat. (If he is really hard-boiled, he may take pleasure in thinking
that the dean knows that he is guilty.) An objector might continue and say
that surely the student intends to deceive someone—his parents or future
employers. However, this isn’t necessarily the case. The student may not
care whether or not others know that he cheated (he might freely and
cynically tell others about his cheating) but simply want to have his grade
changed. (If it helps, suppose that the will of a deceased relative calls
for the student to inherit a great deal of money if he graduates from the
college in question with a certain grade-point average.) These two cases
of lying without intending to deceive others are crucial for my argument
in this article. Later (section 1.10), I will return to these two cases and
explain why my definition counts them both as cases of lying.
Let me digress briefly to explain the notion of an intentional act.
Roughly, the intended consequences of an act are those that one either
(1) aims at for their own sake, or (2) foresees and regards as part of a
causal chain leading to consequences that one desires or aims at for their
own sake.8 Consider the following example. The leader of nation X plans
an air attack on armaments factories in nation Y. The ultimate aim of the
attack is to end the death and suffering caused by the war. The leader of X
regards the destruction of these factories as a necessary means to ending
the war and the attendant suffering. The leader foresees that the attack
will kill civilians living in areas adjacent to the factories. 9 The death of
these people is not a means that (causally) contributes to the goal of end-
ing the war. Rather, it is an unavoidable “side effect” of the bombing. This
consequence is not “intended” but merely foreseen. “Terror bombing,” on
8. See Alvin Goldman, A Theory of Human Action (Englewood Cliffs, N. J.: Prentice-
Hall, 1970), ch. 3; and Michael Walzer, Just and Unjust Wars (New York: Harper and Row,
1977), 153–56. Traditional “just war theory” makes much of the distinction between what
is intended and what is merely foreseen. This distinction also has a very important place in
scholastic ethics and is central to the “doctrine of the double effect.”
9. Franklin Roosevelt and American military leaders were acting in exactly this way
when they ordered massive precision daylight bombing attacks on German armaments
factories and other military targets during World War II.
the other hand, involves the intention to kill innocent civilians. Suppose
that the leader of a nation at war orders the bombing of an enemy city.
She aims to end the war by demoralizing the civilian population of the
enemy country.10 Demoralizing the civilian population is to be accom-
plished by killing large numbers of civilians. Here, the killing of the civil-
ians is not just a side effect but an essential part of the causal chain leading
to the ultimate goal of the bombing. Leaders who order the bombing of
cities for these reasons intend to kill civilians. (This holds even if the lead-
ers in question don’t desire the killing of the civilians as an end in itself
and even if they deeply regret the killing of the civilians.)
1.6 Chisholm and Feehan’s Definition

In their well-known article “The Intent to Deceive,” Chisholm and Feehan
define lying as follows:
[Person] L lies to [person] D = df There is a proposition p such that (i)
either L believes that p is not true or L believes that p is false and L asserts
p to D.11
This definition makes use of the concept of asserting a proposition.

Chisholm and Feehan define this notion as follows:
L asserts p to D = df L states p to D and does so under conditions which,
he believes, justify D in believing that he, L, not only accepts p, but also
intends to contribute causally to D’s believing that he, L, accepts p. 12
They comment on this definition as follows:

A statement that is made merely in play, or in irony, is thus not an asser-
tion, for the speaker is not justified in taking it seriously. When one speaks
10. The British night-time attacks on German cities and the American firebombing
and nuclear attacks on Japanese cities at the end of World War II aimed at killing innocent
civilians as a means to demoralizing the enemy and thereby diminishing their will to
continue fighting. In this sense, it is arguable that Churchill, Roosevelt, and Truman were
all “terrorists.”
11. Thomas Feehan and Roderick Chisholm, “The Intent to Deceive,” Journal of
Philosophy 74 (March 1977), 152. Chisholm and Feehan note that their definition
is “essentially the same as that proposed by Frege” (152). They quote the following
definition of lying from Frege: “In ‘A lied in saying that he had seen B’, the subordinate
clause designates a thought which is said (1) to have been asserted by A (2) while A was
convinced of its falsity.” “Über Sinn und Bedeutung,” in Translations from the Philosophical
Writings of Gottlob Frege, ed. Peter Geach and Max Black (Oxford: Blackwell, 1952), 66.
Charles Fried defends a very similar definition (he acknowledges his debt to Chisholm and
Feehan): “A person lies when he asserts a proposition which he believes to be false”; Right
and Wrong, (Cambridge, Mass.: Harvard University Press, 1978), 55; also see 59. He claims
that in order to assert a proposition one must intend that it be believed (56–57). His
definition implies that the intent to deceive others is a necessary condition of lying and is,
therefore, open to the sort of objection presented above.
12. “Intent to Deceive,” 152.
ironically, one plays oneself down, but one gives a “signal of irony”—perhaps
by means of tone and choice of words—and this signal indicates that one is
not to be taken seriously and hence one is not making an assertion.13
The example in which I am the witness is also a counterexample to

Chisholm and Feehan’s definition. The witness clearly lies in this example,
but their definition doesn’t count this statement as a lie. Since the defen-
dant’s crime was videotaped, the witness knows that the jury won’t be
justified in believing that he believes (accepts) what he says. This objection
can be strengthened if we further stipulate that the witness is known to be
an extremely dishonest person. Given all of this, the witness knows that
the jury will not be justified in believing that he believes (accepts) what
he says. Chisholm and Feehan’s definition has the very odd and unaccept-
able result that a notoriously dishonest person cannot lie to people who
he knows distrust him. Their definition implies that it is self-contradictory
to say that I lie when I know that others know that I am lying (and thus are
not justified in believing that I believe (accept) what I say).14
1.7 My Definition of Lying (A Preliminary Version)

Certain features of my definition are hinted at by W. D. Ross. Ross holds
that all lies are prima facie wrong because they are instances of promise-
breaking; it is prima facie wrong to lie because to lie is to break an implicit
promise to tell the truth that one makes whenever one uses language to
make statements.15 Ross himself does not define lying. However, his view
implies that all lies involve breaking a promise to speak (or communicate)
truthfully. I agree with Ross that lying involves breaking a promise (or
something very similar to a promise) to communicate truthfully.16 The

14. Chisholm and Feehan’s definition implies that one can lie without making a false
statement. I argue against this in section 2. However, since my argument there appeals to
disputed intuitions, I don’t rest my criticisms of Chisholm and Feehan on this point.
15. Ross claims that all moral duties can be reduced to six basic types. The first class of
duties on his list is duties derived from one’s own past actions. He makes a further division
within this class. First, there are duties to make reparations for one’s own previous wrongful
acts. The other kinds of duties that rest on one’s own prior actions are “those resting on a
promise or what may fairly be called an implicit promise, such as the implicit undertaking
not to tell lies which seems to be implied in the act of entering into conversation (at
any rate by civilized men), or writing books that purport to be history and not fiction.”
W. D. Ross, The Right and the Good (Oxford: Oxford University Press, 1930), 21. Nicolai
Hartman also defends the view that lying violates an implicit promise to tell the truth. See
Ethics, trans. Stanton Coit (Atlantic Highlands, N.J.: Humanities, 1975), 286. See also Fried:
“Every lie is a broken promise, and the only reason that this seems strained is that in lying
the promise is made and broken at the same moment. Every lie necessarily implies—as does
every assertion—an assurance, a warranty of its truth” (Right and Wrong, 67).
16. I am not sure that I agree with Ross that an implicit promise is broken whenever
someone lies. Sometimes lying involves breaking an explicit promise to tell the truth.
expression “it’s true that” is redundant in the context of ordinary state-

ments. Consider the following:
(1) The sky is blue.
(2) It’s true that the sky is blue.
Statements (1) and (2) have exactly the same meaning. In ordinary con-
texts, the expression “it’s true that” adds nothing to the meaning of a
statement. This is so because, in ordinary contexts, when one makes a
statement one is understood to be warranting its truth.
Taken together with my earlier criticisms of L1 and L2, the view that
making a statement (ordinarily) involves warranting that what one says is
true suggests the following definition of “lying”:
true), and 3. S states x in a context in which S thereby warrants the truth
of x.
This definition avoids the earlier counterexamples. It counts the witness’s

testimony in court as a lie. When the witness testifies in court, he warrants
the truth of what he says by explicitly promising to tell the truth under
oath. Definition L3 allows us to say that it is possible for me to lie to you
when I know that you know that I am lying so that I have no hope of
deceiving you either about the truth of what I say or about what I believe.
(For more on the implications of my definition for the cases from section
1.5, see section 1.10.)
1.8 Digression: “The Transparency Thesis”

The view that the expression “it is true that” adds nothing to the mean-
ing of a statement is what philosophers of language call the “transpar-
ency thesis.” The classic statements of the transparency thesis are found
in Frege17 and Frank Ramsey. Ramsey writes: “It is true that Caesar was
murdered” means no more than that Caesar was murdered, and “It is
false that Caesar was murdered” means no more than that Caesar was
not murdered.18
Strawson offers a noteworthy criticism of the transparency thesis in
his essay “Truth.”19 He argues that “x” and “it’s true that x” cannot always
be used interchangeably. For example, suppose that I say that Ingrid is
17. “My Basic Logical Insights,” in Posthumous Writings, trans. P. Long and R. White
(Oxford: Blackwell, 1959), 151.
18. “Facts and Propositions,” in Ramsey, The Foundations of Mathematics, (London:
Routledge and Kegan Paul, 1931), 142.
19. Analysis 9 (1949), 83–97.
having an extramarital affair and you deny it. My responding with “But it’s
true that she is having an affair” can constitute a stronger, more emphatic
response than simply repeating “she is having an affair.” In many border-
line cases, it is unclear whether or not making a statement involves war-
ranting its truth. In such cases, saying that x is true warrants the truth of
x, but merely saying that x (without saying that x is true) does not. For
example, suppose that I know of a humorous and improbable fact about
Judy. I wish to state this fact and warrant its truth in the context of a
humorous “bull session.” Simply stating the fact in this situation probably
is not warranting its truth. In order to accomplish this, it may be neces-
sary for me to say something like “It’s true that ; I’m not kidding.”
Strawson’s criticisms of the very strong version of the transparency thesis
according to which “x” and “it’s true that x” can always be used inter-
changeably are consistent with my view that, in ordinary contexts, the
expression “it’s true that” adds nothing to the meaning of a statement. The
fact that saying “it’s true that x” sometimes is a stronger warranty of the
truth of x than just saying “x” is consistent with my view that, ordinarily, a
person who states something warrants its truth (indeed, this fact presup-
poses the truth of my view).
1.9 The Concept of Warranting

A warranty of truth is a kind of guarantee that what one says is true.
It is also a kind of promise that what one says is true. Following Aus-
tin and Searle, contemporary philosophers generally take promising to
be a performative act. To make a promise is to place oneself under
an obligation to do something. This explains the difference between
promising to do x and stating an intention to do x.20 There seems to
be no satisfactory alternative explanation. However, special problems
arise if we attempt to extend this account of promising as an analysis
of warranting the truth of a statement. If one promises to do x, one
is placing oneself under an obligation to perform a specific act (one is
placing oneself under an obligation to do x). However, often when one
warrants the truth of a statement, one is not placing oneself under an
obligation to perform any particular action or kind of action. To war-
rant the truth of a statement x is not necessarily to place oneself under
an obligation to make it true that x, for one is usually not in a position
to affect the truth of the statements one makes. If I warrant the truth
20. See John Searle, “How to Derive ‘Ought’ from ‘Is,’ ” in Theories of Ethics, ed.
Philippa Foot (Oxford: Oxford University Press, 1968), 103–4. The default warranty of
truth is an example of what Paul Grice calls a “conversational implicature.” Like other
conversational implicatures, it is governed by rules and expectations understood by
language users. See Paul Grice, Studies in the Way of Words (Cambridge, Mass.: Harvard
University Press, 1989), chs. 2 and 3.
of my statement that the Moon is 250,000 miles from Earth, I am not

placing myself under an obligation to make it the case that the Moon
is 250,000 miles from Earth. Nor does warranting the truth of x place
one under an obligation to perform acts of compensation in case x turns
out not to be true and others are harmed as a result of relying on one’s
statements. There is no general understanding about what (if anything)
we owe others when they suffer harm as a result of accepting false
claims that we make.
In our linguistic community, and almost all others of which I am
aware, there is a presumption that the warranty of truth is in force in
any situation.21 Convention dictates that one warrants the truth of one’s
statements in the absence of special contexts, special signals, or cues to
the contrary. In the context of a work of fiction or when saying some-
thing in jest, one is not guaranteeing the truth of what one says. So, for
example, one is not implicitly guaranteeing that what one says is true
if one says something manifestly false as a joke to a friend in an ironic
tone of voice. In many cases, it is unclear whether those who speak are
warranting the truth of what they say. For example, suppose that I delib-
erately make a false statement to a person whom I know to be very
gullible, but give a very subtle indication that I might be joking (I might,
for example, raise an eyebrow very slightly). In such a case, it is unclear
whether I am warranting the truth of what I say and, therefore, unclear
whether or not this should be considered a lie. Such cases should be
considered borderline cases for the concept of lying. It is a virtue of my
21. J. A. Barnes describes a case that is arguably a counterexample to the claim

that there is a default warranty of truth for statements in every linguistic community.
Barnes gives the following description of a village in Lebanon: “In this community
liars are typically young men or children. Lies are told for fun, to trick one’s friends.
Success in lying depends on skill, but the final triumph comes when the liar reveals
his lie to the dupe and claims victory. There is thus an attitude of playful competition
towards lying, somewhat similar to the attitude towards tricks played on the first of
April in some countries. Lying is indulged in sometimes for its own sake, without an
instrumental motive. It is not surprising that in an environment of this kind there are
special devices to indicate that what is about to be said is true, and not a lie. These
markers for code switching are phrases such as ‘seriously’, ‘will you believe me’,
‘without joking’, ‘by your life’, and ‘by your father’s life.’ ” J. A. Barnes, A Pack of Lies
(Cambridge: Cambridge University Press, 1993), 70–71. The last part of this quotation
provides an indirect confirmation of my view. The fact that, in ordinary contexts, we
(and most other linguistic communities) do not need special markers to indicate that
what is about to be said is true confirms the idea that in those communities there is
a default presumption that what is said is warranted to be true. Given my definition
of lying, many of the cases Barnes describes as cases of “playful lying” are not genuine
instances of lying, since the false statements in question aren’t warranted to be true.
This, I would suggest, is a virtue of my definition. The villagers Barnes describes have
different understandings about when statements are warranted to be true than people
in most other societies. This is the salient difference between their society and most
other societies.
analysis that they count as such.22 Many instances of bluffing are border-
line cases of warranting/lying.23
1.10 Conditions for Warranting the Truth of a Statement

If one warrants the truth of a statement, then one promises or guaran-
tees, either explicitly or implicitly, that what one says is true. The idea
of explicitly promising to tell the truth is straightforward and needs no
explanation here. A witness who swears an oath explicitly promises that
what she says is true. One can also explicitly promise that what one says
is true by means of such locutions as “I swear on my honor . . . ” or “I am
going to tell it to you straight . . . ” In ordinary circumstances, statements
are warranted to be true; the default is that a statement is warranted to
be true. Because of this default warranty of truth, statements ordinarily
invite trust and reliance.
A certain type of lying merits special attention and comment. Some-
times people lie when they know that others know that they are lying.
I can lie to you in claiming that X, even if I know that you know that X
is false and I also know that you know that I know that X is false. In such
cases, I lie to you, even if I don’t intend to deceive you either about the
truth of X or about what I believe. In these kinds of cases, I invite reliance
on what I say by warranting its truth. It is possible for me to issue you
an invitation, even if I know that you know that I know that you won’t
accept the invitation (and even if I know that you know that I hope you
won’t accept the invitation). I can invite my estranged uncle to attend my
wedding while knowing and hoping that he will not come. This invitation
is not voided by our mutual understanding of the fact that I know and
hope that he will not accept the invitation. Similarly, lies can and do invite
22. It might be argued that the notion of warranting makes implicit appeal to the
notion of intention. It has been suggested to me that “to warrant the truth of a statement
is to make an utterance under conditions where normally the speaker intends that others
believe him.” This suggests that I have removed the notion of the intention to deceive
from the concept of lying only to smuggle it back in under the rubric of the concept
of warranting. Normally, we intend that others believe our statements. It is difficult to
imagine how it could be the case that the use of language involves the default warranty
of truth if this were not the case. There would be no point in warranting the truth of what
we say unless we sometimes intended that others believe what we say. There would be
no point in having a default presumption that any given statement is warranted to be true
unless we usually intended that others believe what we say. (The [or a] purpose of having
such a warranty is to make it more likely that people will believe what we say.) I accept all
of this, but it leaves untouched my earlier objection to the standard dictionary definitions
of lying. According to these definitions, a particular statement on a particular occasion
cannot be a lie unless the person who makes it intends thereby to mislead others. I believe
that I have shown this to be untenable.
23. See my essay “Second Thoughts on Bluffing,” Business Ethics Quarterly 3 (1993),
317–41.
reliance on what is stated, even if the liar hopes and knows that her audi-
ence won’t believe or rely on her statements.
The cases presented in section 1.5 illustrate several distinct sorts of rea-
sons one can have for lying in such cases. It might be the case that one will
benefit simply by making the false statement in question, for example, by
receiving a bribery payment or avoiding a harm that was threatened. So,
one might lie by making a false statement in order to receive the benefit
or avoid the harm. In addition, one might lie in order to “go on record” as
saying something. The student in my earlier example from section 1.5 lies
in order to go on record claiming to be innocent. He does this because
he knows that this will make it less likely that he will be punished for
cheating.
Convention dictates that there are circumstances that remove the
default warranty of the truth for our statements. The great majority of the
cases in which the warranty of truth does not hold involve storytelling
or attempts to be humorous. Indeed, it is difficult to think of statements
that are not warranted as true that do not involve either storytelling or
humor. In the case of statements that are clearly not intended be humor-
ous or tell a story, the default warranty of truth is very strong. The stu-
dent’s statement to the dean is such a case. If the student’s response to the
dean’s questions include obvious winks and nods and unguarded nervous
laughter, then the warranty of truth may be removed (or at least it is cast
into doubt).24 However, if the student “plays it straight” and looks grave
and serious, then his statements are warranted to be true and count as lies
according to my definition. It is not paradoxical or contradictory for me
to promise you that a statement is true, when I know that you know that
I know it isn’t true. I can promise you something, even if you and I both
know that I won’t keep the promise; I can also make a promise to you in
bad faith even if I know that you know that I am making the promise
in bad faith. In such a case, my promise invites your reliance, but you
would not be justified in relying on what I say. Thus, my warranting the
truth of something I say to you justifies you in complaining to me if it isn’t
true, even though it doesn’t always justify you in relying on it.
Whether or not a person warrants the truth of what she says on a
given occasion depends on the context and the relevant local conventions
embedded in that context. Far too many possible contextual factors are
relevant for it to be possible to state necessary and sufficient conditions
for warranting the truth of a statement. However, the following observa-
tions help explicate the notion of warranting the truth of a statement.
24. It might not be possible for the student to lie to the dean with the intention of
causing him to disbelieve what he says by accompanying his statement with a series of
winks and nods. The winks and nods may remove the default warranty of truth. For my
purposes, I don’t need to deny this. I only need to claim that it is possible to lie without any
intention of deceiving others (about the truth of what one is saying, or about one’s own
sincerity, or anything else).
(1) Whether or not a speaker (communicator) warrants the truth of

what she says is, at least in part, independent of her intentions. One can
warrant the truth of a statement without intending to, and one can even
warrant the truth of a statement contrary to one’s intentions. The fol-
lowing example illustrates this. Suppose that I am asked to speak at two
different banquets. At the first banquet, I am asked to give a serious talk
about the current political situation and the job performance of the new
president of the United States. At the second banquet, I am expected to
give a humorous or satirical talk about current politics and political fig-
ures. I become confused about which talk is supposed to be serious and
which is supposed to be humorous. I deliver my humorous satirical talk to
the group that has asked me to give a serious talk. I tell them a story about
the president having “broken wind” during a meeting with foreign digni-
taries. I warrant the truth of this story to my audience, even though I don’t
intend to. I give my serious talk to an audience expecting a humorous talk.
During my talk, I relate certain curious news about the president’s health.
The audience takes it to be a lame joke. Even though I take myself to be
warranting the truth of what I say, I am not. Extending the last example
further, suppose that I concoct a false, but humorous, story to discredit a
political figure before an election. I intend to tell this story and warrant
its truth to an audience interested in current events, but, unbeknownst
to me, I am speaking to an audience expecting to hear political satire and
humor. My statements about the politician are false, I know that they are
false, and I intend to warrant their truth. However, my definition implies
that I haven’t lied, because, contrary to my intentions, I did not warrant
the truth of my statement. In this case, I intended to lie, but failed to due
to my failure to warrant the truth of what I said.
(2) Whether or not the truth of a statement is warranted to an audience
is independent of whether the members of the audience believe that the
speaker (writer) warrants its truth to them. I can be mistaken in believ-
ing that the truth of a statement is or is not warranted to me. Sometimes
people are obtuse and fail to perceive that the things they are told are said
in jest. In such cases, they may be mistaken in believing that the speaker
(writer) warrants the truth of what he says (writes). Similarly, a member
of an audience can be mistaken in believing that a speaker (writer) is not
warranting the truth of what he says (writes). Suppose that a historical
society and an association of comedians are both holding meetings in a
particular hotel. I enter a conference room of the hotel expecting to hear
a comedian perform. However, unbeknownst to me, the speaker in the
room is a member of the historical society. As I walk in, the speaker is
relating a humorous event in the life a well-known historical figure. Even
though I don’t believe that the speaker is warranting the truth of what he
says, he is.
In many contexts, it is unclear whether the speaker/writer warrants
the truth of what she says. Many examples can be given to illustrate this
point. Suppose that an English-speaking acquaintance from a non-Western
country tells me a very bizarre story on April Fool’s Day. I don’t know
whether he knows anything about American practices and understand-
ings about April Fool’s Day. In the absence of other contextual clues, it
is unclear whether he warrants the truth of what he says. Suppose that a
professor is well known for her animated, humorous, and Socratic style of
teaching. She often says false things to her classes to see if they are atten-
tive. When she does this, she berates her students if they fail to catch her.
She begins her lecture with an improbable-sounding historical anecdote.
In this context, it is unclear whether the professor warrants the truth of
what she says. In the course of Socratic classroom debates in which pro-
fessors play the devil’s advocate, it is often unclear whether they warrant
the truth of what they say—philosophy classrooms are rife with this kind
of ambiguity.
This sort of ambiguity is increased if the speaker (writer) and the audi-
ence are members of different societies that have different conventions
and expectations about truth-telling and warranting the truth of state-
ments. There are often contexts in which one party expects that what is
said is warranted to be true and the other party does not.25
1.11 Yet Another Revision

The foregoing example of a person who warrants the truth of a statement
when he intends not to raises problems for my definition of lying. Accord-
ing to my definition, the speaker in the earlier example who mistakenly
believes that his audience is expecting a humorous talk tells a lie (albeit
unintentionally) when he tells the story about the president’s flatulence.
His statement is false, he knows that it is false, and he warrants its truth
to his audience. It seems counterintuitive to call this a lie, even if we
stress that it is an unintentional lie. Our definition of lying seems to need
another condition, namely, that it is not the case that the speaker takes
himself to be not warranting the truth of what he says. The speaker in this
example does not satisfy this condition, because he takes himself to be
not warranting the truth of what he says. With this additional condition,
the definition reads as follows:
true), 3. S states x in a context in which S thereby warrants the truth of x,
and 4. S does not take herself to be not warranting the truth of what she
says.26
25. J. A. Barnes gives numerous examples of this; see Pack of Lies, 114, and elsewhere.
26. Alternatively, we could say that in order to lie one must believe that one is
warranting the truth of what one says. I believe that my condition 4 is preferable to this
alternative condition, because there are many cases of lying in which the liar has no
conscious beliefs about whether or not she is warranting what she says.
1.12 A Complication and My Final Definition

Problems are created by cases in which statements are made to groups
of individuals who have differing levels of knowledge and sophistication,
so that the truth of a given statement on a given occasion is warranted to
some, but not all, of the people to whom the statement is made. Consider,
for example, a greatly exaggerated account of a past event told to a mixed
group containing both sophisticated adults and young children: “The dog
who was chasing me was huge; he was at least ten feet tall.” (Note, I am
supposing that the context of this is not just “a story” but an account of
something that was alleged to have actually happened.) In such a case, one
warrants the truth of what one says to the children but not to the adults.
The very content of the narrative makes it clear to the adults that its truth
is not being warranted to them. However, there is nothing one does and
nothing about the context or content of what one says that removes the
default warranty of truth to the children. We need to relativize our con-
cept of lying and allow for the possibility that, in making a given statement
or utterance on a particular occasion, one might be lying to some members
of one’s audience, but not to others.27 I propose the following:
L5. A person S tells a lie to another person S1 iff: 1. S makes a false
statement x to S1, 2. S believes that x is false or probably false (or,
alternatively, S doesn’t believe that x is true), 3. S states x in a context
in which S thereby warrants the truth of x to S1, and 4. S does not take
herself to be not warranting the truth of what she says to S1.28
1.13 Some Comments on This Definition

Condition 1 says that a statement must be false in order to be a lie.Whether
we count a statement as true or false sometimes depends on the standards
of precision and accuracy we employ. For example, suppose that someone
asks me what time it is. My watch is very accurate and displays the follow-
ing time “10:01:15” (one minute fifteen seconds after ten o’clock). I tell
the person “it’s ten o’clock.” Is my statement true? It depends on the con-
text. If someone asks this question in a context in which it is understood
27. Chisholm and Feehan also relativize the concept of lying in this way. Their
definition is a definition of what it is for one to lie to a particular person. Although they
don’t explicitly say this, their definition allows for the possibility that one and the same
statement could be a case of lying to one person but not a case of lying to the other.
28. Those who don’t share my view that a lie must be a false statement might wish to
modify L5 as follows:
L5'. A person S tells a lie to another person S1 iff: 1. S makes a statement x
to S1, 2. S believes that x is false or probably false (or, alternatively, S doesn’t
believe that x is true), 3. S states x in a context in which S thereby warrants the
truth of x to S1, and 4. S does not take herself to be not warranting the truth of
what she says to S1.
by all that the answer needs to be extremely accurate and precise (e.g., an
engineer at a radio station who needs to know when to begin broadcasting
a program or an astronaut who needs to know when to begin a maneuver)
then my statement is false. On the other hand, if I am asked this question
by my wife while taking a leisurely walk on our vacation when neither
of us needs to know exactly what time it is, my answer should count as a
true statement. My concept of warranting proves useful for dealing with
this issue. In some contexts, one warrants what one says as true to a high
degree of accuracy and precision. In other contexts, statements are not
warranted to be true to a high degree of accuracy.
Some hold that a necessary condition for telling a lie is that commu-
nication actually occurs, that is, that what is said is actually conveyed or
communicated to another person.29 (This could also be construed as a
necessary condition for making a statement.) According to this view, I can
attempt to lie, but fail, if I don’t communicate successfully. Suppose that
I speak to someone and intentionally say something that is false in a situa-
tion in which I know that I am warranting the truth of what I say. On the
view in question, my utterance can’t be a lie unless I succeed in communi-
cating with the other person. If the person to whom I am speaking is deaf
or doesn’t speak the language I am using, then I haven’t succeeded in mak-
ing a statement to her and I cannot be said to have lied to her (although
I attempted to). Or, to take another example, suppose that I am talking
to you in a different room of a house so that we can’t see each other. The
conversation stops for half an hour. After this interval, I intentionally say
something I know to be false intending to warrant the truth of what I say.
But, unbeknownst to me, you have left the house (you are many miles
away) and do not hear what I say. I am inclined to think that in this case
I didn’t lie to you, although I intended to. Others report very different
intuitions about this case. I don’t have any decisive arguments that settle
this question. In any case, we can leave the definition of lying as it is. To
endorse my definition, we don’t need to settle the question of whether
making a (false) statement requires that one succeeds in communicating
with another. This question concerns the proper interpretation of the first
condition of the definition. The question is whether successfully commu-
nicating with another person is necessary for making a statement.30
29. For example, Frederick Siegler, “Lying,” American Philosophical Quarterly 3 (1966),
128–36, and Fried, Right and Wrong, 57.
30. There are some related issues and puzzles that should be briefly noted here. What
should we say about statements that are successfully conveyed to others, but only long
after they are made? Suppose that a historical figure writes many false things in his diary in
the hope of misleading future readers and historians. The diary isn’t read by anyone until a
hundred years later. When he writes this, can we say that he is lying? Can we say that he is
lying, but that his lying (now) depends on what will happen in the future? Suppose that a
politician writes a false account in his diary hoping to mislead future historians. The diary
is later destroyed, and no one else ever reads it. Should we say that he intended to lie but
failed to because his statements were never successfully conveyed to anyone else?
1.14 An Objection: The Concept of Assertion

A number of the alternative definitions of lying discussed in this article
make use of the notion of making an assertion. One might argue that lying
should be defined in terms of making an assertion rather than my concept
of warranting the truth of a statement.31 Taking this suggestion, one might
claim that the following is preferable to (and simpler than) my definition
of lying:
L6. A person S tells a lie to another person S1 iff: 1. S asserts a false
proposition x to S1, 2. S believes that x is false or probably false (or,
alternatively, S doesn’t believe that x is true), and 3. S does not take
herself to be not asserting x to S1.
Whether or not L6 is a plausible definition depends on what sort of expla-

nation of “asserting a proposition” is provided. Without a detailed account
of what is meant by “asserting a proposition,” L6 is vague and ambiguous.
As some understand the concept of an assertion, asserting a proposition
is roughly the same as warranting its truth. Given this, L6 is equivalent
or nearly equivalent to my L5, and the details of my account of warrant-
ing the truth of a statement are still needed to give content to the con-
cept of assertion used in L6. Definition L6 has a superficial appearance of
greater simplicity created by the fact that it includes three clauses instead
of four.
If L6 is combined with a concept of assertion substantially different
from my notion of warranting an assertion, it does not yield an accept-
able definition of lying. I will attempt to show that this is the case for the
other alternative accounts of asserting a proposition presented earlier in
this chapter. Fried, Williams, and Chisholm and Feehan each define lying
in terms of the concept of an assertion. I have offered criticisms of their
definitions of lying. Still, we might ask whether any of their analyses of
the concept of assertion yields a plausible definition of lying when com-
bined with L6.
Fried writes: “As a first approximation: to assert X is to utter X in a
context such that the utterance is intended to cause belief.” 32 If we com-
bine L6 with Fried’s account of assertion, we are committed to the view
that in order for my stating X to be a lie I must intend that others believe
X. However, this is not the case, for reasons that I gave earlier in sections
1.5 and 1.10. I can lie even if I do not intend to cause you to believe what
I say. Definition L6 is not a plausible definition of lying when combined
with Fried’s concept of assertion.
Williams defines what it is to assert a proposition in the following
passage:
31. Bill Tolhurst was the first of several who made this suggestion to me.
32. Right and Wrong, 56.
S utters sentence “S,” where “S” means that P, in doing which either he
expresses his belief that P, or he intends the person addressed to take it
that he believes that P.33
When combined with L6, Williams’s concept of making an assertion

implies that a necessary condition of one’s telling a lie in saying X is
that either one expresses one’s belief that X or one intends to cause the
person(s) one addresses to believe that one believes that X. This is also
false for reasons given in sections 1.5 and 1.10.
Chisholm and Feehan define asserting a proposition in the following
passage:
L asserts p to D = df L states p to D and does so under conditions which,
he believes, justify D in believing that he, L, not only accepts p, but also
intends to contribute causally to D’s believing that he, L, accepts p. 34
When combined with L6, this definition of assertion implies that, in order
to lie to you in stating proposition X, I must state X to you in condi-
tions that I believe justify you in believing that I accept X and intend to
contribute to your believing that I accept X. As I have shown, however, a
notoriously dishonest person can lie to you in stating X even if he knows
that you are not justified in believing that he accepts X and he knows that
you are not justified in believing that he aims or intends to cause you to
believe that he believes that X.
1.15 Reasons to Accept My Definition

Lying is a concept used in everyday language, and moral questions about
lying arise in people’s everyday experience. There are no compelling rea-
sons to revise or reject the ordinary language concept of lying—at least
the burden of proof rests with those who would revise or reject it. There-
fore, consistency with ordinary language and people’s linguistic intuitions
about what does and does not count as a lie is a desideratum of any defi-
nition of lying. My definition provides a better account of our shared
linguistic intuitions about what does and does not count as a lie than
any of the alternative definitions considered in this chapter. My defini-
tion also explains why many cases are unclear or borderline cases for the
concept of lying. (In many cases, it is unclear whether or not one warrants
the truth of what one says.) My definition is near the mark in terms of
ordinary language, but no definition can be completely consistent with
everyone’s linguistic intuitions about what does and does not constitute
lying. People have conflicting intuitions about certain cases and certain
issues, for example, whether a lie must be a false statement.
In every language and culture, there are contextual understandings
about whether a given utterance or use of language should be taken
33. Truth and Truthfulness, 74.

to be a statement that is warranted to be true. 35 These understandings

often determine whether or not an utterance or linguistic act is a lie. One
and the same utterance can be a lie in some contexts but not others. For
example, a false utterance that would ordinarily count as a lie may not
be a lie if it is uttered on April Fool’s Day.36 A sentence that would be a
lie if it appeared in a police report or history book would not be a lie if
it appeared in a novel.37 J. A. Barnes holds that the best explanation of
why such cases are or are not instances of lying is that lying requires the
intent to deceive others.38 According to Barnes, an utterance that is not
intended to be a statement that purports to be true is not a lie, because it
is not made with the intention to deceive others. Barnes’s explanation is
implausible, for reasons given earlier in sections 1.5 and 1.10. My defini-
tion provides a better explanation of this than Barnes’s or any alternative
definition. My definition also provides a good explanation of why there is
often disagreement about whether a given utterance made by a member
of one society to a member of another society is a lie. The explanation is
that often when members of two different societies interact, they don’t
share the same implicit understandings about how statements are to be
taken in given contexts. A context that a member of one society takes to
involve warranting the truth of what one says is not taken to involve a
warranty of truth by a member of the other society.39
Strawson’s criticisms of the “transparency thesis” show that there are
weaker and stronger ways of warranting the truth of a statement. To count
as a lie, a statement must be warranted to a certain minimum degree, but
some lies are warranted to a much greater degree. One’s expressions of
confidence or lack of confidence in the certitude of one’s statements also
strengthen or weaken the degree to which one warrants them as true. The
strength with which a lie is warranted to be true is arguably relevant to
its moral assessment. My definition of lying has the virtue of making per-
spicuous what is arguably a morally relevant consideration. My definition
also helps us to identify and diagnose an important type of deception—
see section 2.1.
My definition doesn’t beg any controversial moral questions about
lying, for example, is lying prima facie wrong? Is lying always wrong? My
definition helps illuminate moral questions by identifying morally salient
features of actions.
A final virtue of my analysis is that it makes sense of the common view
that lying involves a breach of trust. To lie, on my view, is to invite oth-
ers to trust and rely on what one says by warranting its truth, but, at the
35. Barnes, Pack of Lies, defends this claim at length and gives numerous examples.
36. See Barnes, Pack of Lies, 113.
37. Barnes, Pack of Lies, 166.
38. Barnes, Pack of Lies, 15, and 113–14.
39. Barnes gives numerous examples of this; see 114 and elsewhere in Pack of Lies.
same time, to betray that trust by making false statements that one does
not believe.40
2. DECEPTION
2.1 The Concept of Deception

As a first approximation, we might say that to deceive someone is to
cause her to have false beliefs. The New Shorter Oxford English Dictionary
defines the verb “deceive” as to “cause to believe what is false.”41 This defi-
nition is too broad, because not all cases of causing another person to have
false beliefs constitute deception. If a perfectly clear and truthful state-
ment is misinterpreted by others and causes them to have false beliefs it
is not necessarily a case of deception. An automobile salesperson does not
necessarily deceive her customers if her clear and accurate description of
a car, which includes the claim that the car has side and front air bags,
causes a buyer to believe falsely that the car is safe in case of high-speed
collisions with large vehicles.
We might try to define deception as “directly” causing someone else to
have false beliefs. This is unpromising, for two reasons: (1) some of the
subtler forms of deception involve indirection, and (2) the distinction
between direct and indirect causation is too vague to be helpful for the
purposes of conceptual analysis. Since deception is an ordinary language
concept, any definition of deception that employed a very complicated
or torturous distinction between direct and indirect causation would be
implausible. No doubt, some readers will be unpersuaded by these two
points and will be inclined to try to define deception in terms of directly
causing others to have false beliefs. I won’t pursue this matter, since there
is a much simpler and more straightforward way to deal with the two
examples in question.
The best explanation of why these two cases are not instances of decep-
tion is that they involve no intention on one person’s part to mislead the
other person. Deception requires some kind of intention to cause others
to have false beliefs. I take it to be self-contradictory to say that some-
one deceived another person unintentionally. Consider the following case.
40. See Chisholm and Feehan: “The liar would have his victim believe that, at the
moment at least, the liar is someone in whom he may place his faith. Thus we may say,
with St. Augustine: “No liar preserves faith in that about which he lies. He wishes that he
to whom he lies have faith in him, but he does not preserve faith by lying to him” (“Intent
to Deceive,” 152). (The quotation from Augustine appears in “Christian Instruction,” in
Fathers of the Church, vol. 2, Writings of St. Augustine, vol. 4, trans. John Gavigan [New
York: CIMA, 1947], 57.) Also see Fried, Right and Wrong: “A lie invites belief in an
assertion which the speaker knows to be false,” 57.
41. New York: Oxford University Press, 1993.
Suppose that I am told something that is false. I believe what I am told

and repeat it to you, thereby causing you to believe it. (In reporting this to
you, I am not trying to bring it about that you believe anything other than
what I was told and believe to be true.) Some people think that it makes
sense to describe this as a case in which I deceived you “albeit without
intending to do so.”42 Here people’s linguistic intuitions conflict; no pos-
sible definition of deception can be consistent with everyone’s linguistic
intuitions about the word “deception.” There are pragmatic reasons to
accept the view that deception must be intentional. The words “decep-
tion” and “deceive” are typically terms of reproach or condemnation. (The
word “mislead” does not imply the same kind of reproach or negative
evaluation as the word “deceive.”) The negative evaluative connotations
of the term “deception” are inappropriate if we allow for the possibility of
completely unintentional deception (for which people are often blame-
less). There is an important distinction to be drawn between intentionally
and unintentionally causing others to have false beliefs. Rather than coin
new words, we can use the terms “deceive” and “mislead” (or “inadver-
tently mislead”) to mark this distinction. It is not self-contradictory to say
that someone misled another person unintentionally or inadvertently.
In order to deceive you, I must intentionally mislead you or intention-
ally cause you to have false beliefs. However, this is ambiguous. It is not
deception if I intentionally cause you to believe that x where x is false, but
I myself believe that x is true. This would be a case of misleading some-
one, but it would not be deception. Intentionally causing you to believe x
where x is false is not sufficient for deception. We might say that in order
for there to be deception, it is necessary that the deceiver believes that
what she causes the other person(s) to believe is false. Consider the fol-
lowing definition of deception:
D1. A person S deceives another person S1 if, and only if, S intentionally
causes S1 to believe x, where x is false and S knows or believes that x is
false.
This is much closer to the mark than the previous definition, but sev-
eral questions remain. What if I intentionally cause you to believe that
x where x is false and I neither believe that x is true nor believe that x
is false? For example, suppose that I want you to believe x where x is
false and I haven’t given any thought to the question of whether x is true
or false. (Such statements are characteristic of bullshit.) 43 If we want to
count this as a case of deception, we should reject D1 in favor of some-
thing like the following:
42. Al Mele, for one, claims that it is possible to deceive others unintentionally; see his
Self-Deception Unmasked (Princeton: Princeton University Press, 2001), 8–9.
43. Harry Frankfurt claims that lack of concern for the truth and falsity is one of the
most salient features of bullshitters. On Bullshit (Princeton: Princeton University Press,
2005), 33–34, 47–48, and 56–57.
D2. A person S deceives another person S1 if, and only if, S intentionally
causes S1 to believe x, where x is false and S does not believe that x is true.44
Clearly, a modified version of D1 gives sufficient conditions for deception.

A person who intentionally causes another person to believe x, where x is
false and he knows or believes that x is false, has clearly deceived the other
person.45 A modified version of D2 gives necessary conditions for decep-
tion; a necessary condition of my deceiving you is that I intentionally cause
you to believe something false that I do not believe to be true. Ordinary
language does not give us a clear basis for choosing between D1 and D2.
The kinds of cases in which D1 and D2 give conflicting results, for example,
cases in which S intentionally causes S1 to believe x, x is false, and S hasn’t
given any thought to the question of whether x is true or false, are border-
line for the concept of deception. I don’t know of any decisive reasons for
preferring D1 to D2. Rather than claiming that one of the definitions is the
correct account of deception (or is closer to being correct than the other),
I think that we should simply say that there are broader and narrower con-
cepts of deception. Definition D1 defines deception in the narrow sense.
Definition D2 defines deception in the broader or looser sense.
My definition of lying helps to reveal and highlight an important kind of
deception (or attempted deception) short of lying. Suppose that I believe
that X is true, even though I am aware of serious reasons for doubting
the truth of X. I assert the truth of X on a solemn occasion in which
reliance on my statement will be used as a justification for an important
decision (such as the decision to start a war), and I intend to cause oth-
ers to believe X. I give prima facie evidence for the truth of X but make
no mention of the reasons for doubting the truth of X. My statement is
strongly warranted to be true—in this case my statement implies a very
strong assurance of its own truth. My statement is not a lie, since I believe
that X is true. Since I believe that X is true, my intending to cause others
to believe X is not a case of attempted deception. However, my statement
still intends to deceive in virtue of giving/implying a strong assurance of
its truth—an assurance that I hope others will rely on. I intend to cause
others to (falsely) believe that there is strong unambiguous evidence for
X. In this case, withholding information about the counter-evidence is
deception. It is arguable that some members of the Bush administration
were guilty of this kind of deception in their claims to the effect that Iraq
possessed “weapons of mass destruction.”
44. The distinction between D1 and D2 closely parallels the distinction I discussed in
section 1.2 between the view that a necessary condition of telling a lie is that one makes
a false statement that one believes is false (or believes is probably false) and the view that
a necessary condition of telling a lie is that one makes a false statement that one doesn’t
believe to be true.
45. This needs to be qualified in light of the next paragraph; D1' (later) gives sufficient
conditions for deception.
2.2 The Difference between Lying and Deception

Lying differs from deception in two important respects. First, in order to
lie one must make a false statement. Deception does not require that one
make a false statement or make any statement at all. True statements can be
deceptive, and some forms of deception don’t involve making statements.
Consider a variation on a case from 1.1. I am selling a used car that fre-
quently overheats, and I am aware of the problem. You are a prospective
buyer and ask me whether the car overheats. If I say “No,” I am lying. If
I answer by making the true statement “I drove the car 1000 miles on a very
hot day and had no problems,” I am not lying. Even though this statement is
true and I believe it, it might still be deceptive—perhaps the event I describe
happened four years ago and have had lots of trouble with it overheating
since then. Second, unlike “lying,” the word “deception” connotes success.
An act must actually mislead someone if it is to count as a case of deception.
Many lies are not believed and don’t succeed in deceiving anyone.
2.3 Deception versus Withholding Information and Keeping

Someone in the Dark
There is a clear distinction between the concept of withholding informa-
tion and the concept of deception (or attempted deception). To with-
hold information is to fail to offer information that would help someone
acquire true beliefs and/or correct false beliefs. Not all cases of withhold-
ing information are cases of deception. A businessperson who withholds
from her clients information about how much a product she is selling
costs her does not thereby deceive (or attempt to deceive) them about
her costs. However, withholding information can be deception if there
is a clear expectation, promise, and/or professional obligation that such
information will be provided. If a tax advisor is aware of a legitimate tax
exemption her client can claim that would allow him to achieve a con-
siderable tax savings, then her failure to inform him about it is deception.
She thereby intentionally causes him to believe falsely that there is no
way for him to save more money on his taxes.
Cases in which we try to prevent people from discovering the truth
raise interesting questions about the concept of deception. First, consider
cases in which we try to prevent people from learning things that we don’t
want them to know by distracting their attention. Suppose that you and
I are trying to close a business deal, and you are reading the contract that
describes the terms of a loan I am offering you. I don’t want you to read it
carefully, since I fear that you will be distressed by provisions that call for
you to surrender collateral in case of default. You begin to read the contract,
but I distract you from doing this by discussing the fortunes of our favorite
baseball team or engaging you in a political argument. Since you have only
a limited amount of time to look at the contract, you don’t read it care-
fully. You know that there are penalties in case of default, but your true
beliefs about this are vague and incomplete. In this case, I have prevented
you from gaining certain information. However, I haven’t caused you to
acquire any false beliefs and, therefore, I haven’t deceived you about any-
thing. Preventing someone from learning the truth about X is not the same
as causing her to have false beliefs about X (deceiving her about X).
We need a term other than “deception” to describe cases of preventing
others from learning the truth. The English expression that most closely
describes this is “keeping someone in the dark.” Intentionally and actively
preventing someone from learning something counts as keeping him in
the dark. Sometimes withholding information or failing to correct a per-
son’s false beliefs counts as keeping her in the dark. A lawyer keeps her
client in the dark if she fails to inform him that a certain course of action
she is advising him to take is likely to result in his being the subject of a
lawsuit. However, not every case of failing to correct another person’s false
beliefs or remove his ignorance about something one knows about counts
as “keeping someone in dark.” (I am not keeping my neighbor in the dark if
I fail to inform her about my past membership in the Cub Scouts.)
Here is a first stab at a definition of “keeping someone in the dark” (this
is both a proposal and an attempt to fit with ordinary language):
A person S keeps another person S1 in the dark about X (where X is
something that S1 doesn’t know) if, and only if, either: 1. S actively and
intentionally prevents S1 from learning about X, or 2. S fails to inform
S1 about X when either: (1) S knows that S1 wants the information in
question and S can easily give it to S1, or (2) S occupies a role or position
in which he is expected to provide S1 with the sort of information in
question.
Suppose that I attempt to prevent you from discovering the truth by caus-
ing you to be less diligent in your pursuit of the evidence or by tempo-
rarily diminishing your cognitive functioning. For example, suppose that
you are a detective working on a murder case. I know that my daughter
Astrid is guilty of the crime. But there is strong prima facie evidence that
John committed the murder, and no one suspects Astrid. Unless you are
diligent and alert, you will not suspect Astrid and will probably come to
believe that John is the murderer. I might try to cause you to believe that
John is guilty by causing you to have a bad night’s sleep or by tempting
you to get drunk with me. Suppose that I succeed in this. I get you drunk
and make noise during the night to cause you a very bad night’s sleep. The
next day you are hung over and very tired. As a result, you overlook key
clues about Astrid and come to believe that John is guilty. Suppose that
you would have discovered the truth were it not for my actions. In this
case, I have deceived you (in a very indirect way) about John’s guilt and
Astrid’s innocence. My actions intend and bring it about that you believe
falsely that John is guilty and Astrid is innocent.
Briefly consider a variation on this case. I don’t try to cause you to have
false beliefs about Astrid but just try to prevent you from acquiring true
beliefs about her. Suppose that you are unaware of Astrid’s existence. I want
to prevent you from learning the truth (that she is guilty), but I am not try-
ing to cause you to have any false beliefs about Astrid (that she is innocent),
since my purposes are served equally well if you remain unaware of her
existence. If I cause you to believe that John is guilty in order to protect
Astrid, I am deceiving you about John’s guilt and keeping you in the dark
about Astrid’s guilt, but I am not deceiving you about her guilt.
One last puzzle. Consider the following case. I intentionally tamper
with and alter the relevant evidence in order to try to cause you to believe
X when I know that X is false. Unbeknownst to me, you already believe
X. I don’t cause you to believe X, but I strengthen your belief that X. At
a later date, you acquire prima facie evidence for thinking that X is false.
You persist in believing X, but you would have ceased to believe X were it
not for my past action. Is this deception? Have I deceived you about X? It
is unclear what we should say about this case. Tentatively, I would say the
following. At the time I act, I haven’t deceived you into believing that X,
since I did not cause you to believe X. However, my actions will deceive
you into believing X at a later date.
My definition comes fairly close to capturing how most people use
the word “deception.” Since not everyone shares the same linguistic intu-
itions about how the word “deception” should be used, no definition of
deception can be consistent with everyone’s linguistic intuitions. There
are good pragmatic reasons to accept my definition. The concepts of lying
and deception are important from the moral point of view because they
help us pick out morally salient features of actions and classify different
types of actions. It is helpful to classify or categorize different types of
actions before we assess them morally. It is not helpful to assess the moral
status of deception or any other class of actions without first clearly delin-
eating that class of actions. The morally salient feature of deception is that
it involves intentionally causing others to have false beliefs. My concepts
of misleading others (causing others to have false beliefs) and deceiving
others (intentionally causing others to have false beliefs) pick out morally
salient features of actions and help facilitate normative assessments of
action. This is a distinction we should mark in some way. There is a clear
distinction between withholding information and deception. There is also
a clear distinction between intentionally preventing others from learning
the truth and deception.
3 THREE RELATED NOTIONS: BULLSHIT, SPIN, AND

HALF-TRUTHS
3.1 Bullshit and the Relationship between Bullshit, Lying,

and Deception
In his book On Bullshit, Harry Frankfurt claims that (1) bullshit requires
the intention to deceive, (2) producing bullshit is not lying (bullshit is a
kind of misrepresentation “short of lying”), and (3) bullshitters are unlike

liars in that they are unconcerned with the truth of what they say. I argue
that these three claims are all mistaken.
Bullshit and Deception

According to Frankfurt, bullshit necessarily involves deception or the
intention to deceive others. Bullshitters don’t necessarily attempt to
deceive others about the content of what they say; rather they deceive
or attempt to deceive others about themselves. “The bullshitter may not
deceive us, or even intend to do so, either about the facts or what he takes
the facts to be. What he does necessarily [my emphasis] deceive46 us about
is his enterprise. His only indispensably distinctive characteristic is that in
a certain way he misrepresents what he is up to.”47
Frankfurt’s example of the Fourth of July speech illustrates his idea of
misrepresenting what one “is up to.”
Consider a Fourth of July orator, who goes on bombastically about “our
great and blessed country, whose Founding Fathers under divine guid-
ance created a new beginning for mankind . . . ” It is clear that what makes
the Fourth of July oration humbug is not fundamentally that the speaker
regards his statements as false. Rather, just as Black’s account suggests, the
orator intends these statements to convey a certain impression of himself.
He is not trying to deceive anyone concerning American history. What he
cares about is what people think of him. He wants them to think of him as
a patriot, as someone who has deep thoughts and feelings about the origins
and the mission of our country.48
Bullshit and Lying

Frankfurt says that bullshit falls “short of lying.” 49 He claims that bullshit-
ters are unconcerned with the truth of what they say, whereas liars know
that what they say is false and are “guided by the truth” (see below); this
commits him to the view that bullshit cannot constitute lying.
Bullshit and lack of concern with the truth. Frankfurt famously claims
that the essence of bullshit is unconcern with the truth (how things are)
and that the bullshitter is a greater enemy of the truth than the liar (who
is concerned with knowing how things are).
46. Here, Frankfurt’s view would more plausible if he said “intends to deceive us about”
instead of “deceives us about.” Succeeding in deceiving others is not a necessary condition
for bullshit. One can bullshit others even if the fact that one is bullshitting is apparent to
them.
47. Frankfurt, On Bullshit, 54.
48. Ibid., 17–18. Frankfurt describes bullshit as a form of bluffing. He says that bluffing
is a mode “of misrepresentation or deception” (46). Also see 12 and 14.
It is just this lack of connection to a concern with truth—this indifference

to how things are—that I regard as the essence of bullshit.50
This points to a similar and fundamental aspect of the essential nature of
bullshit: although it is produced without concern with the truth, it need
not be false.51
The fact about himself that the bullshitter hides, on the other hand, is that
the truth values of his statements are of no central concern to him This
does not mean that his speech is anarchically impulsive, but that the motive
guiding and controlling it is unconcerned with how the things about which
he speaks truly are.52
Someone who lies . . . is guided by the authority of the truth The bullshit-
ter ignores these demands altogether. He does not reject the authority of
the truth, as the liar does, and oppose himself to it. He pays no attention to
it at all. By virtue of this, bullshit is a greater enemy of the truth than lies
are.53
In a very illuminating passage Frankfurt observes:

Bullshit is unavoidable whenever circumstances require someone to talk
without knowing what he is talking about. Thus, the production of bullshit
is stimulated whenever a person’s obligations or opportunities to speak
about some topic exceed his knowledge of the facts that are relevant to
that topic. This discrepancy is common in public life, where people are fre-
quently impelled—whether by their own propensities or by the demands
of others—to speak extensively about matters of which they are to some
degree ignorant.54
One important kind of bullshit involves trying to convey a certain

impression of oneself to others. For example, one might bullshit some-
one in order to convey the impression of knowledge, intelligence, piety,
or patriotism, as in Frankfurt’s example of the Fourth of July speech.
Another important kind of bullshit (that Frankfurt doesn’t say much
about) involves evasion. Sometimes we are pressured to answer ques-
tions we don’t want to answer. When asked such questions, people often
produce bullshit responses that don’t directly answer the questions. Here
is an example of this kind of bullshit. In a televised presidential debate, a
candidate is asked the following question: “I want to ask you about your
criteria for nominating people to the U.S. Supreme Court. Would you
be willing to nominate anyone who supports the Roe v. Wade decision?
Or, will you make opposition to abortion and Roe v. Wade a require-
ment for anyone you nominate?” The answer is that the candidate is
50. Frankfurt, On Bullshit, 33–34.

53. Frankfurt, On Bullshit, 60–61; also see 47–48.
not willing to nominate anyone who supports Roe v. Wade. She firmly
intends to nominate only people who oppose abortion and the Roe v.
Wade decision and has privately promised some antiabortion leaders that
she will do this. Answering the question truthfully is likely to cost her the
votes of many people who support abortion. Lying and saying that she
would be willing to nominate someone who supports abortion and Roe
v. Wade would cost her the votes of many people who oppose abortion.
The candidate wishes that the question had not been asked and gives the
following bullshit reply that completely fails to answer or address the
question that was asked: “Look, there are lots of things to be taken into
account when nominating someone for the Supreme Court. This isn’t
the only relevant consideration. I want someone with a good legal mind
and judicial experience who supports my judicial philosophy of follow-
ing the constitution as it is written.” (Note that the question does not ask
whether a potential nominee’s views about abortion are the only relevant
considerations. Rather, it asks whether the candidate will make opposi-
tion to abortion and Roe v. Wade a necessary condition for nominating
someone for the court.)
Similar examples of evasion by means of bullshit occur when people
are asked to testify in court or in other semilegal proceedings. Suppose
that I am a witness in court and am asked a question about a friend who
committed a crime I witnessed. An honest answer will incriminate him;
lying will make me guilty of perjury. Refusing to testify will make me
guilty of contempt of court. In this situation, resorting to evasive bullshit
is an attractive option for me.
Bullshit does not require the intention to deceive. Some cases of evasive
bullshit don’t involve the intention to deceive others. In cases of evasion,
we often want to deceive others and make them think we are answering
or trying to answer their questions. Witnesses and politicians who engage
in evasive bullshitting typically intend that others (mistakenly) believe
that they are trying to answer their questions. However, not all cases of
evasion by means of bullshit involve the intent to deceive in this way.
There are some cases in which one is rewarded for making a performance
in response to a question; in some of these cases, one will be rewarded
for saying almost anything rather than saying nothing. There are cases in
which one knows that one can obtain rewards for bullshit responses to
questions without deceiving other people about anything (including the
fact that what one says is pure bullshit). Consider the following example.
I am a student who needs to receive a good grade in a class. I am assigned
to write a short essay on a very clearly and precisely defined topic. I know
nothing about the topic and can’t write on it at all. But I know that my
instructor will give me partial credit for turning in something, however
incompetent and far off the topic. The worst grade I can receive for writing
something that is completely incompetent and off the topic is an F—60
percent. If I write nothing, I will receive a zero—0 percent. In producing
a bullshit answer, I am not attempting to mislead my teacher about my
level of knowledge or about what I am up to (namely bullshitting him).

I don’t care about any of these things; I just want to receive sixty points
instead of zero points. I might even want my bullshitting to be transparent
to the teacher in order to amuse or annoy her.55
Lying Can Involve Producing Bullshit

Contrary to what Frankfurt says, one can tell a lie while producing bullshit.
One can tell a lie as a part of an evasive bullshit answer to a question.
Suppose that I teach at a university that is very intolerant of atheists.
I am asked by an administrator whether a friend and colleague is an athe-
ist. I know that he is an atheist and that it will harm him if I reveal this.
I don’t want to harm my friend, nor do I want to lie and say that he isn’t
an atheist, since I fear that I am likely to be found out if I lie about this. So,
I give an evasive bullshit answer. I say “as a boy he always went to church
and loved singing Christmas carols” even though I know this to be false.
(I am not worried that I will be caught or found out if I lie in this way.)
My answer is evasive bullshit, but since I say what I know to be false in a
context in which I know that I am warranting the truth of what I say, my
answer is also a lie.
Bullshitters Can Be Concerned with the Truth

of What They Say
The bullshitting student might be concerned with the truth of what she
says and all matters relating to it. Suppose that she knows that the teacher
will bend over backward to give her partial credit if he thinks that she
misunderstood the question, but she also knows that if the things she
55. Here is an example of such an answer. A student is asked the following question
on an exam for an applied ethics class: “Briefly describe the facts of the case of Dodge v.
Ford and answer the following question: Was Henry Ford morally justified in his actions
in this case? Defend your answer.” The student hasn’t read the case, nor was she in class
when it was discussed, so she can’t describe the facts of the case. Since she doesn’t know
what Henry Ford did in the case of Dodge v. Ford, she can’t possibly formulate a coherent
argument for thinking that he was or was not morally justified in what he did. She
gives the following bullshit answer: “In today’s society there are many important ethical
questions about the role of business in the larger society. These are important questions
since business and its actions play such a large role in today’s society. We have addressed
these questions in our class. Milton Friedman holds that the only obligation of business is
to make money for the shareholders, provided that it avoids fraud, deception, and unfair
competition. R. Edward Freeman holds that corporations should be run for the benefit of
all their ‘stakeholders.’ Utilitarians hold that corporations should promote the social good.
The Ford Motor company had many obligations in this case. In this case, I think that the
obligation to society was the most important obligation. The company failed to live up
to this obligation, to an extent, but this is not a black or white issue. In any case, Henry
Ford didn’t adequately fulfill his duty to the public.” In the sort of case I have described, a
student could produce this kind of bullshit answer without intending to deceive anyone.
writes are false, she will be marked down. In that case, she will be very
careful to write only things that are true and accurate, even though she
knows that what she writes is not an answer to the question.
A politician who gives evasive bullshit answers to difficult questions
might still be concerned with the truth of what he says. Suppose that
he fears that the media will dissect his every public word and catch him
in false and inaccurate statements. In that case, he will care very much
whether what he says is true. Not only will he carefully craft evasive
bullshit statements and answers, he will be careful to make sure that what
he says is true.
Similarly, the witness in court who gives evasive bullshit answers may
be very concerned to say only things that are true so that he can avoid
charges of perjury. Contrary to Frankfurt’s claim that bullshitters are not
concerned with whether the statements they make are true or false,56 the
bullshitting witness may be very concerned with the truth value of his
statements.
Some bullshit deceives others and much bullshit is intended to deceive
others, but contrary to what Frankfurt says, some bullshit is not intended
to deceive others. One can bullshit even if one knows that one’s bullshit-
ting is completely transparent to others. One can lie without bullshitting,
and most bullshitting does not involve lying, but, Frankfurt to the con-
trary, one can produce bullshit while lying. Although Frankfurt’s claim
that bullshitters are unconcerned with the truth of what they say is an
apt and insightful description of much bullshit, some bullshitters are con-
cerned with the truth of what they say. Therefore, Frankfurt’s claim that
unconcern with the truth of what one says is the essence of bullshit is
mistaken. I don’t have a better alternative definition of bullshit that I am
prepared to defend. I suspect that the concept of bullshit is too loose and
amorphous to admit of a definition in terms of necessary and sufficient
conditions. Frankfurt, himself, says as much. 57 To paraphrase Frankfurt’s
own hopes for his account of bullshit, it is “helpful” (very helpful indeed)
but not “decisive” (definitive).58
3.2 Spin and Half-Truths

“Spinning a story” or “putting spin on a story” involves putting an inter-
pretation on events or facts that, themselves, are not in dispute. We say
that people spin stories/events when they are strongly disposed to place
a particular interpretation on them. Politicians and candidates for public
office often spin stories in such a way as to make themselves and their
policies look good and make their opponents look bad. Ideologues of all

stripes often spin their interpretation of events so that those events seem
consistent with their ideological commitments. Sometimes the interpre-
tations they spin are correct but sometimes not. If someone spins the
interpretation of an event, then his interpretation is biased and unreliable,
but not necessarily incorrect.
One common way of spinning events involves asserting “half-truths.”
Half-truths are true statements or sets of true statements that selectively
emphasize facts that tend to support a particular interpretation or assess-
ment of an issue and selectively ignore, or minimize, other relevant facts
that tend to support contrary assessments. For example, a politician spins
the interpretation of recent events to support the claim that his policies
were successful if he describes the good consequences of those policies in
great and vivid detail and omits any mention of their bad consequences.
A man’s description of his marriage to a friend (or the description he gives
to himself) is a half-truth if it contains a long and accurate account of
unkind and hurtful things that his wife has said and done to him but men-
tions only a few of the equal (or greater) number of unkind and hurtful
things he has said and done to her. Statements or narratives that involve
spin can be misleading if they advance unreasonable interpretations of
events and incline other people to accept those interpretations. Mislead-
ing spin counts as deception if the “spinner” knows or believes that the
interpretation he defends is unreasonable/implausible. If spin involves
making deliberate false claims about one’s state of mind (how one thinks
events should be interpreted), it usually involves lying. Even when there
is no lying or deception involved, spinning often calls into question one’s
intellectual honesty. Often we spin evidence to protect our cherished
beliefs when that evidence ought to cause us to question those beliefs.
ACKNOWLEDGMENTS
I am greatly indebted to many friends and colleagues for helpful criticisms and
suggestions about numerous earlier versions of this chapter. Thanks to Joe Men-
dola, Bruce Russell, Bill Tolhurst, Mike Gorr, Kent Machina, Tom Beauchamp,
George Brenkert, Al Mele, Clancy Martin, J. D. Trout, Mark Chakoian, and Nora
Carson. This article is a shortened version of chapters 1 and 2 of my book Lying
and Deception: Theory and Practice (forthcoming). The first section was published
as “The Definition of Lying,” Nous 40 (April 2005), 284–306. Parts of section 2
(paragraphs 1 and 4–8) are included in my article “Deception and Information
Disclosure in Business and Professional Ethics,” in The Oxford Handbook of Busi-
ness Ethics, ed. Tom Beauchamp and George Brenkert (Oxford: Oxford University
Press, forthcoming).
10
Deception and the Nature of Truth
Michael P. Lynch
Philosophers who worry about the nature of truth rarely worry about
the nature of deception. This is partly due to a sensible division of
labor. When philosophers ask “What is truth?” they take themselves
to be asking about a particular property of our beliefs. Defining that
property—or as is fashionable, claiming that it can’t be defined—is a
matter of metaphysics and the philosophy of language. To talk about
deception or lying on the other hand is to talk about the messy realm
of human interaction. And that is the province of the ethicist. The
general assumption seems to be that the philosophy of deception is
to the philosophy of truth as bioethics is to the philosophy of biology.
Nice stuff to know, undoubtedly important, but really not quite to the
point.
To the layperson, this is apt to seem completely backward. Most
folks, if they think about truth at all, think of deception first and truth
second. Deception, after all, is a real human universal; it knows no
boundaries. Most people are interested in what philosophers say about
truth because they are fascinated by deception, not the other way
around.
In this essay, I will try to say something about both deception and
truth, with an eye toward vindicating the layperson’s sense of what is
important. I think that our attitudes toward deception tell us something
important about both the nature and value of truth. These reflections in
turn underline a more general lesson: that truth is a concept best under-
stood in terms of the role it plays in our overall cognitive life.
This chapter is organized as follows. In part 1, I lay out what I take
deception to be, and argue that, unlike lying, it is conceptually linked
to truth. In part 2, I give some reasons for thinking that certain theories
of truth are implausible in virtue of what they imply about deception.
Since we arguably have a better grip on deception than on truth, this is
bad news for such theories. In part 3, I reflect on what our understanding
of deception tells us about how and why we value truth. I conclude by
drawing out a general lesson from these reflections about the possibility
of giving a meaningful theory of truth.
188
1. DECEPTION
First there was the Word. And then there was the Lie. Lying is a public act,
an act of speech. Deception, on the other hand, runs the behavioral gamut.
As everyone knows, you can deceive without lying, indeed, without even
speaking: sleight of hand, emotional misdirection, or simply pointing in
the wrong direction (“He went that way”) will do just as well. This line of
thought encourages us to think of lying as a form of deception.
This is understandable, but mistaken. It is understandable since one
can’t define lying without appealing to deception. Lying isn’t just saying
what one doesn’t believe. Actors do that in the course of their profession,
and they aren’t lying, they are acting. To lie is to say what one doesn’t
believe with the intention of deceiving.1 Lying and deception are concep-
tually linked in this way, but in fact lying is not a form of deception. For
one can also lie but not deceive. Lying requires an intention to deceive,
but what is intended may not happen: the listener may not be deceived.
This might happen in two ways. First you might be skeptical and not
believe what the liar says. Second, you might believe what he says but he
inadvertently says what is true. In either case you were lied to, but you
were not deceived.
It is this second case that is most important. Supposed I believe that
the butler killed Col. Mustard in the library with the candlestick. Being
a friend of the butler, I tell you what I think is false—that the butler was
sunning himself in Hawaii at the time. You believe what I say, and take
me to be sincere. But in fact I am mistaken, and the butler was in Hawaii,
as you already know. I have lied: I’ve told you what I thought was false in
order to mislead you. But I didn’t succeed. I wanted to deceive you about
the butler’s whereabouts at the time of the foul deed in question, but you
were not deceived about his whereabouts.
Of course, this is consistent with my deceiving you about something
else. Since you believed that I was sincere, then I have deceived you about
the fact that I wanted to deceive you. I’ve concealed my intentions, as
we say. Indeed, any time I lie to you, and you don’t detect the lie, I have
deceived you about my lying to you. But this is an independent matter
from whether I deceive you about the subject matter of the lie.
So it is wrong to say that lying is a form or kind of deception. One can
lie without deceiving and deceive without lying. We might say that a suc-
cessful lie is a kind of deception. But even here we must be careful. If by
“successful lie” we mean an act that succeeds at being a lie, then as just
1. For a defense of this view of lying, see Michael P. Lynch, True to Life (Cambridge,
Mass.: MIT Press, 2004). Compare Donald Davidson, “Deception and Division,” in The
Multiple Self, ed. J. Elster (Cambridge: Cambridge University Press, 1987), 79–93. A
standard conception is Sissela Bok, Lying: Moral Choice in Public and Private Life (New
York: Vintage, 1979); see also Bernard Williams’s treatment in his Truth and Truthfulness
(Princeton: Princeton University Press, 2002), 102–4.
Deception and the Nature of Truth 189
indicated, lies, successful or not, don’t always lead to deception. But if, as
we probably should, we take “successful lie” to mean a lie that succeeds
at doing what the liar intends, then successful lies are a type of deception.
But not all lies, thankfully, are successful.
One way of putting the point I’m making here is that “deceive” is a
success term. I deceive you only insofar as I actually succeed at misleading
you or directing you away from the truth. Moreover, this misleading must
be willful or nonaccidental. This is because those who give us wrong infor-
mation by accident are not deceitful, but simply in error. In the admit-
tedly technical way I will be using the term here, a willful action is one
that is the result of a motivating reason, whether or not one is conscious of
that reason. I can willfully mislead you without doing so with conscious
intention: I may deceive you about Lancelot’s devotion to you because
I desire you for myself, even if I am unaware of my true motivations.2
Likewise, I might deceive myself about your feelings by causing myself to
turn away from the evidence simply because of my unconscious desire to
want your feelings to be other than what they are.
With “willfulness” understood in this very broad way, we can suggest a
particular constraint on deception:
X deceives Y only if X willfully causes Y to have a false belief.
But this is not quite right. I can deceive you even if I don’t cause you
to believe something false.3
Consider the shell game. The con man presents three shells, one of
which has a penny underneath. He moves the shells around and asks you
to pick the shell with the penny. If done right, it looks easy, but isn’t. The
reason is that he distracts you (usually with subtle hand movements) so
that you fail to track the right shell. This causes you to fail to know where
the penny is. But one can lack knowledge without having a false belief.
One can be simply confused, and that is typically the case with such
tricks. You don’t know what to think, and so simply guess. If so, then the
con has succeeded—you’ve been deceived—because then the odds are in
the con man’s favor. This suggests that one can be deceived not only be
believing what is false but by not believing what is true. That is:
X deceives Y only if X willfully causes Y to fail to believe what is true.
But even this isn’t quite right. Can’t I deceive you into believing the
truth? Suppose you believe falsely that Guinevere hates you when in fact
she is madly in love with you. Pretending to be Guinevere, I write you a
2. For more on nonintentional accounts of deception, see A. Mele, Self-deception

Unmasked (Princeton: Princeton University Press, 2002), and M. Johnston, “Self-deception
and the Nature of Mind,” in Perspectives on Self-Deception, ed. B. McLaughlin and A. O.
Rorty (Berkeley: University of California Press, 1988); and J. Bermúdez, “Self-deception,
Intentions, and Contradictory Beliefs,” Analysis 60, 4 (2000), 309–19.
3. Thanks to Joel Kupperman for helping me to see this point.
love note, you change your mind, and everyone lives happily ever after.
I’ve deceived you but I’ve also caused you to believe what is true; but
what I’ve deceived you about is not the truth I’ve caused you to believe.
I’ve deceived you about who wrote the letter, but I’ve not deceived you
about Guinevere’s devotion. Nonetheless, this teaches us that deception
is always deception with regard to something:
X deceives Y with regard to f only if X willfully causes y to fail to believe
what is true with regard to f.
For my purposes, we need only to treat this as stating a necessary con-

dition. But I think there are good reasons to go further and treat the “only
if” as a “if and only if.” If we do, then we can say that to deceive is to
prevent someone from grasping what is true about something. To put it
differently, I deceive another when I willfully cause them to be in a state of
ignorance or error; I deceive myself when I do the same to myself. And this in
turn suggests a sway of saying what deception in general is: if we say for
simplicity’s sake that error is a form of ignorance, we can say that decep-
tion is willful ignorance.
2. TRUTH AND VULNERABILITY

That deception is willful ignorance tells us that when we are deceived,
we lack a belief with a particular property—truth—about some matter
before us. And that, in turn, suggests that our understanding of the nature
of truth and deception are apt to be intertwined. A particularly stark way
of illustrating this point is the simple consensus view of truth:
S’s belief that p is true if and only if S’s belief that p is accepted within S’s
community.
This is the sort of position you end up with if you define truth in terms
of warranted assertibility or belief, and then adopt Richard Rorty’s posi-
tion that “warrant as a sociological matter, to be ascertained by observing
the reception of S’s statement by her peers.” 4 As Rorty once infamously
put it, truth then becomes a matter of what your peers let you get away
with.
One—among many—reasons to reject the simple consensus view is
that it doesn’t jibe with what we know about our ability to be deceived.
If my earlier reflections on deception were correct, you are deceived only
when someone willfully causes you to not believe what is true. This sug-
gests that on the foregoing theory of truth, there are at least two ways
you might deceive me: by causing me to lack a belief that is otherwise
accepted in my community, or by getting the community to accept some-
thing I don’t. Either way, you willingly cause me to not believe what is
4. “Putnam and the Relativistic Menace,” Journal of Philosophy 90 (1993), 450.

true. But that seems off, to say the least. Suppose I don’t believe you are
honest but you wish to deceive me into thinking you actually are. If by
use of a clever advertising campaign you convince everyone else in my
community that you are honest, peer pressure may eventually cause me
to doubt my previous belief; but if I am stubborn and don’t change my
mind, the mere fact that you have convinced others won’t make you honest,
or my assessment of your character mistaken. You can’t deceive me by
convincing someone else that something is the case.
If this weren’t bad enough, the simple consensus theory entails that
it is impossible to deceive the entire community. According to the view,
if what is accepted within my community is p, then it is true that p.
Thus, the community cannot be caused to be in ignorance or error about
what they accept, for it is by definition true. But of course this is non-
sense. Entire communities can and have been deceived about all sorts of
things.
Most folks—we hope!—don’t take the simple consensus view seriously.
But the foregoing objections are worth noting because they illustrate how
a theory of truth can go awry simply by way of what it implies about the
possibilities for deception. Since we have a better grip on deception than
on truth, a theory of truth that gets deception wrong is, to that degree,
implausible.
The foregoing points also suggest another lesson: that there is a con-
ceptual linkage between how objective an account takes truth to be and
the extent to which the account makes us vulnerable to deception. The
simple reason for this is that the twin hallmarks of objectivity are igno-
rance and error. What we believe may not be true, and what is true we may
not believe. The more room an account leaves for ignorance and error, the
more objective it intuitively counts as being. And since deception is the
willful causing of ignorance and error, this suggests that the more vulner-
able a theory of truth leaves us to deception, the more objective it is. But
the linkage also goes the other way as well: the more objective a theory of
truth, the more vulnerable it leaves us to deception.
To see this second point, consider radical deception. We are vulnerable
to being radically deceived about some matter just when we are vulner-
able to being caused to be in perpetual and undiscoverable ignorance or
error about it. Consider a traditional correspondence theory according
to which truth consists in correspondence with mind-independent fact.
Such theories are nonepistemic: whether a belief corresponds to the facts
does not depend in any way on whether we believe, justifiably or not,
that it does. Humans do seem vulnerable to radical deception if truth is
radically nonepistemic. For if truth has nothing to do with the epistemic
status of our beliefs, then we might be deceived in trusting even our best
theories of the world. We may be deceived by the Cartesian demon, or
be brains in vats. Hence, it may seem that we can’t be sure that we really
know what we think we know. Thus the usual complaint against cor-
respondence, nonepistemic theories of truth: they make us vulnerable to
skepticism because they make us vulnerable to the possibility of radical

deception.5
Antirealist theories of truth have been traditionally motivated by this
complaint. Thus, suppose we hold a Peircian view like:
(P) is true if and only if would be accepted at the end of inquiry.
Radical deception of the Cartesian variety is ruled out by fiat by such

a view. For while we might be deceived about what will or will not be
accepted at the end of inquiry, what we do in fact accept at the end
of inquiry must be true. Even the evil demon cannot deceive us at the
hypothetical limit of science, because truth is defined as what we believe
when we reach that limit. What we believe at the ideal limit can’t be
wrong. This means that there is less space for ignorance and error on
this account. Intuitively, therefore, we might say that the Peircian view
is more objective than the simple consensus view, but less objective than
the correspondence view canvassed earlier. And one reason for this is that
it allows greater room for the possibility of deception, although it rules
out radical deception. And this in turn suggests the other direction of the
aforementioned conceptual linkage. Not only is it the case that the more
vulnerable a theory of truth leaves us to deception, the more objective it
is, the less vulnerable to deception an account leaves us, the less objective
we will take it to be.
The point holds even for antirealist views of truth that are not moti-
vated just by fear of skepticism. Consider, for example, a theory of truth
built out of a notion like superwarrant:6
Superwarrant: is superwarranted if and only if the belief that p is
warranted at some stage of inquiry and would remain warranted at every
successive stage of inquiry.
Here a “stage of inquiry,” as the name suggests, is a state of warranted

information or evidence available in principle in the actual world to some
open-minded, receptive inquirer. Stages are understood as being extensi-
ble (additional information might always come in) and inclusive (the addi-
tional information is just that—additional; all successive stages of inquiry
include the information warranted at prior stages). Again, superwarrant
does not posit an idealized “end of inquiry.” A superwarranted belief is
one that is warranted by some state of information available to ordinary
inquirers and that, in fact, would never be defeated or undermined by
subsequent increases of information also available to ordinary inquirers.
5. For a similar point linking skepticism and realism, see J. Heil, “Mind and
Knowledge,” in Oxford Handbook on Epistemology, ed. P. Moser (Oxford: Oxford University
Press, 2002), 316–35.
6. See Wright’s Truth and Objectivity (Cambridge, Mass.: Harvard University Press,
1992). His term is “superassertibility.” I use “superwarrant” both because my account of the
notion is a bit different from his own, and because I find it a more perspicuous label.
Moreover, superwarrant is a stable property: if a belief is superwarranted,

then it is superwarranted at any stage of inquiry. Thus we might suggest:
(SW) A proposition is true if and only if it is superwarranted.
This claim is clearly distinct from (P). Nonetheless, it has some of the
same consequences. In particular, it rules out the possibility of radical
deception. For (SW) and our account of superwarrant seemingly under-
write the claim that
If is superwarranted, is knowable.
And consequently, we can deduce that

If is true, then is knowable.
Now given the T-schema, or the principle that

 is true if and only if p,
we can infer that

If p, then is knowable.
This is just to draw out the obvious: that any account of truth like (SW),
which defines truth partly in terms of warrant or a belief’s epistemic sta-
tus, will have to admit that truth is “epistemically constrained.” But now
it is a simple matter to show that any such theory limits the possibilities
for radical deception. For if we accept that if p, it is knowable that p, then
presumably we should also accept the following:
If is knowable, then p is not something about which reflective
humans may remain perpetually and undetectably ignorant.
But if the fact that p is not something about which reflective humans
may remain forever ignorant, then it is not something about which we can
be eternally deceived. We are immune from radical deception because at
some point, to someone, the truth will out, so to speak. In short, (SW)
appears to imply that
If p, then it is impossible to be radically deceived about p.
This point exposes the weakness of (SW) as a theory of truth. For it

seems very likely that there are some truths about which we cannot rule out,
a priori, the possibility that we might be radically deceived about them. We
don’t need evil demons to make this point either. It seems possible that we
might be perpetually and undetectably deceived about some event in the
distant past by certain documents that were willfully created for that pur-
pose. We will never have any means by which to see through the deception,
or even any evidence that it is a deception. If this is possible, then there are at
least some propositions about which it is possible to be radically deceived.
This suggests that superwarrant is not a plausible theory of the nature
of truth. That is, it is not plausible that truth just is superwarrant. But if
we were to stop here, we would overlook and important point. For even
if truth can’t be identified with superwarrant, it may still be plausible that
some propositions are made true by being superwarranted. Indeed, this is
just the sort of possibility that a pluralist theory of truth allows for.7 Such
theories, at a minimum, are committed to the idea that
(ST): Necessarily, for any proposition, if it is true, then it has some
property F such that, necessarily, if a proposition if F, it is true.
What (ST) proposes is that truth is a supervenient property, in that it

strongly covaries with other properties. And (ST) is compatible, clearly, with
two further thoughts: first, that truth does not just covary with these further
properties but is metaphysically dependent in some way on them, and sec-
ond, that which property determines truth can vary. That is, not only:
Necessarily, there are some propositions such that if they are F, then they are T
but also:
It is possible that there are some propositions that are T but not F.
This limited pluralism about the base properties for truth says noth-
ing about the property of truth itself other than it is a single higher-level
property that is asymmetrically dependent on other properties. Intuitively,
these other properties are those that make propositions true. Perhaps for
some sorts of propositions, the property that makes them true is super-
warrant. Indeed, the foregoing reflections help to show us which sorts of
propositions are likely to be candidates: propositions about which it is a
priori the case that we cannot be radically deceived. What sort of proposi-
tions might these be? One suggestion—and here I only put this forward as
a representative suggestion—would be moral propositions. Many writers,
holding quite different normative ethical theories, think moral wrongness
is conceptually tied to responsibility and blameworthiness. “We do not
do call anything wrong,” as Mill writes, “unless we mean to imply that a
person ought to be punished in some way or other for doing; if not by
law, by the opinion of his fellow creatures; if not by opinion, then by the
reproaches of his own conscience.”8 What is wrong is what we can be jus-
tifiably held responsible for doing—what is worthy of blame, in short. If
so, then the following argument seems cogent:
7. For two versions of alethic pluralism see Michael Patrick Lynch, Truth as One and
Many: A Pluralist Manifesto (Oxford: Oxford University Press, 2009); Wright, Truth and
Objectivity and Saving the Differences (Cambridge, Mass.: Harvard University Press, 2004);
and “Truth and Multiple Realizability,” Australasian Journal of Philosophy 82 (2004),
384–408.
8. Mill, Utilitarianism, ch. 5. See also Stephen Darwall, The Second-person Standpoint,
(Cambridge, Mass: Harvard University Press, 2006), 27. Robert Adams, Finite and
Infinite Goods (New York: Oxford University Press, 2002), 233–38; Allan Gibbard, Wise
Choices, Apt Feelings (Oxford: Oxford University Press, 1990), 42; John Skorupski, Ethical
Explorations (Oxford: Oxford University Press, 2002), 142.
If an action is wrong, then it is blameworthy.

If an action is blameworthy then it is knowable that it is wrong.
Therefore, if an action is wrong, then it is knowable that it is wrong.
The argument appears to be valid. The first premise is Mill’s concep-

tual truth. The second premise, too, seems to be a conceptual truth about
blameworthiness. It says that it is a necessary condition of being a blame-
worthy action that it is possible that some ordinary human observer could
at some point recognize it as wrong. My action is blameworthy when it is
possible for me to be held accountable for that action. But I can’t be held
accountable, surely, for an action that no one, including myself, would ever
know that it is one for which I should be censured. Given the first prem-
ise, to think otherwise would simply mean not only that there are unno-
ticed moral wrongs but also that there are unnoticeable moral wrongs.
It might, for example, be wrong that I used more that six words in this
sentence. And that seems absurd. Hence, the conclusion seems warranted,
and so, presumably, would be a parallel conclusion about praiseworthiness
and rightness. If so, then, thanks to the schema that it is true that p if and
only if p, we know that there are no unknowable truths about what is
morally right or wrong.
Similar reasoning shows that radical deception about moral wrongness
is deeply implausible. Such deception would be possible if it were pos-
sible that an evil demon could make us believe that it is permissible that
I write this sentence when in fact it is morally wrong for me to do so. But
if no one is ever, even if in principle, able to detect the demon’s deceit,
then on one could ever know that my action in writing that sentence was
morally wrong. But if so, that is, if no one is ever able to know it is wrong,
then by the foregoing argument, it isn’t wrong.
Of course, the fact that I can’t be radically deceived about the moral
truth doesn’t mean that I can’t be substantially deceived. The fact that
something is unknowable in principle is consistent with its being the case
that I and everyone on the face of the earth are deceived about it right
now. Radical deceit may be ruled out for morality, but global deceit is not.
Nonetheless, this fact about moral truths makes them noticeably different
from truths about the natural world, where we are quite willing to accept
unknowable truth and radical deception to boot.9
These reflections help to illustrate not only the conceptual linkage
between deception and the nature of truth, but the general lesson I drew
9. Obviously the foregoing considerations aren’t, all by themselves, intended to prove

that moral judgments are made true by being superwarranted, or even that moral truth
is epistemically constrained. Extreme utilitarians might resist the argument by appealing
to the possibility of incalculable utility functions. Others might acknowledge its force but
insist that it is a sure fact about the world and our cognitive capacities that we happen
to be able to recognize the moral facts when they obtain. Lucky devils that we are, we
are just built to be able to discover such facts. For my part, I see no reason to think the
world is so cooperative. I take it to be more plausible that the foregoing argument tells us
something about how those beliefs are made true.
out of that linkage earlier. Where truth is maximally objective, we are vul-
nerable to radical deception. It is plausible that we are vulnerable to such
deception, and therefore theories of truth that artificially limit the nearly
limitless ways humans might be in ignorance or error, hence the ways they
may be deceived, should be rejected. But where the possibility for radical
deception is in fact already limited, such theories may in fact be plausible
accounts of what makes some propositions true.
3. DECEPTION AND THE VALUE OF TRUTH
In the previous section, I reviewed some of the ways the conceptual link-
age between deception and truth can, and should, affect how we think
about the metaphysics of truth. I now turn to question of how that link-
age affects how we think about the value of truth. Since I’ve addressed
this issue at length elsewhere, I’ll be brief.10
People don’t like to be deceived—by their mechanic, their boss, their
friend, their lover, anyone. One reason for our dislike is obvious: when
you are deceived, someone has caused you to not believe what is true, or
even to believe what is false. Either way, they’ve willfully placed them-
selves between you and the facts. And that can be a dangerous thing. But
even if it is not overtly dangerous, it is likely to interfere with one’s plans,
to change them, or to simply make them go awry. Either way, we are
less likely to get what we want. And that of course is typically the point:
the deceiver deceives to get his way, not (generally speaking) to facilitate
yours. There are exceptions of course, but generally speaking, we hate to
be deceived because deception has negative practical consequences.
But our dislike of deception goes deeper than that. One reason to think
so is that most folks would prefer not to be deceived even if it would make
no difference to one’s experiences. Suppose you had to choose between
two doors. Once you make the choice of which door to enter, you will
forget that you ever made a choice at all. Behind door number one is your
life just as it is now. Your friends are friendly, and your lover loves you.
Behind door number two is a very similar life, with one very important
exception: here some of your friends and your lover really despise you.
But you will never discover that fact: their deceit will be perfect. From
the inside, both lives will be indistinguishable: where the first causes you
joy, the second does also; where the first causes you pain, the second does
as well; and so on to the grave. Yet in the second, your life is the life of the
fool: you are deceived.
Forced to choose, almost all of us will prefer the first life over the sec-
ond. Perhaps some may be ambivalent; they’ll flip a coin. Presumably, no
one will actively prefer the second over the first. Either way, your reac-
tion tells you something about how deeply you dislike deception. If you
10. See True to Life, and “Replies to Critics,” Philosophical Books 46 (2005), 331–42.
are ambivalent, then deception matters less to you than it does to others.
What matters is how you feel and experience life; if the truth of your
beliefs has no effect on those feelings and experiences, then you don’t care
whether you are being deceived or not. One door is as good as another.
Most of us, however, will find this attitude odd, even repugnant. We don’t
just want to seem to have friends and lovers, we want friends and lovers,
even if there were no discernible difference between the one case and the
other. Moreover, we want to want to be that way: we care about not being
deceived. We would no more wish to be ambivalent about which door to
enter than we would wish to willingly enter into a deception.
Our attitude toward such choices also tells us something about our
attitudes toward truth. Given the conceptual connection between truth
and deception, this is not surprising. The fact that we prefer not to be
deceived—even when the deception is undetectable—suggests that our
preference for believing whatever is true over not doing so remains even
when it would have no effect on how we experience life. And this in turn
suggests that it is a basic preference—not derived from a preference for
something else. Moreover, it is not a mere preference—I don’t just want
to believe whatever is true; I care about doing so. And if I care about not
being deceived for more than instrumental reasons, I care about believing
what is true for more than instrumental reasons. I care about it “for its
own sake.”11
It is worth emphasizing that this line of reasoning is not intended to
show, absurdly, that we want all of our actual beliefs to be true. I believe
many propositions that I don’t want to be true. Beliefs about the future of
global warming or the continuing spread of AIDS in Africa number among
them. But the fact that I don’t want these particular propositions to be
true is entirely consistent with it being the case that I care about believing
what is true and only what is true, whatever that turns out to be.
Nor does our disvaluing of deception and consequent valuing of true
belief mean that truth is our only or ultimate value. It obviously is not.
Sometimes other things matter more than truth. Thus, more of us would
be willing to be deceived, or to deceive ourselves, if we thought that more
good than bad would come of it overall, or that the matter was so trivial
that the point was essentially moot. But this fact is entirely consistent
with the fact that considered by itself, deception is still something we
deeply wish to avoid, and believing what is true is something we care
about achieving. What this shows is that, like almost everything else we
care about, true belief is a pro tanto value. It is something we care about
other things being equal.
11. What we care about is the state of affairs of believing what is true. This value or
end is to be distinguished from the fact that it is correct to believe a true proposition—or
that the standard of correctness for belief is truth. See Lynch, True to Life, and “The Truth
of Values and the Values of Truth,” in Epistemic Value, eds. A. Haddock, A. Millar, and
D. Pritchard (Oxford: Oxford University Press, forthcoming).
What we care about we typically believe is worth caring about. Thus,

what these reflections do show is that we (or most of us) believe that
truth is worth caring about, and caring about for its own sake. 12 And that
in turn tells us that truth is a value: the state of affairs of believing what is
true is something we take to be a good.
4. CONCLUSION
Philosophers like me who think about truth for a living are, funnily
enough, often the least likely to believe that there is anything all that
interesting to say about truth. One reason for this is that most of us know
too well the failures of traditional theories like the correspondence theory
or Peirce’s pragmatic view. Such theories are reductive: they try to identify
truth with some single property that all and only true propositions have in
common. But such attempts are beset by counterexamples. Theories that
seem plausible when applied to propositions about the middle-sized dry
goods of everyday life (like certain versions of the correspondence theory)
seem much less plausible when applied to propositions about abstract
entities like numbers or norms. And theories that seem more plausible
when applied to propositions about norms (such as superwarrant) are
much less plausible when applied to propositions about the physical world
around us.13 Partly as a result, many philosophers working on truth today
are attracted to one form or other of deflationism, according to which,
roughly, everything that needs to be said about truth can be gleaned from
our inclination to accept instances of the T-schema.14 From this standpoint,
it is simply a mistake to think with the traditional theories that truth has
any sort of “nature” that is worth explaining. It is a useful concept for sure,
providing a handy semantic ladder by which we can ascend and generalize
over infinite strings of propositions, but it is not to be confused with a
property that needs deep metaphysical investigation. In my view, the
considerations raised in this article suggest that there may be more to say
about truth than jaundiced deflationists believe, even if it isn’t the sort of
thing traditionalists look for. What the foregoing thoughts suggest is that
truth is a concept that is intimately related to a host of other concepts—
deception, ignorance, objectivity, value. Truth seems in fact to sit in a
network of such interrelations. Consequently,
12. Of course, this is distinct from showing that truth actually is worth caring about.
For arguments to that effect, see ch. 8 of True to Life.
13. For further arguments to this effect, see Lynch, “Truth and Multiple Realizability.”
14. Representative deflationists include P. Horwich, Truth, 2nd ed. (Oxford: Oxford
University Press, 1998), and H. Field, Truth and Absence of Fact (Oxford: Oxford University
Press, 2001). For some remarks that overlap with the foregoing, see Donald Davidson,
“The Folly of Defining Truth,” in The Nature of Truth, ed. M. P. Lynch (Cambridge, Mass.:
MIT Press, 2001).
there is reason to think that revealing these interrelations will shed light
on the concept of truth, by illuminating what can be described as the
functional role truth plays within our overall cognitive economy. Inves-
tigating truth’s role will most likely not aid us in a reductive analysis of
truth. It will not reveal the secret essence of truth. But it will tell us more
about what truth does for us—how it functions in our thought—not just
in logic or in epistemology, but in the broader realm of messy human
interactions, the realm that the layperson lives in, the realm we live in
ourselves.
ACKNOWLEDGMENTS
Thanks to the following individuals for helpful discussion: Paul Bloomfield, David
Capps, Daniel Massey, Joel Kupperman, and Terry Berthelot.
11
The Truth about Kant on Lies
James Edwin Mahon
As one philosopher has remarked, “one cannot help noticing the heat
with which [Kant] treats the question of lying.”1 To this it must added,
one cannot help noticing the heat with which other philosophers treat the
question of Kant on lying. More than any other element of his moral phi-
losophy, Kant’s writings on lies have elicited an unprecedented amount of
abuse. One philosopher has accused Kant of being “hysterical” in his ethi-
cal writings on lies. Another has accused Kant of “repellent fanaticism” 2 in
his essay On a Supposed Right to Lie from Philanthropy from 1797, where
Kant argues that it would be a crime to lie to a would-be murderer even
to save an innocent life.3 Those scholars who have argued that this con-
clusion in his late work is not entailed by his philosophy of right have
seen fit to accuse him of “bad temper in his old age.”4 This is surely a
case of philosophers shedding more heat than light on a subject, however.
A thorough examination of Kant’s writings on lies—both in the ethics and
in the philosophy of right—is required before accusing him of hysteria,
fanaticism, or bad temper. The importance of this examination can hardly
be underestimated, since it is still true that “some philosophical textbooks
give such prominence to the Kantian view of the lie that the rest of Kant’s
1. Nathan Rotenstretch, “On Lying,” Revue Internationale de Philosophie 10 (1956), 420.

2. Bernard Williams, Truth and Truthfulness: An Essay in Genealogy (Princeton:
Princeton University Press, 2002), 106.
3. W. I. Matson, “Kant as Casuist,” Journal of Philosophy 51 (1954), 859. An eclectic
assortment of critics of Kant’s conclusion in the Right to Lie can be found in Heimo
Hofmeister, “The Ethical Problem of the Lie in Kant,” Kant-Studien 63 (1972), 353–68.
4. H. J. Paton, “An Alleged Right to Lie: A Problem in Kantian Ethics,” Kant-Studien 45
(1954), 201. Many commentators express sympathy with Paton’s verdict, if not outright
agreement. Roger Sullivan, for example, says that “Perhaps there is some merit after all in
the accusation that here he was being cantankerous!” Immanuel Kant’s Moral Theory (New
York: Cambridge University Press, 1989), 350 n. 24. Sally Sedgwick also comments: “More
than any other passage in the essay, I admit that this one makes Paton’s appeal to Kant’s
‘bad temper’ and ‘old age’ extremely seductive”; “On Lying and the Role of Content in
Kant’s Ethics,” Kant-Studien 82 (1991), 61.
201
ethics is either overshadowed and ignored, or summarily dismissed as

inseparable from this obviously impossible notion.” 5 This essay will con-
tribute to that examination by arguing that there are three senses of a lie
to be found in Kant’s moral philosophy, and three corresponding duties
not to lie.6
5. Hofmeister, “The Ethical Problem of the Lie in Kant,” 353. Although the following
instance of a rejection Kant’s entire ethics (or a significant part thereof) due to his
prohibition on lying is an old one, nevertheless it is instructive: “The notion of the test
of universalisation as a practical criterion has been unanimously rejected by the critics,
and doubtless with good reason. The arguments against it are probably familiar to every
student in the elementary stages of moral philosophy. We have all been introduced very
early to the figure of the innocent man pursued by murderers whose life can be saved by
a timely lie. There is no need to work over this well-trodden ground again”; G. C. Field,
“Kant’s First Moral Principle,” Mind 41 (1932), 19. It is interesting to note that a cheap
edition of a translation of the Groundwork aimed at undergraduates is now accompanied
by a translation of the Right to Lie as a supplementary text, giving the Right to Lie great
prominence: Grounding for the Metaphysics of Morals, with On a Supposed Right to Lie Because
of Philanthropic Concerns, trans. and ed. James W. Ellington (Indianapolis: Hackett, 1993).
6. References to Kant’s works in this chapter are given parenthetically, using the
abbreviations of individual works listed here. First are cited the volume and page
number in the standard edition of Kant’s works, Kants gesammelte Schriften, ed. Königlich
Preußischen Akademie der Wissenschaften [subsequently Deutsche, now Berlin-
Brandenburg Akademie der Wissenschaften], original ed. Wilhelm Dilthey (Berlin: Reimer
[subsequently de Gruyter] 1900–), followed by the page number in the translation listed
here. Unless otherwise indicated, all emphases are in the original.
A: Anthropology from a Pragmatic Point of View [Anthropologie in pragmatischer Hinsicht]
(1798), trans. Mary J. Gregor (The Hague: Nijhoff, 1974).
AN: Announcement of the Near Conclusion of a Treaty for Eternal Peace in Philosophy
[Verkündigung des nahen Abschlusses eines Traktes zum ewigen Frieden in der
Philosophie] (1796), trans. Peter Fenves, in Raising the Tone of Philosophy: Late Essays
by Immanuel Kant, Transformative Critique by Jacques Derrida, ed. Peter Fenves
(Baltimore: Johns Hopkins University Press, 1993), 83–100.
C: Correspondence [Kant’s Briefwechsel], trans. and ed. Arnulf Zweig (Cambridge:
Cambridge University Press, 1999).
CPR: Critique of Practical Reason [Kritik der praktischen Vernunft] (1788), trans. Mary
J. Gregor, in Practical Philosophy, trans. and ed. Mary J. Gregor and Allen W. Wood
(Cambridge: Cambridge University Press, 1996), 133–271.
G: Groundwork of the Metaphysics of Morals [Grundlegung zur Metaphysik der Sitten]
(1785), trans. Mary J. Gregor, in Practical Philosophy, trans. and ed. Mary J. Gregor and
Allen W. Wood (Cambridge: Cambridge University Press, 1996), 37–108.
LE: Lectures on Ethics [Vorlesungen über Ethik] (1924), trans. Peter Heath and ed. Peter
Heath and J. B. Schneewind (Cambridge: Cambridge University Press, 1997).
MM: The Metaphysics of Morals [Die Metaphysik der Sitten] (1797), trans. Mary J. Gregor,
in Practical Philosophy, trans. and ed. Mary J. Gregor and Allen W. Wood (Cambridge:
Cambridge University Press, 1996), 353–603.
M: On the Miscarriage of All Philosophical Trials in Theodicy [Über das Misslingen aller phil-
osophischen Versuche in der Theodizee] (1791), trans. George di Giovanni, in Religion and
Rational Theology, ed. and trans. Allen W. Wood and George di Giovanni (Cambridge:
Cambridge University Press, 1996), 24–37.
RL: On a Supposed Right to Lie from Philanthropy [Über ein vermeintes Recht aus Menschen-
liebe zu lügen] (1797), trans. Mary J. Gregor, in Practical Philosophy, trans. and ed. Mary
J. Gregor and Allen W.Wood (Cambridge: Cambridge University Press, 1996), 611–15.
The Truth about Kant on Lies 203
WHAT IS A LIE?
Kant does not give a single answer to the question of what is a lie. Instead,
he tells us what a lie is in the ethical sense, what a lie is in the juristic
sense, and what a lie is in the sense of right. 7 Nevertheless, it is possible to
say what is common to a lie in all three senses, as well as what differenti-
ates a lie in one sense from a lie in the other senses. What is common to
a lie in all three senses is that it is an intentional action of a certain kind.
More specifically, it is the making an untruthful statement with the intention
that that statement be believed to be true. This entails that three conditions
must be met for a lie in all three senses.
The first condition that must be met for a lie in all three senses is that a
statement must be made. A person may be said to make a statement when
a person believes that there is some expression, and some language, such
that one of the standard uses of that expression in that language is that of
expressing some proposition, and the person utters, writes, or signs that
expression with the intention that it be believed that she intended to utter
(etc.) that expression with that standard use.8 Making a statement there-
fore requires the use of conventional signs, as opposed to natural or causal
signs. A person lies “when a person gives signs indicative of thoughts that
he does not have” (LE, 27:700 [p. 426]). Kant sometimes speaks of this
statement condition in terms of words: “communication of one’s thoughts
to someone through words that yet (intentionally) contain the contrary of
what the speaker thinks” (MM, 6:429 [p. 552]). However, conventional
signs are not limited to spoken or written words; they include sign lan-
guage, smoke signals, semaphore signals, Morse code, and so forth.9 They
also include nodding one’s head, winking one’s eye, and shrugging one’s
shoulders.10 In the case of a person who does not utter (etc.) a declara-
tive sentence, but who, for example, curses, or utters an interjection or an
exclamation, or issues a command or an exhortation, or asks a question,
or says “Hello,” then, if it is not one of the standard uses of that expression
in that language to express some proposition, and the person does not
intend that it be believed that she intended to utter (etc.) that expression
7. The claim that there are three senses of a lie in Kant’s moral philosophy is not
original to this essay. John E. Atwell says about Kant that “he condemns lying on three
grounds: (1) as a lying promise, it violates a perfect duty to others, and thus infringes
someone else’s rights; (2) as a lie in general, it violates a strict duty to oneself; and, as we
shall see here, (3) a lie violates the rights of mankind”; Ends and Principles in Kant’s Moral
Thought (Dodrecht: Nijhoff, 1986), 193–94 (references omitted). These three grounds
for the condemnation of lying correspond to the three senses of a lie, namely, a lie in the
juristic sense, a lie in the ethical sense, and a lie in the sense of right.
8. Roderick M. Chisholm and Thomas D. Feehan, “The Intent to Deceive,” Journal of
Philosophy 74 (1977), 150.
9. Sissela Bok, Lying: Moral Choice in Public and Private Life (New York: Random
House, 1978), 13.
10. Chisholm and Feehan, “Intent to Deceive,” 149.
with that standard use, it follows that she is not making a statement when
she does any of these things. If she is not making a statement when she
does any of these things, it follows that she cannot be lying when she does
any of these things, although she may be attempting to deceive.11
Kant gives at least two examples of actions in which nonconventional
signs are used in an attempt to deceive. The first is packing one’s bags—a
natural sign of intending to leave on a journey—in order to have a thief
falsely believe that one is intending to leave, so that one may catch the
thief: “a pretended journey, to uncover and thwart a crime” (LE, 27:699–
700 [p. 426]). The second involves the Scottish speculator John Law,
whose company for investment in Louisiana, Compagnie Perpetuelle des
Indes, went bankrupt. Law continued to build his house in France—a nat-
ural sign of intending to stay—in order to have people falsely believe that
he was not leaving, so that he could flee France. In neither case is a lie told,
since in neither case is a statement made:
if I pack my bags, for example, people will think I am off on a journey, and
that is what I want them to believe; but they have no right to demand any
declaration of will from me. That is what the famous John Law did; he
kept on building, and when everyone was thinking: He’ll never leave, off he
went. (LE, 27:447 [p. 202–3])
Since a lie in all three senses requires that a statement be made, there
can be no so-called lies of omission or concealment lies, according to Kant.
In Anthropology from a Pragmatic Point of View, after stating that “the mere
fact that any prudent man finds it necessary to conceal a good part of his
thoughts makes it clear enough that every member of our race is well
advised to be on his guard and not to reveal himself completely” (A, 7:332
[p. 192]), Kant distinguishes between three different things: dissimulation
(concealment, reticence), that is, intentionally not making a statement,
which may involve no deception; deception, which may involve keeping
silent with the intention that something false be believed to be true, that
is (deception by omission), and hence, that may involve no lying; and
finally, lying:
So it already belongs to the basic composition of a human creature and to
the concept of his species to explore the thoughts of others but to withhold
one’s own—a nice quality that does not fail to progress gradually from dis-
simulation to deception and finally to lying. (A, 7:153 [p. 32])
Kant does insist that lying requires that a statement be made:

This reticence, however, this want of candor—a candor that, taking man-
kind en masse, we cannot expect of people, since everyone fears that to
reveal himself completely would make him despised by others—is still very
11. For the contrary argument that it is possible to lie by using, for example,
interrogatives and imperatives, see Henry S. Leonard, “Interrogatives, Imperatives, Truth,
Falsity and Lies,” Philosophy of Science 26 (1959), 172–86.
different from that lack of sincerity that consists in dishonesty in the actual
expression of our thoughts. (C, 11:332 [p. 411])
Kant believes that it is possible to tell an “internal lie” or “inner lie,”

to lie to one’s “inner judge, who is thought of as another person” (MM,
6:429–30 [p. 552–53]). Since, in the case of an “internal” lie, one lies to
one’s “inner judge,” who is “thought of as another person,” but who is not
another person, an “internal” lie is understood by Kant to be a lie to one-
self. That is, Kant holds that it is possible to make an untruthful statement
to oneself with the intention that one believe it to be true. Although he
says that “to deceive oneself on purpose seems to contain a contradiction”
(MM, 6:430 [p. 553]), he does not believe that it contains a contradiction.
Hence, it is possible to lie, in one sense of lie, namely, the ethical sense
of lie, without this being an external or public action. However, even in
the case of the “internal lie,” one must “profess,” or make a statement, to
one’s “inner judge,” in order to tell such a lie. Hence, even in the case of an
“internal lie,” one must make a statement:
Insincerity is mere lack of conscientiousness, that is, of purity in one’s profes-
sions before one’s inner judge, who is thought of as another person when
conscientiousness is taken quite strictly; then if someone, from self-love,
takes a wish for the deed because he has a really good end in mind, his
inner lie, although it is indeed contrary to his duty to himself, gets the name
of frailty, as when a lover’s wish to find only good qualities in his beloved
blinds him to her obvious faults.—But such insincerity in his declarations,
which a human being perpetrates upon himself, still deserves the strongest
censure. (MM, 6:430 [p. 553–54])
The second condition that must be met for a lie in all three senses is
that the statement that is made must be believed to be false by the per-
son who makes the statement. That is, the statement that is made must
be untruthful. Importantly, it is not required that the statement be false.
The statement that is made may be true. However, if the statement that
is made, whether true or false, is believed to be false by the person who
makes it, then the statement that is made is untruthful:
One cannot always stand by the truth of what one says to oneself or to
another (for one can be mistaken); however, one can and must stand by the
truthfulness of one’s declaration or confession, because one has immediate
consciousness of this. For in the first instance we compare what we say with
the object in a logical judgment (through the understanding), whereas in
the second instance, where we declare what we hold as true, we compare
what we say with the subject (before conscience). Were we to make our
declaration with respect to the former without being conscious of the latter,
then we lie, since we pretend to something else than what we are conscious
of. (M, 8: 267–68 [p. 34])
Since untruthfulness is required for a lie in all three senses, it follows that
it is not a lie, in any of the three senses, if someone makes a truthful, but
false, statement. However, it may be a lie, in all three senses, if someone
makes an untruthful, but true, statement. In particular, it may be a lie, in

all three senses, if someone makes an untruthful, but true, statement with
the intention that it be believed to be true.
In the Right to Lie, Kant gives an example of someone making a truthful,
but false, statement with the intention that it be believed to be true, and
an example of someone making an untruthful, but true, statement with
the intention that it be believed to be true. In the first case, a person falsely
believes that his friend is in his house, and answers “Yes” to a would-be mur-
derer on his doorstep asking if his friend is in his house. In the second case,
the same person, with the same false belief, answers “No.” In the first case,
the person does not lie, in any of the three senses of lie. In the second case,
the person does lie, both in the sense of right and in the ethical sense:
It is still possible that, after you have honestly answered “yes” to the murder-
er’s question as to whether his enemy is at home, the latter has nevertheless
gone out unnoticed, so that he would not meet the murderer and the deed
would not be done; but if you had lied and said that he is not at home, and he
has actually gone out (though you are not aware of it), so that the murderer
encounters him while going away and perpetrates his deed on him, then you
can by right be prosecuted as the author of his death. (RL, 8:427 [p. 612])
Since untruthfulness is required for a lie in all three senses, it follows that
it is not a lie in any of the three senses to make a truthful statement with
the intention that some believed-false statement be believed to true. Kant
gives an example of making a truthful statement with the intention that a
believed-false statement be believed to be true in the course of discussing
intentionally deceptive truthfulness:
A moral casuistic would be very useful, and it would be an undertaking
much to the sharpening of our judgement, if the limits were defined, as
to how far we may be authorized to conceal the truth without detriment
to morality. Along with lying we may include: (a) aequivocatio moralis,
i.e. moral ambiguity, insofar as it is deliberately employed to deceive the
other; for example, a Mennonite swore an oath that he had handed over the
money he owed to his creditor, and in a literal sense he could swear this, for
he had hidden that very sum in a walking stick and asked his adversary to
hold it. (LE, 27:701–2 [p. 428])
In this example the Mennonite makes a believed-true statement, that

he has given the creditor the money he owes, with the intention that
a believed-false statement—that it is not the case that he has given the
creditor the money he owes—be believed to be true. The Mennonite is not
lying in any of the three senses, although he is attempting to deceive.12
12. For more on the distinction between lies and intentional deception, see my “Kant
and Maria von Herbert: Reticence vs. Deception,” Philosophy 81 (2006), 417–44, and “A
Definition of Deceiving,” International Journal of Applied Philosophy 21 (2007), 181–94.
Note that Kant does not consider the case of someone making an untruthful statement
without the intention that that statement be believed to be true, but with the intention
that some other believed false statement be believed to be true.
The third condition that must be met for a lie in all three senses is that
it must be intended that the untruthful statement be believed to be true.
To make an untruthful statement with the intention that it be believed
to be true is to have an intention to deceive about the contents of the
untruthful statement. Hence, an intention to deceive about the contents
of the untruthful statement is necessary for a lie in all three senses. Mak-
ing an untruthful statement without the intention that it be believed to be
true—without an intention to deceive about the contents of the untruth-
ful statement—is not a lie in any of the three senses:
A white lie is often a contradictio in adjecto [contradiction in terms]; like
pretended tipsiness, it is untruth that breaches no obligation, and is thus
properly no lie. Joking lies, if they are not taken to be true, are not immoral.
But if it be that the other is ever meant to believe it, then, even though no
harm is done, it is a lie, since at least there is always deception. (LE, 27:62
[p. 28])
One example that Kant gives of making an untruthful statement that is

not intended to be believed to be true is writing “your obedient servant” at
the end of a letter: “Can an untruth from mere politeness (e.g., the “your
obedient servant” at the end of a letter) be considered a lie? No one is
deceived by it” (MM, 6:431 [p. 554]). Another example that Kant gives
of making an untruthful statement that is not intended to be believed to
be true is bragging or telling a tall tale: “Hence the telling of tall stories,
or braggings in company, demean us, and can only pass as a jest if the
judgment of others about the content of their truth cannot be in doubt”
(LE, 27:700 [p. 427]). Other examples Kant gives of making untruth-
ful statements that are not intended to be believed to be true are giving
compliments, being gallant, and making excessively warm protestations
of friendship:
Courtesy (politesse) is a semblance of graciousness that inspires love. Mani-
festations of deference (compliments) and the whole of courtly gallantry,
along with the warmest verbal protestations of friendship, are not always
the truth (“My dear friends: there is no such thing as a friend.” Aristotle);
but this still does not make them deception, because everyone knows how to
take them, and especially because these tokens of benevolence and respect,
though empty at first, gradually lead to real attitudes of this kind. (A, 7:152
[p. 31])13
Although it is necessary for a lie in all three senses that the untruthful
statement that is made be intended to be believed to be true, it is not neces-
sary for a lie in any of the three senses that it be believed to be true. That
13. Kant may not be consistent on this matter, however. In his lectures he says that
courtiers and politicians are lying in the ethical sense (at the very least): “If untruth
presupposes cleverness and skill, we get artful lying and repute; courtiers and politicians,
for example, have to achieve their aims by lying, and everyone should flee any position in
which untruth is indispensable to him” (LE, 27:62 [p. 28]).
is, although an intention to deceive about the contents of the untruthful

statement is necessary for a lie in all three senses, deception is not neces-
sary for a lie in any of the three senses. This is true for two reasons. First,
an untruthful statement that is made with the intention that it be believed
to be true may be true. However, if an untruthful but true statement is
believed to be true, then deception about its contents does not occur.14
Nevertheless, this may be a lie, in all of the three senses. For example,
untruthfully answering “No” to the would-be murderer’s question about
whether one’s friend is in one’s house, with the intention that he believe it
to be true, when, unbeknownst to one, one’s friend has left one’s house, is
a lie in the ethical sense and in the sense of right, even though the would-
be murderer is not deceived about one’s friend whereabouts if he believes
what one says to be true. Second, an untruthful statement (whether true
or false) that is made with the intention that it be believed to be true may
not be believed to be true—it may be disbelieved. For example, to modify
the foregoing case, if one’s friend is in one’s house, and one untruthfully
answers “No” to the would-be murderer’s question about whether one’s
friend is in one’s house, with the intention that he believe it to be true, then
the would-be murderer may not believe one’s answer to be true. However,
one is lying in this case, in the ethical sense and in the sense of right. Unlike
“deceive,” which is a success or achievement verb, “lie” is not a success or
achievement verb, in any of the three senses. Lying is not a perlocutionary
act, in any of the three senses. Lying does not require a response of any
kind, and in particular, does not require the response of being believed to
be true, in order for it to be lying, in any of the three senses.15
Granted that what is common to a lie in all three senses is that it is the
making an untruthful statement with the intention that that statement be
believed to be true, it is now possible to distinguish between a lie in the ethi-
cal sense, a lie in the juristic sense, and a lie in the sense of right.
A LIE IN THE ETHICAL SENSE: FALSILOQUIUM DOLOSUM
A lie in the ethical sense is extremely broad in scope. It is simply the

making an untruthful statement with the intention that that statement
be believed to be true. As Kant says: “In the ethical sense it compromises
14. It may well be that, in such a case, the person is deceived about what the liar
believes and about what the liar intends. However, it is possible to imagine a case in which
one knows that someone will lie about some matter, but one also knows that the person is
mistaken about this matter, and that the lie will be true, although one does not know what
the lie will be. In such a case, when one believes to be true what the liar says, one is not
being deceived about anything.
15. D. S. Mannison, “Lying and Lies,” Australasian Journal of Philosophy 47 (1969), 135.
Charles Fried, however, calls a “case in which the hearer not only does not believe what
he is being told, but does not even believe that the speaker believes it” an “attempted lie”
rather than a lie; Right and Wrong (Cambridge, Mass.: Harvard University Press, 1978), 59.
every intentional untruth, or every intentionally false [declaration] 16 of

my disposition” (LE, 27:605 [p. 351]). A lie in the ethical sense is, as Kant
would say, a falsiloquium dolosum (intentionally deceptive untruthful
statement): “Lying (in the ethical sense of the word), intentional untruth
as such” (MM, 6:430 [p. 553]).
Importantly, a lie to oneself, an “internal lie,” is a lie in the ethical
sense:
It is possible that not everything a person holds to be true is true (for every-
one can err), but in everything that one says, one must be truthful (one
ought not to deceive); it may be that a confession is merely inward (before
God) or also outward. The transgression of this duty of truthfulness is called
lying, and, for this reason, there can be external lying as well as internal
mendacity; as a result, it can happen that both sorts of mendacity are united
or that they contradict each other. But lying, whether it be inward or out-
ward, is of two kinds: (1) if one states something to be true that one knows
to be untrue; (2) if one states something to be certain that one nevertheless
knows to be subjectively uncertain. (AN, 8:421–42 [p. 93])
A lie to oneself is intentionally harmless to others. Hence, a lie in the

ethical sense may be intentionally harmless to others, since it may be a
lie to oneself. However, actions that are intentionally harmless to others
are not punishable by law. Hence, a lie in the ethical sense may not be
punishable by law, since it may be a lie to oneself. As he says: “In ethics,
though, every falsiloquium, every knowing deception, is impermissible,
even though it be not immediately coupled with an injury, and would
not be imputable coram foro juridico [before a court of law]” (LE, 27:700
(426–27]).
A LIE IN THE JURISTIC SENSE: FALSILOQUIUM DOLOSUM IN

PRAEJUDICIUM ALTERIUS
A lie in the juristic sense is much narrower in scope than a lie in the ethi-
cal sense. It is the making of an untruthful statement to a particular other
person (or particular other persons) with the intention that this person
believe that statement to be true and be harmed by believing that state-
ment to be true: “jurists insist upon adding for their definition, that it must
harm another (mendacium est falsiloquium in [praejudicium]17 alterius [a lie
is an untruthful statement that harms another])” (MM, 8:426 [p. 612]).
16. I have substituted “declaration” for “statement,” since the word used by Kant is
Erklärung, which is best translated as “declaration.”
17. Although it says “praeiudicium” in the text, there is good reason to believe that
there is a misprint in the Latin in Kants gesammelte Schriften here (possibly as a result of a
misprint in the original publication) and that it should be praejudicium.
In one place in his lectures on ethics, Kant calls a lie in the juristic
sense a falsiloquium in praejudicium alterius: “In law a mendacium is afal-
siloquium in praejudicum alterius, and cannot be anything else there” (LE,
27: 448 [p. 203]). In several other places in his lectures, Kant is more
precise, and says that a lie in the juristic sense is a falsiloquium dolosum
in praejudicum alterius, an intentionally deceptive untruthful statement
that harms another: “The jurist recognizes and applies this only insofar
as it involves a violation of the duties towards others (officii juridicorum
[juridical duties]), and he understands thereby a falsiloquium dolosum in
praejudicum alterius; he is therefore looking to the consequences and rela-
tion to others” (LE, 27:604–45 [p. 350–51]).
In sensu juridico [legal/juridical sense] the mendacium is a falsiloquium dolo-
sum in praejudicium alterius, but in sensu ethico [ethical sense] it is already
any deliberate untruth. (LE, 27:701 [p. 427])
In his lectures, Kant explains the distinction between a lie in the ethical
sense and a lie in the juristic sense:
Hence an untruth differs from a lie in this, that both, indeed, contain a
falsiloquium, i.e., a declaration whereby the other is deceived, but the latter
is uttered with an associated intention to injure the other by the untruth.
Hence, too, a lie is subject to judicial reprimand, at least an offence, but not
as an untruth. In ethics, though, every falsiloquium, every knowing decep-
tion, is impermissible, even though it be not immediately coupled with an
injury, and would not be imputable coram foro juridico [before a court of
law]. (LE, 27:700 [p. 426–27])
By an intention to harm a particular other person Kant means an inten-

tion that this person’s rights be violated. A lie in the juristic sense, there-
fore, is the making an untruthful statement to a particular other person (or
particular other persons) with the intention that this person believe that
statement to be true and that this person’s rights be violated by believing
that statement to be true.
In the Doctrine of Right, part 1 of The Metaphysics of Morals, Kant says
that the original right that belongs to every person by virtue of her or
his humanity is the right to freedom. The right to freedom is the right to
act without the interference of others “insofar as it can coexist with the
freedom of every other in accordance with a universal law” (MM, 6:237
[p. 393]). This right to freedom includes the right to act towards others
in such a way that “does not in itself diminish what is theirs.” Making
an untruthful statement to a particular other person with the intention
that this person believe it to be true “does not in itself diminish what is
theirs”:
This principle of innate freedom already involves . . . his being authorized to
do to others anything that does not in itself diminish what is theirs, so long
as they do not want to accept it—such things as merely communicating his
thoughts to them, telling or promising them something, whether what he
says is true or sincere or untrue and insincere (veriloquium aut falsiloquium

[truthful statement or untruthful statement]); for it is entirely up to them
whether they want to believe him or not. (MM, 6:237–38 [p. 393–94])
By “diminish what is theirs” Kant means deprive the particular other

person of what is rightfully hers or his. What is rightfully a person’s
includes the person’s property, and through contract, the promised per-
formance of something by a person (MM, 6:245–87 [p. 401–34]). Making
an untruthful statement to a particular other person with the intention
that this person believe it to be true does not, by itself, entail intending that
this person be deprived of what is rightfully hers or his by believing it to
be true. Not every untruthful statement is such that, if a person believes it
to be true, it deprives this person of what is rightfully hers or his. However,
certain untruthful statements are such that, if a person believes them to
be true, they deprive this person of what is rightfully hers or his. In the
case of such an untruthful statement, making this untruthful statement
to a particular other person with the intention that this person believe it
to be true does entail intending that this person be deprived of what is
rightfully hers or his by believing it to be true:
Telling an untruth intentionally, even though merely frivolously, is usually
called a lie (mendacium) because it can also harm someone, at least to the
extent that if he ingenuously repeats it others ridicule him as gullible. The
only kind of untruth we want to call a lie, in the sense bearing upon rights, is
one that directly infringes upon another’s right, e.g., the false allegation that
a contract has been concluded with someone, made in order to deprive him
of what is his (falsiloquium dolosum). (MM, 6:238 n. 1 [p. 394 n. 1])
Making the untruthful statement to a particular other person, “I have

fulfilled my contract with you,” with the intention that this person believe
it to be true, entails intending that this person be deprived of what is right-
fully his or hers by this person believing it to be true, namely, the promised
fulfillment of the contract. It entails intending that this person’s rights be
violated by believing it to be true. Hence, making the untruthful statement
to a particular other person, “I have fulfilled my contract with you,” with the
intention that this person believe it to be true, is a lie in the juristic sense. It
is a falsiloquium dolosum in praejudicum alterius, and it is punishable by law.
Every lie in the juristic sense is also a lie in the ethical sense. However,
most lies in the ethical sense are not also lies in the juristic sense.
A LIE IN THE SENSE OF RIGHT: FALSILOQUIUM DOLOSUM IN

PRAEJUDICIUM HUMANITATIS
A lie in the sense of right is broader in scope than a lie in the juristic sense,
but narrower in scope than a lie in the ethical sense. It is the making of an
untruthful statement to others with the intention that others believe that
statement to be true and that “humanity generally” be harmed by them

believing that statement to be true:
Thus a lie, defined merely as an intentionally untrue declaration to another,
does not require what jurists insist upon adding for their definition, that
it must harm another (mendacium est falsiloquium in [praejudicium]18 alte-
rius). For it always harms another, even if not another individual, neverthe-
less humanity generally, inasmuch as it makes the source of right unusuable
[sic]. (RL, 8:426 [p. 612])
In his lectures, Kant refers to a lie “from the moral viewpoint,” which
includes right, as a falsiloquium in praejudicium humanitatis, an untruthful
statement that harms humanity:
A mendacium is thus a falsiloquium in praejudicium humanitatis even when
it is not also in violation of any particular jus quaesitum [special right]
of another. In law a mendacium is a falsiloquium in praejudicium alterius,
and cannot be anything else there, but from the moral viewpoint it is a
falsiloquium in praejudicium humanitatis. (LE, 27:448 [p. 203])
By an intention to harm “humanity generally” is meant an intention to

violate a “right of mankind,” or a right of “humanity generally,” namely, the
right to enter into and maintain a society:
It is therefore possible for a falsiloquium to be a mendacium—a lie—though
it contravenes no right of any man in particular. Whoever may have told me
a lie—I do him no wrong if I lie to him in return, but I violate the right of
mankind; for I have acted contrary to the condition, and the means, under
which a society of men can come about, and thus contrary to the right of
humanity. (LE, 27: 448 [p. 203])
It is required, for a society to come about and to continue in existence,

that there be rights based on contracts. To make an untruthful statement
to others with the intention that others believe it to be true is to act in a
way such that “statements (declarations) in general are not believed,” and
hence, that trust in contracts is undermined, and hence, that rights based
on contracts are undermined:
I nevertheless do wrong in the most essential part of duty in general by such
falsification, which can therefore be called a lie (though not in the jurist’s
sense); that is, I bring it about, as far as I can, that statements (declarations)
in general are not believed, and so too that all rights which are based on
contracts come to nothing and lose their force; and this is a wrong inflicted
upon humanity generally. (RL, 8:426 [p. 612])
Making any untruthful statement to others with the intention that oth-
ers believe it to be true entails intending to undermine rights based on
contracts. Therefore, making any untruthful statement to others with
the intention that others believe it to be true entails intending to harm
18. See note 17.

“humanity generally.” Hence, any untruthful statement made to others

with the intention that others believe it to be true is a lie in the sense of
right, or a falsiloquium in praejudicium humanitatis—or better, a falsilo-
quium dolosum in praejudicium humanitatis.
Every lie in the juristic sense is a lie in the sense of right. However, not
every lie in the sense of right is a lie in the juristic sense. In the case of
some untruthful statements made to others with the intention that others
believe them to be true, there is not an intention that particular other per-
sons be harmed by believing them to be true, since there is not an inten-
tion that particular other persons believe them to be true. Nevertheless, in
these cases, since there is an intention that others believe them to be true,
there is an intention that “humanity generally” be harmed. These lies are
lies in the sense of right, but they are not lies in the juristic sense:
The question arises, whether a lie that affects nobody’s interests, and does
nobody any harm, is likewise a lie? It is, for I promise to speak my mind, and
if I fail to speak it truly, I do not, indeed, act in praejudicium of the particular
individual concerned, but I do so act in regard to humanity. (LE, 27:449
[p. 204])
Kant gives several examples of lies in the sense of right that are not lies
in the juristic sense. One is the lie to the would-be murderer that one’s
friend is not in one’s house. This is a lie in the sense of right, but it is not
a lie in the juristic sense. Since the would-be murderer has forfeited his
rights by setting out on a course of murder, one cannot intend to violate
his rights by making the untruthful statement “No” to him with the inten-
tion that he believe it to be true. One is, however, intending to violate the
right of humanity:
I indeed do no wrong to him who unjustly compels me to make the state-
ment if I falsify it, I nevertheless do wrong in the most essential part of duty
in general by such falsification, which can therefore be called a lie (though
not in the jurist’s sense). (RL, 8:426 [p. 612])
Another example of a lie in the sense of right that is not a lie in the
juristic sense is the lie to someone who has lied to one. Since the liar has
forfeited his rights by his lie, one cannot intend to violate his rights by
making an untruthful statement to him with the intention that he believe
it to be true. One is only intending to violate the right of humanity: “Who-
ever may have told me a lie—I do him no wrong if I lie to him in return,
but I violate the right of mankind” (LE, 27:447 [p. 203]).
Another example of a lie in the sense of right that is not a lie in the
juristic sense is that of publishing an untruthful statement about an event
with the intention that it be believed to be true by others. Here there is
no particular other person to whom the untruthful statement is made
with the intention that it be believed to be true, and hence, no intention
to violate the rights of a particular other person. However, it is intended
that the right of humanity be violated:
If a man publishes a false report, he thereby does no wrong to anyone in par-

ticular, but offends against mankind, for if that were to become general, the
human craving for knowledge would be thwarted; apart from speculation,
I have only two ways of enlarging my store of information: by experience,
and by testimony. But now since I cannot experience everything myself, if
the reports of others were to be false tidings, the desire for knowledge could
not be satisfied. (LE, 27:447–48 [p. 203])
Every lie in the sense of right, every falsiloquium dolosum in praejudi-

cium humanitatis, is punishable by law. This means that any untruthful
statement made to others with the intention that it be believed to be
true, even if it is not made with the intention to violate the rights of a par-
ticular other person—either because it is not made to a particular other
person, or because the particular other person to whom it is made has
forfeited her or his rights—is punishable by law. As Kant says about the
lie in the sense of right to the would-be murderer:
Such a well-meant lie . . . can be condemned as wrong even in accordance with
external laws. That is to say, if you have by a lie prevented someone just now
bent on murder from committing the deed, then you are legally accountable
for all the consequences that might arise from it. (RL, 8:427 [p. 612])
As another commentator has said about this example: “But suppose, what
is more likely to occur, that the house-owner lies to the would-be mur-
derer, the innocent friend is saved, and soon thereafter the police appre-
hend the intruder. Everything turns out well—except, according to Kant,
the house-owner may be charged with violating the juridical duty not
to lie.”19 In order not to confuse the second and third senses of lie here,
however, it is preferable to refer to the “juridical duty not to lie” as the
duty of right not to lie.
While every lie in the sense of right is also a lie in the ethical sense, not
every lie in the ethical sense is also a lie in the sense of right. In particular,
lies to oneself are not lies in the sense of right.
With a lie in the ethical sense, a lie in the juristic sense, and a lie in
the sense of right explained, it is possible to look at the different duties
not to lie in the Groundwork of the Metaphysics of Morals, the Critique of
Practical Reason, The Metaphysics of Morals, and On a Supposed Right to
Lie from Philanthropy.
THE DUTY NOT TO LIE IN THE GROUNDWORK

AND THE CRITIQUE OF PRACTICAL REASON
It is a matter of some controversy as to which duty not to lie is at issue in

the Groundwork of the Metaphysics of Morals. In the preface, Kant gives
as an example of a moral law “‘thou shalt not lie’ ” (G, 4:389 [p. 44–45]).
19. Atwell, Ends and Principles in Kant’s Moral Thought, 200.

However, he does not explain which sense of lie he has in mind, and
hence, which duty not to lie he has in mind.
In Section I he asks “may I, when hard pressed, make a promise with
the intention not to keep it?” (G, 4:402 [p. 57]) As later versions of the
same example make clear, by “hard pressed” he means hard pressed for
money, and by “a promise with the intention not to keep it” he means a
lying promise to get money. Kant argues that “I could indeed will the lie,
but by no means a universal law to lie” (G, 4:403 [p. 57]), and concludes
that one must “be truthful from duty” (G, 4:402 [p. 57]). Granted that
a duty to “be truthful” is a duty not to lie (as opposed to a duty to be
candid),20 his conclusion is that there is a duty not to lie. However, once
again, he does not say which sense of a lie he has in mind, and hence,
which duty not to lie he has in mind. The example of a lie that he gives—a
lie to get money—is an untruthful statement made to a particular other
person with the intention that this person believe it to be true and be
harmed by believing it to be true (by depriving him of what is rightfully
his, namely, his money). It is a lie in all three senses of a lie. Hence, the
duty not to lie in question could be a broad ethical duty, a narrower duty
of right, or an even narrower juristic duty.
In section 2, Kant gives examples of “duties to ourselves and to other
human beings” that are “perfect and imperfect” (G, 4:421 [p. 73]). Here
he argues that a person who acts in accordance with the maxim “when
I believe myself to be in need of money I shall borrow money and promise
to repay it, even though I know that this will never happen” (G, 4:422
[p. 74]) violates a “necessary duty to others or duty owed them” (G, 4:429
[p. 80]), that is, a perfect duty to others.21 Since this is the example of
a lie that he gives in section 1, and since he makes the same argument
against this lie that he makes in section 1 (that a maxim of making a lying
20. Kant can be said to agree with Marcus Singer that the dictum “‘one ought to
tell the truth’ ” is “subject to certain understood but unstated conditions,” namely, “if one
is called upon to speak, or if one says something, then one should tell the truth; if one
is asked a question, and one decides to answer it, then one should answer it truthfully.
The rule as so stated clearly does not require anyone to answer every question that he is
asked . . . The positive formulation of the rule, as well as the negative formulation, leaves it
open to one to be silent. .... The rule ‘One ought to tell the truth’ is logically and morally
equivalent to the rule ‘One ought not to lie’ ”; “Negative and Positive Duties,” Philosophical
Quarterly 15 (1965), 99–100. In the Doctrine of Virtue, pt. 2 of The Metaphysics of Morals,
Kant clearly distinguishes between being truthful and being candid: “Between truthfulness
and lying (which are contradictorie oppositis [contradictory opposites]) there is no mean;
but there is indeed a mean between candor and reticence (which are contrarie oppositis
[contrary oppositis]), since one who declares his thoughts can say only what is true without
telling the whole truth” (MM, 6:433 n. 1 [p. 556n. 1]). For more on the distinction between
truthfulness and candor, see my “Kant on Lies, Candour and Reticence,” Kantian Review 7
(2003), 102–33.
21. As Mary J. Gregor points out, “Less frequently, Kant uses ..... ‘necessary’ as
synonymous with ‘perfect’ duty”; Laws of Freedom (Oxford: Blackwell, 1963), 97.
promise in order to get money cannot be universalized), it can be argued

that the duty not to lie at issue in the Groundwork is a perfect duty to oth-
ers not to lie to others. This would appear to make the duty in question
either a duty of right or a juristic duty, and not a (directly) ethical duty.22
Two reasons can be given in support of this conclusion. First, an ethical
duty not to lie would prohibit all lies in the ethical sense, including lies
to oneself; however, the perfect duty to others not to lie to others does
not prohibit all lies in the ethical sense, since it does not prohibit lies to
oneself.23 Second, in The Metaphysics of Morals Kant does not allow for
perfect duties to others that are ethical duties.24
This leaves the question of whether the duty not to lie in the Ground-
work is a duty of right or a juristic duty. Some commentators have argued
that the duty at issue is a juristic duty:
The illustration of perfect duties to others—the duty not to promise false-
ly—is a prototype for the duties of justice which are later elaborated in the
Rechtslehre. The action such duties demand (in this case not making false
promises) is relatively well-defined and is owed to specifiable others; it can
be legally enforced, for example, by giving those who are defrauded by false
promisers rights of redress which can be pursued in the courts.25
22. In the Doctrine of Right, Kant says that all duties of right (and all juristic duties) are
indirectly ethical duties: “So while there are many directly ethical duties, internal lawgiving
makes the rest of them, one and all, indirectly ethical” (MM, 6:221 [p. 385]).
23. At least one commentator has argued that the perfect duty to others at issue in the
Groundwork cannot be a duty not to lie, but must be a duty to keep promises to others,
on the basis that the duty in question is a duty to others, and not a duty to oneself: “I
take it that a lying promise, as Kant calls it, is a promise one never intends to keep. Qua
promise its not being kept is a violation of duty to the promisee, and hence is classified
as a violation of a perfect duty to another. In contrast, as we shall see below, when Kant
discusses lying in the The Metaphysics of Morals, he discusses it as a violation of a duty to
oneself”; Nelson Potter, “Duties to Oneself, Motivational Internalism, and Self-Deception
in Kant’s Ethics,” in Kant’s Metaphysics of Morals, ed. Mark Timmons (Oxford: Oxford
University Press, 2002), 373 n. 7; emphases in original). However, this is wrong. Kant is
well aware of the distinction between making a lying promise and breaking a promise
made truthfully: “There are also lies whereby the other is cheated. To cheat is to make a
lying promise. Breach of faith is when we promise something truthfully, but do not have
so high a regard for the promise as to keep it” (LE, 27:449 [p. 204]). As Otfried Höffe has
said, “in the case of false promises, Kant is not, as is often assumed, concerned with the
injunction to keep promises under all circumstances.......Kant is not concerned with the
observable sequence of events, in which a promise is made and then kept or broken, but is
instead concerned with honesty as the subjective principle determining the will. He asks
whether it is morally permissible for someone in need to give a promise that he does not
intend to keep. The false promise ...... is seen as an instance of lying and deceit; Immanuel
Kant, trans. Marshall Farrier (Albany: State University of New York Press, 1994), 154.
24. Despite what is said here, for an argument that the duty not to lie at issue in the
Groundwork is an ethical duty, see my “Kant and the Perfect Duty to Others Not to Lie,”
British Journal for the History of Philosophy 14 (2006), 653–85.
25. Onora O’Neill, “Kant’s Virtues,” in How Should One Live? Essays on the Virtues, ed.
Roger Crisp (Oxford: Clarendon Press, 1996), 83.
Most lies are not lies in the juristic sense. A lie to obtain money is a lie in
all three senses of a lie, and therefore is a lie in the juristic sense of a lie.
Kant surely did not choose his example of a lie in the Groundwork lightly.
There is some reason to hold, therefore, that the duty not to lie to others
in the Groundwork is a juristic duty. Nevertheless, it is not certain that the
duty is a juristic duty.
It is also not certain what kind of duty not to lie is at issue in the Cri-
tique of Practical Reason. Indeed, it is not certain that there is a single duty
not to lie at issue in the Critique. In the Critique, Kant gives, as an example
of a duty, “[one] ought never to make a lying promise” (CPR, 21 [p. 154]).
He also gives, as example of a lie, giving untruthful testimony in court,
and argues for a duty to be truthful on the basis of this example:
When the maxim on which I intend to give testimony is tested by practical

reason, I always consider what it would be if it were to hold as a universal
law of nature. It is obvious that in this way everyone would be necessitated
to truthfulness. For it cannot hold with the universality of a law of nature
that statements should be allowed as proof and yet be intentionally untrue.
(CPR, 5:44 [p. 175])
It is not clear if giving untruthful testimony in court with the intention

that it be believed to be true is necessarily a lie in the juristic sense. If,
in such a case, there is necessarily a particular other person or persons
to whom the untruthful testimony is given with the intention that she
or he believe it to be true and be deprived of her or his rights, then it
is necessarily a lie in the juristic sense. If that is so, then a duty not to
lie in which everyone would be “necessitated to truthfulness” in giving
testimony in court could be a juristic duty. However, if, in such a case,
there is not necessarily a particular other person or persons to whom the
untruthful testimony is given with the intention that she or he believe
it to be true and deprived of her or his rights, then it is not necessarily a
lie in the juristic sense. If that is so, then a duty not to lie in which every-
one would be “necessitated to truthfulness” in giving testimony in court
could be either a duty of right or an ethical duty, although it seems more
likely to be a duty of right.
Later in the Critique, Kant appears to consider an example of an inten-
tionally “harmless” lie:
Has not every even moderately honorable man sometimes found that he
has abstained from an otherwise harmless lie by which he could either
have extricated himself from a troublesome affair or even procured some
advantage for a beloved and deserving friend, solely in order not to have
to despise himself secretly in his own eyes? When an upright man is in the
greatest distress, which he could have avoided if he could only have disre-
garded duty, is he not sustained by the consciousness that he has maintained
humanity in its proper dignity in his own person and honored it, that he has
no cause to shame himself in his own eyes and to dread the inward view of
self-examination? (CPR, 5: 87–88 [p. 210–11])
However, by “harmless lie” here Kant surely means harmless to the par-
ticular individual to whom the lie is told, since such a lie would still be
harmful to humanity. The duty not to lie here could therefore be a duty
of right not to lie or an ethical duty not to lie. However, given what Kant
says about it—“solely in order not to despise himself secretly in his own
eyes”—it seems that he is talking about an ethical duty not to lie.
Nevertheless, it is not certain what duty not to lie at issue in the Critique,
or indeed, that there is a single duty not to lie at issue in the Critique.
THE JURISTIC DUTY NOT TO LIE IN THE DOCTRINE OF RIGHT
The Doctrine of Right is exclusively concerned with a lie in the juristic

sense, a falsiloquium dolosum in praejudicum alterius. As Kant says, “in the
doctrine of right an intentional untruth is called a lie only if it violates
another’s right” (MM, 6:429 [p. 552]).
Kant does not mention a specific duty not to lie in the Doctrine of Right,
nor does he provide an argument for such a duty. The reason for this is
that there is no further wrong in telling a lie in the juristic sense than
that of intending to harm a particular person by depriving that person of
what is rightfully hers or his. Harming a person by depriving that person
of what is rightfully hers or his is wrong insofar as it is a violation of her or
his property rights, her or his contract rights, and so on. The wrongness of
telling a lie in the juristic sense, therefore, just is the wrongness of intend-
ing to violate a person’s property rights, rights based on contract, and so on.
An argument for a duty not to lie in the juristic sense is thus superfluous.
Nevertheless, there is a juristic duty not to lie in the Doctrine of Right.
It is a duty of omission of narrow obligation. It is a duty to refrain from
performing an external action (viz., making an untruthful statement to a
particular other person with the intention that this person believe that
statement to be true and be harmed by believing it to be true). It is a duty
that makes no reference to the person’s motivation for fulfilling the duty.
It is a duty that may be legislated by someone other than the person who
has to fulfill it (in particular, by the state), and the person may be com-
pelled to fulfill the duty by coercion.
THE ETHICAL DUTY NOT TO LIE IN THE DOCTRINE OF VIRTUE
The Doctrine of Virtue, part 2 of The Metaphysics of Morals, is exclusively

concerned with a lie in the ethical sense, a falsiloquium dolosum. Here
Kant pointedly distinguishes between a lie in the juristic sense and a lie
in the ethical sense:
In the doctrine of right an intentional untruth is called a lie only if it vio-
lates another’s right; but in ethics, where no authorization is derived from
harmlessness, it is clear of itself that no intentional untruth in the expres-

sion of one’s thoughts can refuse this harsh name. (MM, 6:429 [p. 552])
Since a lie in the sense of right also does not necessarily violate the
rights of a particular other person (although it necessarily “harms . . .
humanity generally”; RL, 8:426 [p. 612]), the claim that a lie in the
ethical sense does not necessarily violate the rights of a particular other
person is insufficient to distinguish between a lie in the ethical sense
and a lie in the sense of right. However, Kant also says that a lie to one-
self is a lie in the sense of ethics, even though it is intentionally harmless
to others:
A lie can be an external lie (mendacium externum) or also an internal lie.—

By an external lie a human being makes himself an object of contempt in
the eyes of others; by an internal lie he does what is still worse: he makes
himself contemptible in his own eyes and violates the dignity of humanity
in his own person. And so, since the harm that can come to others from
lying is not what distinguishes this vice (for if it were, the vice would consist
only in violating one’s duty to others), this harm is not taken into account
here. (MM, 6:429 [p. 552])
Although a lie to oneself is intentionally harmless to others, Kant does

add that the telling of lies to oneself is the source of telling lies to others,
which are intentionally harmful to others:
But such insincerity in his declarations, which a human being perpetrates

on himself, still deserves the strongest censure, since it is from such a rotten
spot (falsity, which seems to be rooted in human nature itself) that the ill of
untruthfulness spreads into his relations with other human beings as well,
once the highest principle of truthfulness has been violated. (MM, 6:430–31
[p. 554])
The “duty of truthfulness” (MM, 6: 404 [p. 532]) in the Doctrine of

Virtue is an ethical duty not to lie. The ethical duty not to lie is a duty to
oneself not to lie to oneself or to others. As Kant says in the Anthropology,
“the sole proof a man’s consciousness affords him that he has character is
his having made it his supreme maxim to be truthful, both in his admis-
sions to himself and in his conduct toward every other” man (A, 7:295 [p.
160]). The wrongfulness of lying, in the ethical sense, consists in what one
does to oneself when one lies. The human being, as a natural being, has a
natural capacity to communicate her or his thoughts. The human being,
as a moral being, can use the human being, as a natural being, to commu-
nicate her or his thoughts (either to herself or himself or to others). This
is an end that is “in agreement with” the end that the human being has
as a natural being. However, the human being, as a moral being, can also
use the human being, as a natural being, to communicate what are not
her or his thoughts (either to herself or himself or to others). This is an
end that is contrary to the end that the person has as a natural being. To
do this is to lie, in the ethical sense. When one lies, one (as a moral being)
uses oneself (as a natural being) as a mere means to an end. When one
lies, one (as a moral being) treats oneself (as a natural being) as a “speaking
machine,” that is, as a thing:
But communication of one’s thoughts to someone through words that
yet (intentionally) contain the contrary of what the speaker thinks on the
subject is an end that is directly opposed to the natural purposiveness of
the speaker’s capacity to communicate his thoughts The human being
as a moral being (homo noumenon) cannot use himself as a natural being
(homo phaenomenon) as a mere means (a speaking machine), as if his natural
being were not bound to the inner end (of communicating thoughts), but
is bound to the condition of using himself as a natural being in agreement
with the declaration (declaratio) of his moral being and is under an obliga-
tion to himself to truthfulness. (MM, 6:429–30 [p. 553])
Since to lie is to treat oneself as a thing, it follows that to lie is to treat

oneself as something less than a human being: “By a lie a human being
throws away and, as it were, annihilates his dignity as a human being”
(MM, 6: 429 [p. 552–53]). However, one is a human being, and to treat
oneself as something less than a human being, that is, as a thing, is a
wrongful act against oneself: “But his way of pursuing this end is, by its
mere form, a crime of a human being against his own person and a worth-
lessness that must make himself contemptible in his own eyes” (MM,
6:430 [p. 553]). From this it follows that one has a duty to oneself not to
lie to oneself or to others: “The human being as a moral being is under
obligation to himself to truthfulness” (MM, 6:430 [p. 553]).26
The ethical duty not to lie in the Doctrine of Virtue is a duty of omis-
sion of wide obligation. It is a duty to refrain from acting on maxims of
making an untruthful statement with the intention that that statement
be believed to be true—which may be maxims of internal lying (lying to
oneself) or maxims of external lying (lying to others)—from the motive of
duty. It is a duty that may only legislated by the person who has to fulfill
it. It is a duty that requires that the person fulfill it through self-constraint.
The violation of this duty, that is, acting on a maxim of lying, is a vice
(MM, 6:428 [p. 552]). Indeed, lying is the greatest violation of the duty to
oneself as a moral being to preserve oneself as a moral being: “The greatest
violation of a human being’s duty to himself regarded merely as a moral
being (the humanity in his own person) is the contrary of truthfulness,
lying” (MM, 6:429 [p. 552]). The fulfillment of this duty is a virtue: “Truth-
fulness in one’s declarations is also called honesty and if the declarations
are promises, sincerity; but, more generally, truthfulness is called rectitude”
(MM, 6:429 [p. 553]).
26. Nelson Potter has said in various places that “somewhat surprisingly Kant regards
lying, not as a violation of a duty to another, but as a violation of a duty to oneself; “How
to Apply the Categorical Imperative,” Philosophia 5 (1975), 415 n. 5 (and “Duties to
Oneself in Kant’s Ethics,” 386). However, the ethical duty not to lie must be a duty to
oneself, in order to prohibit lies to oneself.
THE DUTY OF RIGHT NOT TO LIE IN THE RIGHT TO LIE
The Right to Lie is exclusively concerned with a lie in the sense of right,
a falsiloquium dolosum in praejudicium humanitatis. The case Kant is con-
cerned with in this essay is a case raised by Benjamin Constant 27, that of
making the untruthful statement “No” to a would-be murderer asking if
one’s friend is in one’s house, with the intention that the would-be mur-
derer believe the untruthful statement to be true. This is not a lie in the
juristic sense, since in so acting one is not intending to violate the rights
of the would-be murderer (he has forfeited his rights by setting out on a
course of murder). However, it is a lie in the sense of right, since one is
necessarily intending to violate the right of humanity:
although I indeed do no wrong to him who unjustly compels me to make
the statement if I falsify it, I nevertheless do wrong in the most essential
part of duty in general by such falsification, which can therefore be called a
lie (though not in the jurist’s sense); that is, I bring it about, as far as I can,
that statements (declarations) in general are not believed, and so too that
all rights which are based on contracts come to nothing and lose their force;
and this is a wrong inflicted upon humanity generally. (RL, 8:426 [p. 612])
It is a right of “humanity generally” to enter into and maintain a society.

This right is violated by any untruthful statement made to others with the
intention that it be believed to be true, since in so acting one is necessarily
intending to bring it about that “statements (declarations) in general are
not believed,” and hence, that trust in contracts is undermined, and hence,
that rights based on contracts are undermined, and hence, that society is
undermined, since rights based on contracts are required to create and
maintain a society. The duty of right not to lie is the basis for all duties
(and correlative rights) based on contracts: “truthfulness is a duty that
27. Constant’s On Political Reactions [Des réactions politiques], first published in

1796, was translated into German and published in the journal Frankreich im Jahr 1797.
As Kant says, in this work, Constant claimed that “a German philosopher” had maintained
that “it would be a crime to lie to a murderer who asked us whether a friend of ours
whom he is pursuing has taken refuge in our house” (RL, 8:425 [p. 611]). Allen Wood has
said that “Constant is perhaps responding to an example Kant was to use in the Doctrine
of Virtue—a servant lies to the police in saying that his master is not at home, and this
lie enables his master to slip away and commit a crime; Kantian Ethics (Cambridge:
Cambridge University Press, 2007), 244. Wood criticizes Constant for ignoring the fact
that “the example occurs in a casuistical question not about lying as a violation of right
but lying as a violation of a perfect duty to oneself, grounded on self-respect” and says
that “Constant’s reworking of the example totally changes the issue that is supposed to
be raised by it” (326 n. 10). However, Constant could not have been responding to Kant’s
example in the Doctrine of Virtue, since Constant’s On Political Reactions was published in
1796, a year before the Doctrine of Virtue. Kant himself says, in reply to Constant, “I hereby
grant that I actually said this somewhere or other, though I cannot now recall where” (RL,
8:425 [p. 611]), but there is no trace of this example in Kant’s writings or lectures prior to
the Right to Lie.
must be regarded as the basis of all duties to be grounded on contract, the

law of which is made uncertain and useless if even the least exception to
it is admitted” (RL, 8:427 [p. 613]).
Some commentators have claimed that the duty at issue in the Right to
lie is the ethical duty not to lie: “Kant’s notorious argument that one must
not lie even to a would-be murderer in order to protect the life of his
innocent target thus does not turn on a claim that one owes the truth to
the murderer, but on the claim that one owes it to oneself only to tell the
truth.”28 However, when Kant says here that “truthfulness in statements
that one cannot avoid is a human being’s duty to everyone” (RL, 8:426
[p. 612]), and that “to be truthful (honest) in all declarations is therefore
a sacred command of reason prescribing unconditionally, and one not to
be restricted” (RL, 8:427 [p. 612]), the duty he has in mind is the duty of
right not to lie. As he says in a footnote:
I prefer not to sharpen this principle to the point of saying: “Untruthful-
ness is a violation of duty to oneself.” For this belongs to ethics, but what is
under discussion here is a duty of right. The doctrine of virtue looks, in this
transgression, only to worthlessness, reproach for which a liar draws upon
himself. (RL, 8:426 n. 1 [p. 612 n. 1])
The duty not to lie is a duty of omission of narrow obligation. It is

a duty to refrain from performing an external action (viz., making an
untruthful statement to others with the intention that others believe that
statement to be true). It is a duty that makes no reference to the person’s
motivation for fulfilling the duty. It is a duty that may be legislated by
someone other than the person who has to fulfill it (in particular, by the
state), and the person may be compelled to fulfill the duty by coercion.
It has been said about the duty of right not to lie that it is concerned
with “Deklarationen (declarations), and Erklärungen (statements),” rather
than with “Aussagen (acts of speech, speakings),” and hence that it is “a
formal duty to speak truthfully when you speak in such a way that you
will likely be taken by others to be speaking your mind,” with the result
“not all acts of speech, not all utterances, are covered by the formal duty.”29
However, this is just to say that the duty is only concerned with the mak-
ing of untruthful statements to others with the intention that they believe
them to be true, and that it does not prohibit non-intentionally-deceptive
untruthful statements, such as jokes, polite untruths, and tall tales.30
28. Paul Guyer, Kant (New York: Routledge, 2006), 403 n. 14.
29. Paul Griffiths, Lying: An Augustinian Theology of Duplicity (Grand Rapids, Mich.:
Brazos Press, 2004), 192.
30. Allen Wood has said that “the term “declaration” (Aussage, Deklaration, Latin
declaratio)” only refers to a statement that is made “in a context where others are
warranted or authorized (befugt) in relying on the truthfulness of what is said, and makes
the speaker liable by right, and thus typically subject to criminal penalties or civil damages,
if what is said is knowingly false” (Kantian Ethics, 241). However, in the Doctrine of Virtue,
when Kant says that “The human being . . . is bound to the condition of using himself as
Although Kant does not discuss the duty of right not to lie in the Doc-
trine of Right, he does discuss the duty in the lectures on moral philosophy
in 1784–85 (LE, 27:446–50 [p. 202–5]). This undermines the claim that
the argument in the Right to Lie for the duty of right not to lie is primarily
a response to Constant,31 as well as the claim that it is a product of bad
temper in his old age.32
Some commentators have argued that there is an “apparent discrep-
ancy”33 between the Right to Lie and the Doctrine of Right. They argue
that Kant “explicitly accepts here [Doctrine of Right] the view which he
rejects in the essay ‘On the Right to Lie’—the view, namely, that in law
an untruth is a lie only if it is to the prejudice of some one else (in praeju-
dicium alterius).”34 They argue that, according to the Doctrine of Right,
there is a right to tell lies to others, even if such lies are vicious, so long
as they are not intended to violate the rights of particular other persons:
“lying is always morally wrong, but he also thinks that as long as a lie is
not intended to deprive someone of her rights or property it should not be
prohibited by law.”35 Since the Doctrine of Right “expressly permits lying”36
to others, when it is not intended to violate the rights of particular other
persons, and since the Right to Lie expressly prohibits lying to others, there
is an apparent discrepancy between the Right to Lie and the Doctrine of
Right.
However, there is no apparent discrepancy between the Right to Lie
and the Doctrine of Right. When, in the Doctrine of Right, Kant says that
“the only kind of untruth we want to call a lie, in the sense bearing upon
rights [im rechtlichen Sinne], is one that directly infringes upon anoth-
er’s right, e.g., the false allegation that a contract has been concluded
with someone, made in order to deprive him of what is his (falsiloquium
a natural being in agreement with the declaration (declaratio) of his moral being” (MM,
6:430 [p. 553]), he is referring to declarations to oneself and to God as well as to others.
A declaration (declaratio) does not, therefore, only refer to a statement made in a context
where others are warranted or authorized in relying on the truthfulness of what is said.
31. See Robert J. Benton, “Political Expediency and Lying: Kant vs. Benjamin
Constant,” Journal of the History of Ideas 43 (1982), 135–44.
32. As Manfred Kuehn says, “while some have wanted to explain it [Right to Lie] away
as a product of Kant’s old age, it seems clear that it represents his considered view on the
subject, and that he would have presented essentially the same arguments at the time he
was writing the Groundwork; Kant: A Biography (Cambridge: Cambridge University Press,
2001), 403.
33. Sedgwick, “On Lying and the Role of Content in Kant’s Ethics,” 58.
34. Paton, “An Alleged Right to Lie,” 199–200.
35. Allen D. Rosen, Kant’s Theory of Justice (Ithaca, N.Y.: Cornell University Press,
2003), 70 n. 88.
36. Hiram Caton, “Truthfulness in Kant’s Metaphysical Morality,” in Essays in
Metaphysics, ed. Carl G. Vaught (University Park: Pennsylvania State University Press,
1970), 38 n. 57.
dolosum)” (MM, 6:238n. 1 [p. 394 n. 1]), he is talking about a lie in the
juristic sense. He is saying that the only kind of intentionally deceptive
untruthful statement that is called a lie, in the juristic sense, is one that
is intended to directly violate the rights of a particular person (a “par-
ticular jus quaesitum [special right] of another” (LE, 27:448 [p. 203]).
This is compatible with saying that the kind of intentionally deceptive
untruthful statement that is called a lie, in the sense of right, is one that
is intended to violate the right of humanity (“Recht der Menschheit” (LE,
27:447 [p. 203]). It is just that every intentionally deceptive untruthful
statement made to others is a lie in the sense of right.
ACKNOWLEDGMENTS
I would like to thank Allen Wood for discussions about Kant on lies, and for pro-
viding me with an advance copy of the chapter on lies from his Kantian Ethics.
I would also like to thank Andrew Chignell and the Sage School of Philosophy
at Cornell University for facilitating these discussions during Wood’s visit to
Cornell. Much earlier versions of parts of this essay were read at Duke Univer-
sity, Washington and Lee University, Harvard University, and the University of
Pittsburgh in 1999 and 2000, as well as at the Eastern Division meeting of the
American Philosophical Association in 2000, and at San Diego University in 2003.
I would like to thank all respondents and audiences on those occasions for their
comments. This essay was written while I was a Visiting Fellow in the Philoso-
phy Department at Princeton University in 2006–2007. I would like to thank
the Philosophy Department at Princeton, and especially Desmond Hogan, for
their warm hospitality. My sabbatical leave at Princeton was made possible by a
Hewlett-Mellon Fellowship from Washington and Lee University, and I would
like to thank my university for their continued generosity.
12
On the Supposed Duty of Truthfulness

Kant on Lying in Self-Defense
David Sussman
[I]
Kant’s uncompromising views about lying seem to be a reductio of his

entire moral philosophy. Kant presents lying as not just one form of wrong-
doing, but as the archetype of all immorality. Lying is simultaneously a
violation of a duty to others, a duty to oneself, and a duty to humanity in
general. Although Kant holds that as rational agents, even wicked people
have a kind of unconditional value, he claims that “by a lie a human being
throws away and annihilates his dignity as a human being” such that he
“has even less worth than if he were a mere thing” (Metaphysics of Morals
[MM], 6:430, also 6:481).1 Kant approvingly notes that scripture presents
evil coming into the world not through the first murder, but through the
first lie (MM, 6:431), and considers our endemic propensity to hypocrisy
and self-deception to be at the heart of the “radical evil in human nature”
that is the fundamental source of all moral corruption.
Despite such claims, we might hope that Kant can recognize special
circumstances in which lying, if not required, is at least morally permit-
ted. These hopes die with Kant’s notorious essay “On a Supposed Right
to Lie from Philanthropy” (SRL), a brief work written near the end of his
career. In this essay, Kant considers whether we may deceive a would-be
murderer about the whereabouts of his intended victim, a friend of ours
whom we are hiding in our home. Kant supposes that in this case there is
no option of remaining silent or otherwise evading the question. Suppos-
edly, anything other than an outright lie will betray the truth. Although
Kant offers no more details about the case, he also says that our admission
is being “compelled” by some “unjust constraint” (SRL, 8:426). Even so,
1. All references to Kant’s works are by the Akademie numbering of Practical

Philosophy, trans. and ed. Mary J. Gregor (Cambridge: Cambridge University Press, 1996)
and Lectures on Ethics, trans. Peter Heath, ed. Jerome Schneewind (Cambridge: Cambridge
University Press, 1997).
225
Kant insists that so long as we cannot avoid responding to the assailant’s

demands,2 we must answer truthfully: “Truthfulness (if he must speak) is
an unconditional duty . . . [that] makes no distinction between persons to
whom one has this duty and those to whom one can exempt oneself from
it, since it is, instead, an unconditional duty, which holds in all relations”
(SRL, 8:429).
Kant goes on to suppose that we manage to deceive the murderer, but
unbeknownst to us our friend has fled, so that our lie actually helps the
murderer find him. In this case, Kant claims, we “can by right be pros-
ecuted as the author of [our friend’s] death” (SRL, 8:427). We are morally
and legally responsible for all the bad consequences of our lie, no matter
how remote or unforeseeable they may be. In contrast, Kant holds that
if we truthfully answer the assailant, knowing that in doing so we are
betraying our friend to his death, we bear no such responsibility, even
though this consequence is obvious and immediate. Here we have sup-
posedly only done what morality demands, such that the murder is only a
foreseen but unintended consequence of our performance of our duty.
These outrageous conclusions are a gift to Kant’s enemies and a calam-
ity for his friends. Kant’s concern for avoiding moral and legal liability
smacks of the sort of “clean hands” fetish that is often claimed to be an
endemic failing of deontological ethics. Even worse,“On a Supposed Right”
seems to destroy any chance of defending Kant against the long-standing
charge of “rigorism” brought against his moral theory. This charge faults
Kant for issuing absolute prohibitions of entire classes of actions (e.g. false
promising, coercion, suicide) that admit of no exceptions or qualifications
for the moral complexity that real circumstances can frequently present.
Seen in this light, Kantian morality appears fundamentally pharisaical in
spirit. Rather than honestly confront the messy details of human life, Kant
seems to care only about an unthinking conformity to rules and the nar-
cissistic cultivation of one’s own sense of moral purity.
If not for “On a Supposed Right,” the charge of rigorism could be easily
laid to rest. Although Kant’s ethics are ultimately based on an absolute
and exceptionless “categorical imperative,” none of the formulations
of that imperative refer to specific act-types, and Kant does not derive
any absolute prohibitions against any such acts, considered in abstrac-
tion from the reasons for which they might be performed. The principle
2. Kant does not hold that we have to be particularly forthcoming about such
information, and generally thinks that a fair degree of reserve is necessary for the
maintenance of social life and self-respect. In the Lectures on Ethics, he says that we may
even dissemble, so long as we do not actually assert anything contrary to what we believe:
“We may knowingly deceive the other in a permissible way, if we try by our action or
utterance to promote the truth, or avert an evil; e.g. a pretended journey, to uncover or
thwart a crime.” (LE, 27:700). So long as another “has no right to infer from my utterance
a declaration of intent . . . I have told him no lie” (LE, 444–47 [Collins]). However, we may
not dissemble simply to “sniff out” another person’s views.
On the Supposed Duty of Truthfulness 227
formulations of the categorical imperative are framed with respect to such

considerations as the universalizability of maxims, respect for humanity
as an end-in-itself, and the potential legislation of a “kingdom of ends.”
There is no obvious reason to think that the correct interpretation of
these notions cannot be sensitive to whatever degree of moral detail is
appropriate. As “subjective principles of volition,” our maxims properly
represent everything that plays a role in our practical reasoning, every-
thing we take to be relevant to our decision to act in a certain way in a
particular circumstance. If so, then in principle Kantian duties can be as
fine-grained as our own deliberations about what to do.3
Kant does not generally claim that whole classes of action are pro-
hibited without exception or qualification. Although he recognizes that
there are strong moral objections to killing and coercion, he allows for
morally permissible instances of both in cases of self-defense, just war,
and legal punishment. He similarly condemns suicide in emphatic terms,
but in the “casuistical” questions of the Doctrine of Virtue he considers
whether it might be permissible to kill oneself when succumbing to
rabies, or to avoid being used as a hostage against one’s own country.
The Doctrine of Virtue also entertains casuistical questions about lying.
Here Kant is willing to strain at such gnats as whether it is wrong to
close letters with such untruthful expressions as “your obedient servant”
even though everyone recognizes them to be mere formalities. 4 He also
wonders whether we may lie when an author asks us our opinion of her
work, when she will take “the slightest hesitation in answering as an
insult” (MM, 6:431).
Kant offers no answers to these challenges in the Doctrine of Virtue.
However, by seriously considering these questions, he intimates that
there are real difficulties here, such that the correct answer cannot be
immediately read off of the moral law. If Kant were really a rigorist, there
should be little room for any interesting sort of casuistry or practical
judgment. Yet he insists that correct moral deliberation involves a crucial
element of “mother wit” that cannot be reduced to any sort of algorith-
mic decision-procedure (Theory and Practice, 8:275). Nevertheless, Kant
3. See Christine M. Korsgaard, “The Right to Lie: Kant on Dealing with Evil,” in
Creating the Kingdom of Ends (Cambridge: Cambridge University Press, 1996), 135–37.
Korsgaard argues that while the formula of universal law can allow maxims of lying
to the murderer, those maxims cannot pass the tests of the formula of humanity. The
latter conclusion assumes, however, that the formula of humanity admits of determinate
application without having to introduce the idea of a right that is the focus of Kant’s
dispute with Constant.
4. Although Kant himself exploited such language to escape from the promise he made
to Frederick Wilhelm II to never again write on religious topics. In an instance of the sort
of reservatio mentalis that he classes as a lie (LE, 27:702 [Vigilantius]) Kant claimed that
because he made the promise only as “Your Majesty’s most loyal subject,” he was released
from his obligations at the event of Frederick’s death (Conflict of the Faculties, 7:10).
considers nothing like the case inquiring assassin in his casuistical discus-
sions.5 Even if he is not a consistent rigorist, Kant still seems to think that
we have stronger reason to lie to protect an author’s vanity than to defend
the life of a friend who is under our protection.
Admittedly, the Doctrine of Virtue’s casuistical discussion is preceded by
a condemnation of lying that is almost as severe as that of “On a Supposed
Right.” Kant argues that all lies are violations of a basic duty to ourselves as
moral beings, in that they contravene the teleology of basic human pow-
ers. A lie “is directly opposed to the natural purposiveness of the speaker’s
capacity to communicate his thoughts, and is thus a renunciation by the
speaker of his personality, and such a speaker is a mere deceptive appear-
ance of a human being, not a human being himself” (MM, 6:429). For
Kant, lies renounce one’s personality even when “done out of frivolity or
even good nature . . . or to achieve a really good end” (MM 6:429).
This condemnation of lying seems to depend on Kant’s general ten-
dency to inflate the normative demands of teleological thinking. Some
such teleological claim is certainly plausible. It is hard to see how anything
like speech could have arisen if its characteristic use had been to deceive
or manipulate others. How can any individual come to master a language
if she does not take the vast majority of what she is told or of how others
present themselves at face value? For communication to be possible, com-
petent speakers must generally take it for granted that one needs a special
reason to lie, but no such reason to tell the truth. This is not to deny that
there can be pathological liars, who lie habitually and gratuitously. For
the basic teleological point to go through, it is enough that such cases are
necessarily exceptional, being parasitic on an implicit trust in the truth-
fulness of others commonly shared throughout a society. Although there
can be pathological liars, there cannot be any self-sustaining communities
of them.
However, such teleological reflections do support anything like an
absolute prohibition of lying. Kant often assumes that having identi-
fied a natural purpose of some power, we may immediately conclude
that we have moral obligation never to use that power in ways that
would deviate from or defeat that purpose. In the Groundwork, Kant
argues that we may not commit suicide out of self-love, in part because
5. The closest he comes is in the example of a servant who has been ordered to lie
about his employer’s whereabouts. Kant claims that should the employer then be able to
commit a crime, the servant shares in his guilt, since the servant “violated a duty to himself
by his lie, the results of which his own conscience imputes to him” (MM, 6:431). Here
Kant does not appeal to any of the considerations that will be central to his analysis of
the “murderer at the door” case. Kant does not claim that the servant violates any duty to
humanity in general with respect to the basis of contracts or political life. Kant does not
even claim that the servant has wronged those to whom he has lied. He argues only that
the servant has violated a duty to himself, which is apparently enough by itself to make him
culpable for any resulting crimes.
the natural purpose of self-love is the continuance of life. He similarly

asserts that we have some obligation to develop our talents because
these abilities were “given” to us to be put to use in some way. Kant
condemns gluttony, lust, and all nonprocreative intercourse as violations
of the natural purposes of our digestive and sexual capacities. Yet he
never explains why the teleology that characterizes us as natural beings
should be morally authoritative for us as free agents. Kant’s occasional
appeals to something like natural law seem to conflict with the basic
autonomy we are supposed to possess by virtue of pure practical reason
alone. If so, then Kant’s claim that all lies violate a duty to humanity in
one’s own person based in the natural purpose of speech would seem to
be on no firmer ground than his similarly categorical denunciations of
homosexuality and masturbation.
Kant wrote “On a Supposed Right to Lie from Philanthropy” in response
to a perceived challenge from Benjamin Constant. In 1797, a Berlin jour-
nal published a translation of Constant’s Des reactions politiques, in which
Constant claims that
the moral principle “it is a duty to tell the truth” would, if taken uncondi-
tionally and singly, make any society impossible. We have proof of this in
the very direct consequences drawn from this principle by a German phi-
losopher, who goes so far as to maintain that it would be a crime to lie to a
murderer who asked us whether a friend of ours whom he is pursuing has
taken refuge in our house. (quoted in SRL, 8:425)
The journal’s editor notes that Constant had told him that Kant was
the “German philosopher” in question, and in response Kant allows “that
I actually said this somewhere or other, though I cannot recall where”
(SRL, 8:426). Yet Kant seems never in fact to have advanced this remark-
able position in any prior work. In his earlier ethics lectures, he had instead
argued that lying may be permitted as form of self-defense, if such deceit
is necessary to prevent a wrong being done to oneself:
For example, somebody, who knows that I have money, asks me: Do you
have money at home? If I keep silent, the other concludes that I do. If I say
yes, he takes it away from me; if I say no, I tell a lie; so what am I to do? So
far as I am constrained, by force used against me, to make an admission, and
a wrongful use is made of my statement, and I am unable to save myself by
silence, the lie is a weapon of defense…. Hence there is no case in which
a necessary lie should occur, save where the declaration is wrung from me,
and I am also convinced that the other means to make a wrongful use of it.
(Lectures on Ethics (LE), 27:448 [Collins]).
No memory of this view seems to survive in “On a Supposed Right.”

Instead, Kant chooses to defend a bizarre position that he had never
advanced instead of the perfectly sensible view that he had actually put
forward. Here it is hard not to sympathize with H. J. Paton, who sees “On
a Supposed Right” as just evidence of the failing intellectual powers and
notorious irascibility of Kant’s last years. Paton speculates that the essay is
little more than Kant’s overreaction to the perceived slight of being caus-
ally characterized as merely “a German philosopher” by Constant.6
In any event, Kant probably had little love for Constant to begin with.
Robert Benton notes that Constant had written Des reactions politiques as
an apology for the Directory that came to power in France in opposition
to the Revolution’s more radical elements.7 Kant, “the old Jacobin,” may
well have seen the Directory as betraying the highest aspirations of the
Revolution (such as universal suffrage and economic equality). Constant
apparently wrote the pamphlet not out of political conviction but only
as an attempt to secure a position with the new regime in France.8 We
should not be surprised, then, to find Kant attacking not Constant’s actual
argument but rather the political opportunism that Kant may have seen
as the real motivation behind the essay.9
[ II ]
Yet for all its problems, we should not dismiss “On the Supposed Right to
Lie from Philanthropy” as merely a product of Kant’s wounded pride or
failing powers. Although the conclusions of the essay are wildly implau-
sible, they do have substantial motivation within Kant’s practical philoso-
phy. To appreciate the significance of the essay, we must first recognize
the essentially political nature of the problem Kant is considering. 10 He
does not treat the inquiring murderer as an ordinary moral problem that
requires direct application of the categorical imperative to some suitably
formulated maxim. Kant is quite clear that we would do no wrong to the
murderer by lying to him in these circumstances, as we would in cases of
6. H. Paton, “An Alleged Right to Lie: A Problem in Kantian Ethics,” Kant-Studien 45

(1953), 190–203.
7. See Robert J. Benton, “Political Expediency and Lying: Kant vs. Benjamin
Constant,” Journal of the History of Ideas 43, 1 (January–March 1982), 135–44.
8. Ibid., 137 n. 10.
9. “On a Supposed Right” does give evidence of pique. Kant begins by taking
Constant to task for talking about a “right to the truth” as if this meant a right over the truth-
value of some proposition, such that its value be under the control of the possessor, and
wholly subject to his will. Such a right would make the truth of some claim an item of
personal property, to be altered or traded as the owner saw fit. As Kant says, such a right
“would give rise to an extraordinary logic” (SRL, 8:426). Yet it is doubtful that anyone
has ever understood “a right to the truth” in this way. By “a right to the truth” Constant
clearly means, as Kant recognizes, a right to truthfulness (that is, a right to be told only
what the speaker actually believes). It is unlikely that Kant would have indulged in such
uncharitable pedantry had he not felt some personal animus against its target.
10. See John E. Atwell, Ends and Principles in Kant’s Moral Thought (Dordrecht:
Nijhoff, 1986), 193–202.
deceit that are directly proscribed by the moral law. For Kant, the prob-
lem of how to respond to the murderer only arises when the situation is
considered as a matter of right (Recht), that is, in terms of those aspects of
morality that bear on our relations as free and equal members of a politi-
cal community, as they may be enforced by a just legal system. 11 For Kant,
the murderer presents no dilemma outside of such a civil condition. In a
“state of nature,” everyone would be
authorized to do to others anything that does not in itself diminish what
is theirs, so long as they do not want to accept it—such things as merely
communicating his thoughts to them, telling or promising them something,
whether what he says is true and sincere or untrue and insincere . . . for it
is entirely up to them whether they want to believe him or not. (MM,
6:238)
In approaching the inquiring assassin as a problem of right, Kant

merely follows Constant, who frames his initial challenge in essentially
political terms. Constant claims not just that an unconditional duty to tell
the truth would have morally unacceptable results, but that such a duty
would “make any society impossible.” He argues that our duties of truth-
fulness must depend on more basic rights to the truth, the sort of right
that the murderer presumably forfeits through his criminal design. Kant
agrees that the murderer is not morally entitled to demand the truth, and
we would do him no wrong by lying to him. Yet against Constant, Kant
holds that this fact does not settle the moral status of such deceit. Kant
aims to show that although we may have a morally significant reason to
lie, and would do no wrong to anyone in particular by so doing, such lying
could nevertheless wrong the body politic and perhaps even humanity
as a whole. Supposedly, what Constant fails to appreciate is that we can
violate the necessary preconditions of “rightful relations” in general even
when we do not violate anyone’s actual rights in particular.
Kant has independent reasons to locate the problem of the inquiring
assassin within the domain of right and not the broader realm of virtue.
After all, Kant had developed his Doctrine of Right precisely to address
problems that are raised by the possibility of transgressions committed
by others.12 For Kant, morality requires that we stand in determinate rela-
tions of right, relations that define and maintain a certain sphere of “exter-
nal” freedom for each citizen. In addition to the “internal freedom” that
we attain through commitment to the moral law, morality demands that
we must have some range of activity in which we are not subject to the
11. If so, then Kant’s teleological reflections about lying as a violation of a duty to
oneself, even if they could be sustained, would not bear on the case. Kant is clear that all
questions of duties to oneself fall within the domain of virtue, which is quite distinct from
issues of right.
12. If so, then the “non-ideal theory” that Korsgaard thinks Kant needs is already largely
present, if not complete, in his Doctrine of Right. See Korsgaard, “The Right to Lie,” 147–54.
“arbitrary” choices of others. Although Kant considers coercion to be mor-

ally objectionable, he holds that our rights may be coercively enforced,
since otherwise the effective exercise of those rights would still depend
on the judgment and good graces of particular individuals. Insofar as coer-
cion serves as a “hindrance to a hindrance to freedom,” it is justified as a
necessary conditional that makes basic social relations of freedom and
equality possible.
Since morality demands that we stand in such relations, Kant holds
that we are not merely entitled but obligated to enforce our rights. How-
ever, no one may properly enforce his rights as an individual, since in so
doing someone’s right would again come to depend on the discretion and
goodwill of some particular individual or individuals. For Kant, we may
enforce our rights only collectively, through a common power that we
jointly recognize as our agent in this regard. The authority of the state
supposedly rests on its role as the sole legitimate enforcer of our rights.
Absent such an authority, the defense of any right would depend on the
judgment of particular defenders (both those who think their rights vio-
lated, and those who are in a position to support or resist them). Any
such “unilateral” defense of right is itself in violation of basic conditions
of right, even when this defense is guided by judgments that are substan-
tively correct. For Kant, a true state of nature would have to be a morally
impossible world. We would find ourselves obligated to defend our rights,
even though every way of doing so would itself be a violation of the
rights of others. Here there would be a conflict not just between “grounds
of obligation,” but between strict, determinate obligations themselves.
Morality would be in contradiction with itself, and thereby lose any claim
to rational authority over us.
Yet Kant insists that the rational authority of morality is absolutely
unconditional. If so, then we can never countenance any suggestion that
we really do or might come to occupy a state of nature with respect to
others. Since we cannot recognize the possibility of a real state of nature,
we must interpret our social life, however bad it seems to be, as something
other than this condition. For Kant, the only alternative to the state of
nature of nature is the “civil condition,” in which we collectively acknowl-
edge a common authority as the proper enforcer of our rights. Insofar as
I recognize the state as my agent, the state’s action can satisfy my duty
to defend my rights. Yet, since the state’s authorization is based equally
in the rights of all citizens, such acts of enforcement will not involve the
subordination of anyone to the will of anyone else in particular. Through
the agency of the state, we jointly enforce the rights of each of us, leav-
ing each individual in relations of freedom and equality with respect to
everyone else.
Kant argues that regardless of the “empirical character” of the polity in
which we find ourselves, we must approach the government as if it had
been authorized through something like a social contract of all its citizens
to serve as the sole enforcer of our rights. It does not matter whether such
contract has actually been transacted, or even if we would choose this

government over some alternative from any real or hypothetical starting
point. For our actual relations to other citizens to be minimally accept-
able morally, these relations must be construed in light of such a sup-
posed social contract that defines the “intelligible character” of the state.
We must fundamentally identify ourselves as members of a “general will”
whose intentions can be expressed, however imperfectly, only through
our political and legal institutions.
[ III ]
In “On a Supposed Right to Lie from Philanthropy,” Kant recognizes that

the murderer has no right to the truth from us, and that we would do him
no wrong by deceiving him in order to frustrate his evil plans. Lying to
the murderer cannot be wrong in the same way as are ordinary lies told to
exploit or manipulate others, which presumably do directly wrong their
recipients. Where Kant parts from Constant is over whether, in order to
violate rights, a lie must harm or violate the rights of anyone in particular.
Kant argues that in addition to violating the rights of a specific person, a
lie might also violate the right of humanity as a whole. The problem with
lying to the murderer is that
I bring it about, as far as I can, that statements (declarations) in general are
not believed, and so too that all rights which are based on contracts come
to nothing and lose their force; and this is a wrong inflicted upon humanity
generally . . . inasmuch as it makes as it makes the source of right unusable.
(MM 8:426)
Here Kant may have a plausible general reply to Constant’s claim that,
in order to have a potentially enforceable duty to tell the truth, there
must be a particular individual to whom we owe the truth. Kant recog-
nizes that we may also have duties based not in the claims of assignable
persons, but also with respect to what is needed to sustain the minimal
trust needed for any sort of just social order. Yet this reply hardly seems
adequate to the example Constant presents. By lying to the would-be
murderer, I hardly repudiate my commitments to speak truthfully and
uphold contracts in normal circumstances. Here my target is simply those
who would seek to exploit such honesty in order to violate the basic
rights of others. Such an attitude would seem to admit of being publicly
shared and avowed without having to renounce the basic commitments
needed to sustain a political community.
It is also hard to believe that such a defensive lie would have the effect
of undermining our general trust in contracts and promises. If my lie is
effective, it may well remain secret, and in this case there will be no such
repercussions whatsoever. Even if my lie is revealed, my motives should
become apparent as well, such that neither my honesty nor that of anyone
else need come into doubt. Criminals might lose faith in any declara-
tions that they extort from innocent bystanders, but this would seem to
strengthen rather than undermine any effective regime of rights. While
we might not be able to effectively deceive such criminals, we would at
least be able to give them a useless response, and so resist their attempts
to make us complicit in their evil designs. In contrast, an absolute duty
of truthfulness would put us in the power of anyone who can intelligibly
address a question to us.13 Yet Kant insists that “truthfulness is a duty that
must be regarded as the basis of all duties grounded on contract, the law
of which is made uncertain and useless if even the least exception to it is
admitted” (MM, 8:427). He maintains this even though the actual laws
of contract have in fact tolerated all sorts of exceptions and qualifications
without thereby becoming completely “uncertain and useless,” at least for
ordinary purposes.
[ IV ]
In “On Lying and the Role of Content in Kant’s Ethics,” Sally Sedgwick
argues that in “On a Supposed Right” Kant is only opposing the general
principle that duties of right may be set aside whenever this would help
another person, at least if so doing does not harm anyone else. The full
title of the essay is after all “On a Supposed Right to Lie from Philan-
thropy” (Menschenliebe). Kant would be on firm ground in arguing that we
could not stand in relations of right if each individual were authorized to
lie, break a promise, ignore a contract, or lay hold of another’s property
whenever she sincerely thought that enough good might come to another
because of it. In this case, our rights would once again depend on the
judgment and goodwill of particular individuals. The true extent of our
freedom would depend on whether others think that ignoring our rights
would contribute to someone’s welfare. Our right would be conditioned
by the particular degree of importance that others happen to accord the
imperfect duty of beneficence and the various sorts of attachments they
might have to those to whom it might be shown.
Kant observes that “from one’s right to require another to lie to one’s
advantage would follow a claim opposed to all lawfulness” (SRL, 8:428)
and concludes that the duty of honesty, as a “sacred command of rea-
son,” is one “not to be restricted by any inconveniences” (SRL, 8:428).
Sedgwick contends that Kant is considering only the basic principles
of our political relations, and does not mean to rule out more specific
“principles of application” that might allow lying in special cases like the
one Constant proposes. Supposedly, Kant holds only that lying can never
be justified by direct appeal to the benefits it might bring. Any such
13. Korsgaard contends that we have in fact a “perfect duty of virtue” to ourselves not
to allow our goodwill to be used as a “tool of evil” (“The Right to Lie,” 145–46).
justification would instead have to be framed in the language of rights, in

terms of relations of equal freedom rather than maximal welfare. While
there may be no right to lie from philanthropic motives as such, other
reasons might provide adequate justification. Kant would then be mak-
ing a point only about the beginning, rather than the end, of proper
moral deliberation.
So read, “On a Supposed Right” advances a position that is plausible if
unremarkable. Yet if Sedgwick is correct, then Kant’s dispute with Con-
stant is based on a misreading of Constant’s position that is almost willfully
obtuse. After all, Constant never proposes anything like a general “right
to require another to lie to one’s advantage.” Even if Constant thought
that we were at liberty to lie whenever doing so would be to someone
else’s substantial benefit, such an entitlement need not give that person
any right to demand that we do so. I am at liberty to donate a kidney to
whoever is most likely to benefit from it, and have strong moral reasons
to do so, but these facts do not in themselves give any person a right to
demand a kidney from me.
Moreover, it would be exceptionally uncharitable of Kant to think that
Constant supposes that we should be free to lie for the sake of any benefi-
cent concern at all. Admittedly, Kant frames his discussion in terms of
generic “philanthropic motives,” “expediency,” and “advantage,” and by so
doing obscures the features of the case that motivate the real moral prob-
lem. To keep someone from being murdered is not merely to promote his
happiness. It is to defend what is arguably his most vital right. Such defense
is not merely a “convenience,” but something we have a compelling moral
reason to do out of recognition of any of that person’s rights at all. Kant
stubbornly ignores other features of the case that further distinguish it
from a mere matter of expedience. In Constant’s example, the potential
victim is supposed to be my friend, and I have apparently accepted and
encouraged his trust by allowing him to take refuge in my home (had the
victim known about my scruples about lying, he might well have sought
refuge elsewhere). Kant also neglects the fact that the murderer is trying
to make me complicit in a profoundly wrongful act in a way that would
involve invading both my privacy and my property.
Sedgwick claims that Kant is only arguing that fundamental principles
of right cannot include exceptions for the sake of expediency, and that he
can accept that special cases of emergency may be recognized in subor-
dinate rules of application. Yet if this were so, Kant and Constant would
have no real disagreement. Kant quotes approvingly Constant’s observa-
tion that “every time that a principle proved to be true seems inapplicable,
this is because we do not know the intermediary principle, which contains
the means of application” (SRL, 8:427). Although Kant accepts a general
need for such intermediary principles of application, he explicitly claims
that there can be no such principle that would allow for a defensive lie:
“And yet [Constant] himself had abandoned the unconditional principle
of truthfulness because of the danger to society it brought with it, since
he could discover no intermediary principle . . . and here there is actually no

such principle to be inserted” (SRL, 8:428, my emphasis).14
For Kant, a proper intermediary principle specifies a particular way
requirements of a more general requirement might be instantiated in spe-
cial contexts. He gives the example of how the basic demand for demo-
cratic participation in legislation can be satisfied in large societies through
representative institutions. In contrast, Kant sees Constant as calling for
not a more exact specification of the principle of truthfulness, but rather
a wholesale exception to it. Such exceptions would specify special cases
in which the general principle did not apply at all, or ceased to have
the degree of authority it would otherwise have. Kant explains that “so-
called intermediary principles can contain only the closer determination
of their application to cases that come up . . . but never exceptions from
those principles; for exceptions would nullify the universality on account
of which alone they are called principles” (SRL, 8:430).
Unfortunately, Kant never explains how to distinguish an acceptable
“closer determination” of a principle from such an objectionable “excep-
tion.” The distinction seems to turn largely on the difference between a
correct and an incorrect interpretation of the principle. Until we have
settled just what the principle calls for in particular cases, how can we
decide what would count as making an exception to it? Kant’s rejection
of any entitlement to lie to the murderer seems to depend on an equivo-
cation about “exceptions” that conceals the question-begging nature of
the conclusion. Such lying is clearly an “exception” to our basic principles
only in the sense that it is of a kind of act that the principles prohibit in
most other situations. This establishes only that such lies are atypical and
require special justification, not that they would have to count as cases
where the basic principle no longer applies at all.
[V]
Kant claims that in lying to the murderer, we would “do wrong formally
though not materially” (SRL, 8:429; also LE, 27:449 [Collins]) but never
explains what the distinction between formal and material wrongs comes
to in this context. The contrast seems to be between wronging someone
in particular (i.e., violating a particular person’s rights) and acting from a
principle that conflicts with the possibility of rightful relations in general,
even when no particular person’s right is violated. Yet if Kant thinks that
14. See Alasdair MacIntyre, “Truthfulness and Lies: What Can We Learn from Kant”?
in MacIntyre, Ethics and Politics (Cambridge: Cambridge University Press, 2006), 129–30.
Sedgwick recognizes the difficulty Kant’s remarks here pose for her reading, but dismisses
them as part of a polemical overreaction by Kant to Constant’s political opportunism;
Sally Sedgwick, “On Lying and the Role of Content in Kant’s Ethics,” Kant-Studien 82
(1991), 42–62.
the objectionable principle at stake in this case is that our rights may be
ignored when doing so promotes utility, he is just refusing to think seri-
ously about the case Constant presents.
However, there is another way we might think about the distinction
between formal and material wrongs that might give Kant’s objections
more force. In matters of right, every act we take has a twofold nature that
may provide a plausible way of interpreting the material/formal distinc-
tion. Our acts have a material aspect determined by the end at which they
aim. This aspect of my act, and its moral significance, would remain the
same even if I were in a state of nature. Yet in a civil condition, our choices
also have a political form or aspect. In addition to adopting an end, in
each choice I arrogate a certain power to myself. Insofar as I take myself
to be in rightful relations, I am implicitly claiming that I am entitled to act
on such a decision in such a circumstance. Kant thinks that for an act to
be rightful, it must accord with the “universal principle of right” in both
these respects. The pursuit of the end must be consistent with the law-
governed external freedom of others with respect to similar pursuits, and
any implicit assumption of entitlement or authority must be consistent
with the equal and law-governed assignment and free exercise of such
powers by everyone else as well.
These two requirements can come apart for Kant. The most notorious
example is his other outrageous conclusion: the claim that all political
revolution is morally wrong. Kant holds that a government’s authority
derives from its role as the embodiment of our collective authority to
specify and enforce our rights. As such, any law must at least be such that
it could have proceeded from the “general will” of a body politic commit-
ted to maintaining rightful relations among its members (regardless of the
actual mechanisms of legislation). Kant recognizes that a particular gov-
ernment may completely fail in these tasks: its laws might be incompat-
ible with anything that could count as a general will, and the government
may even have abandoned the pretense of maintaining rightful relations
altogether. Such a government would be illegitimate, and could not prop-
erly demand our obedience. Nevertheless, Kant holds that it would still be
wrong for citizens in such circumstances to revolt against this govern-
ment. Although he thinks that we must not comply with immoral laws or
commands, we nevertheless may not offer resistance in any way that
would undermine or repudiate of the government’s authority in general.
In such grim situations, revolution is not materially wrong; since the
government is indeed illegitimate, the revolutionary violates no real claim
it might have to his allegiance. Yet revolution would still be formally wrong
in much the same way the lying to the murderer might be. The revolu-
tionary assumes the right to judge the adequacy of the government, and to
release himself from obedience if he finds these institutions unsatisfac-
tory. For Kant, no set of rightful relations could assign such a prerogative
to individuals. The problem is not merely the risk that people might mis-
judge a decent government and undermine truly legitimate institutions.
For Kant, it is the mere fact that an individual would be taking himself
to be entitled to make such a decision, even if he can do so infallibly, that
would immediately vitiate any rightful relations he might have.15
For Kant, political authority is grounded in the need to sustain rights in
such a way that no person’s right depends on the judgment or goodwill of
any other person. No government could perform this function if individu-
als were authorized to decide, as individuals, whether and when they owed
it their allegiance. In this case, each citizen’s rights would still ultimately
depend on discretion and good will of others, and so we would remain in
what is effectively a state of nature, however peaceful and harmonious it
might be. For Kant, a government could be properly repudiated not by
individuals taken severally, but only by the body politic acting as a whole.
Yet Kant thinks that we must take the extant government as the only
institution through which such collective decisions can be expressed or
collective actions taken. As a result, for there to be a permissible way for
the people to challenge the government, they must already have some
alternate way of acting as a people, independent of that government. And
so a revolution would have already to be successfully completed before it
could be even begun.
A similar analysis can be applied to the supposed right to lie in defense
of someone’s rights.16 Unlike the revolutionary, the defensive liar is not
challenging the legitimacy of anyexisting political institutions. Instead, she
is only considering whether the assailant is entitled to the protections of
his rights in the usualway. By choosing to lie, she claims an entitlement to
do what would otherwise be wrong insofar as this is necessary to prevent
a serious violation of another’s rights. Like the revolutionary, the liar may
be perfectly right about the particular case. The murderer has no more
of a legitimate complaint against the liar than a despotic government has
against the revolutionary. Nevertheless, the liar might still be committing
a “formal wrong” against the body politic by essentially taking the law into
her own hands. She takes herself to be released from her normal obliga-
tions because of her own assessment of the ultimate intention behind
murderer’s question, even though he is within his rights simply to ask.
For Kant, any authorization to so judge the overall intentions of others
would be inconsistent with the common allegiance to the body politic
that makes determinate relations of right possible. If our rights are not to
depend on anyone’s personal discretion, then the state cannot be under-
stood as just a tool that individuals use to adjudicate and manage their
15. Here I am indebted to Christine M. Korsgaard, “Taking the Law into our Own
Hands: Kant on the Right to Revolution,” in Reclaiming the History of Ethics, ed. Andrews
Reath, Barbara Herman, and Christine M. Korsgaard (Cambridge: Cambridge University
Press, 1997) 297–328. See also Sarah Williams Holtman, “Revolution, Contradiction, and
Kantian Citizenship,” in Kant’s Metaphysics of Morals, ed. Mark Timmons (Oxford: Oxford
University Press, 2002), 209–32.
16. For a similar suggestion see Atwell, Ends and Principles in Kant’s Moral Thought, 198.
prior rights and interests. If this were the case, citizens would be entitled
to quit civil society whenever they judged that the state had failed to
properly perform its function, and all rights would once again ultimately
rest on the goodwill of individuals. Citizens can avoid this result only
if they refuse to acknowledge any position outside of civil society from
which they might call its basic legitimacy into question. Every individual
must fundamentally identify with the body politic, and take her role as
part of that collective authority to be who she most truly is. No matter
how bad things become, she must understand her rights not as an individ-
ual possession, but as something fundamentally assigned to her through
the exercise of a more basic joint sovereignty.
The revolutionary repudiates this self-understanding by taking himself
to be released from his political obligations once he judges the govern-
ment to be defective. In defensive lying, I would similarly usurp the proper
enforcement authority of the entire community. Here I take myself to be
entitled to release myself from my normal obligations on the basis of my
assessment of someone’s ultimate intentions. In so doing,
we have implicitly broken the pactum, and acted contrary to the right of
humanity The case of emergency subverts the whole of morality, since
if that is the plea, it rests upon everyone to judge whether he deems it an
emergency or not; and since the ground here is not determined, as to where
emergency arises, the moral rules are not certain. (LE, 27:449 [Collins]).
Since the rules defining an emergency cannot be made fully deter-

minate and obvious, such an authorization would have to subordinate
our rights in general to the personal discretion of other individuals. Both
the revolutionary and the defensive liar would be asserting a normative
identity that is prior to their common identity as citizens of a particular
political community, and so would make any rightful relations with them
impossible.
[ VI ]
Kant may be right to worry that defensive lying presupposes a principle

that is incompatible with the basic, quasi-contractual commitments that
are the “source of right” (Rechtsquelle). that makes a just social order possi-
ble. In order for us to avoid being in a state of nature, these commitments
must be taken as absolute, allowing no exceptions or qualifications that
would allow an individual, as an individual, to reassess their legitimacy.
This interpretation may answer the charge that “On a Supposed Right”
is merely the fruit of Kant’s senility, national pride, or general crankiness.
Yet this reading also makes the conclusion of the essay even more incred-
ible. It was bad enough to argue that we may not lie to the murderer at
the door. Now it seems that we may not do anything in self-defense (or
the defense of others) that would count as a violation of the right of the
assailant in a more normal context. It seems that I may not lie, threaten, or
use even mild force even if I am the person the murderer is seeking. If all
enforcement must proceed through the proper institutions, then all I can
do is call the police. If they cannot or will not act in time, it would seem
that my only appeal is to heaven.17
Unfortunately, Kant never discusses emergency self-defense at any
length. In the introduction to the Doctrine of Right, he claims that we may
sometimes defend ourselves in ways that would normally violate some
right. Kant considers what he takes to be the spurious “‘right’ of neces-
sity,” which would authorize a person to do anything she needs to save
her own life. In arguing against this supposed right, Kant distinguishes
such cases from “the issue of a wrongful assailant upon my life whom
I forestall by depriving him of his life (ius inculpatae tutelae), in which
case a recommendation to show moderation (moderamen) belongs not to
right but only to ethics” (MM, 6:235). Here there is no prohibition on the
use of deadly force, but merely a suggestion that we try to limit the force
used that falls outside of the domain of right. Yet if in an emergency I may
respond to the threat of violence with violence, why not merely with a
lie? Could my justification for doing so in Constant’s example be weaker
because I would be trying to save the life of another person, rather than
my own?
Given that Kant accepts the possibility of legitimate individual self-
defense, “On a Supposed Right” may only be protesting a particular way
of conceiving of such resistance. We might think that when the murderer
comes to my door, and the police cannot help me, I enter into a local and
temporary state of nature with him. I would then be entitled to act in
whatever ways would be permissible in such a state, such as by lying or
using force. Kant must reject this tempting interpretation, because any
state of nature, even a supposedly temporary and local one, would com-
mit practical reason to a self-contradiction that would permanently vitiate
its authority. By countenancing the option of withdrawing to such a state,
even as an emergency measure, a citizen would renounce her fundamen-
tal identity as part of a particular body politic, and with it any possibility
of rightful relations.
However, Kant has the resources to understand self-defense in a way
that does not invoke any such temporary or limited state of nature. He
holds that rights must be generally enforceable in ways that do not rest on
individual discretion and initiative. If self-defense were always illegitimate,
then our rights would once again end up depending on the goodwill of
particular others in every instance where the police could not effectively
17. See John Locke, Second Treatise of Government, ed. C. B. Macpherson (Indianapolis:
Hackett, 1980), 91–92.
intervene. If so, then we can know a priori that any general will com-
mitted to maintaining rightful relations would delegate some of its basic
enforcement authority to individuals who find themselves in such situa-
tions. Just as Kant thinks that any general will would have to specify some
rights of property, contract, and household authority, so too it would have
to assign some of the police power to individuals who confront these
emergencies.
As a matter of what pure reason demands, the extent of this power
is largely indeterminate, and will depend on actual political institutions
for its full specification, just like our other basic rights. Even so, we can
know a priori that there must be some such emergency powers, and so
can reject out of hand the position that we have no special authorization
of self-defense at all. No general will could leave us utterly powerless
to resist attacks on our own rights or those of others. We may then be
entitled to assume at least authorization to lie in circumstances supposed,
insofar as this seems to be the most minimal effective resistance that
can be conceived. The murderer is after all threatening one of the most
important rights, and the lie is specifically targeted to frustrate only his
attack, leaving his other rights and interests largely unaffected (unlike the
use of force). If I am not authorized to do even this, I am left completely
at the murderer’s mercy.
Admittedly, in such self-defense an individual would still have to act
on her own judgment, but this need be no more problematic than the
fact that particular police officers have to make their own decisions in
the performance of their duties. Yet the police do not conceive of their
actions as being guided directly by their own moral judgment as private
citizens. Instead, public institutions exist to hold the police accountable
in the exercise of the special discretion that comes with their role, and
police officers properly deliberate and act in light of this fact. Similarly, in
self-defense a person has to make up her own mind about many things. In
so doing, she may take on new legal responsibilities, becoming specially
liable for the exercise of good judgment here in ways that go beyond
what might be required of citizens generally.18 What is crucial here is that
we approach self-defense not as private individuals suddenly thrust into
a state of nature, but as agents implicitly deputized to enforce the law in
the event of certain kinds of emergency. We do so in a morally appropri-
ate way so long as we accept in good faith any restrictions and special
responsibilities associated with this role that would issue from any truly
general will.
18. Although it is hard to believe that anyone could be held responsible for such
unlikely calamities as those Kant describes, when the lie unwittingly leads the murderer
to his victim. Arguably, no general will would assign such strict liability to individuals who
have to act in such emergencies.
[ VII ]
Constant’s example may present a special problem in that the liar lies not
in response to any wrong the assailant has committed, but only in antici-
pation of crimes that he is expected to commit. The would-be murderer
does not obviously violate anyone’s rights by simply asking after his vic-
tim. In contrast, when I kill an attacker in self-defense, I am obstructing
a wrongful act that is already under way. For Kant, as fellow members
of the body politic we must relate to one another as fundamentally free
and responsible beings. This attitude must govern not just personal rela-
tions, but the relations between individuals and various political and legal
institutions. In preemptive self-defense (like preventative detention), we
would condition someone’s rights not on what he is actually doing, but on
the basis of some prediction of what he will do. His rights as a free, ratio-
nal agent would then be subordinated to a view of him as a determined
natural system, to be managed and controlled in ways not that different
from other physical phenomena.
For Kant, such a fundamentally predictive stance toward people is
incompatible with basic relations of right. No general will could recognize
such a police power that would generally regard others in this way. Yet if
such a power could not be assigned to the police, it cannot be available to
individuals acting as emergency deputies. This limitation need not apply
to other kinds of self-defense. When I kill my attacker, I am responding to
his free choices as an intelligible being, rather than anticipating his future
behavior as a merely sensible one. An authorization to resist in such cases
need not conflict with the basic understanding of ourselves that is the
precondition of all substantial rights, in that it is not conditioned by any
sort of psychological prediction. The important distinction is not between
lying and using force. Instead, it is between frustrating a crime already in
progress and forestalling a crime that has yet to begin. Kant could allow
that we may lie to defeat some criminal effort (e.g., by telling a mugger
that one has no money). The problem with Constant’s case is that in his
inquiries, the murderer has yet to overstep his rights, and so it seems that
the liar could not yet be authorized to step beyond hers.
The question here ulimtately turns on what counts as part of an
attempted crime. If the assassin’s inquiry is an act that is really distinct
from his attempt at murder, then I cannot be authorized to lie to him,
regardless of what I can reasonably and confidently predict about what
he will do with my answer. However, it may be more accurate to see his
inquiry as just the first part of his attempt to kill. This becomes all the
more plausible if, as Kant suggests, the murderer is already exerting some
kind of pressure on me to answer. The question would then be a compo-
nent of a wrongful act already in progress that authorizes me to lie or use
some degree of force against him in response.
Admittedly, we cannot appreciate the wrongness of the assailant’s
question by considering it in isolation from the overall pattern of activity
of which it is a part. Similarly, there is nothing obviously wrong with

cocking a fist or the hammer of a gun, apart from the broader actions
in which such behaviors are embedded. Of course, people ask after one
another all the time without any criminal intent, while the cocking of a
fist is usually the prelude to an assault. Because of this difference, we may
more readily recognize the latter as part of a criminal action already under
way. In contrast, the question asked by the murderer can serve all sorts of
innocent purposes, and we would normally be less confident of its broader
aim than we would in the case of a drawn gun. Yet Kant allows that in
Constant’s example, there is no doubt about why we have been asked to
reveal the location of our friend. With normal epistemic worries put aside,
we can recognize this question as the initial stage of an attack that we
may resist by either lies or force, as need be. Insofar as we are implicitly
deputized with respect to such emergencies, it may well be our moral and
legal duty to do so.19
19. The situation is more vexed if the would-be murderer is himself acting as an agent
of the state. If I am right, this case should not be conceived of as a private person opposing
the political authority, but rather as a conflict between two different enforcement powers
of the state.
13
User-Friendly Self-Deception
A Traveler’s Manual
Amelie Rorty
Since many varieties of self-deception are ineradicable and useful, it

would be wise to be ambivalent about at least some of its forms. 1 It is
open-eyed ambivalence that acknowledges its own dualities rather than
ordinary shifty vacillation that we need. To be sure, self-deception remains
dangerous: sensible ambivalence should not relax vigilance against pre-
tense and falsity, combating irrationality and obfuscation wherever they
occur.
The animus against self-deception has an honorable origin: the motto
“Know thyself” was inextricably linked to the Socratic enlightenment proj-
ect, to the systematic critical examination of belief, its clarification and
Note: An early version of this chapter was delivered at colloquia at the East-West Center
in Honolulu, Hawaii and at Williams College. I am grateful to Annette Baier, Brian
McLaughlin, Sam Fleischacker, and Steven Gerrard for comments.
1. One variety of self-deception: X is self-deceived about p when
(1) X believes that p at t (where t covers a reasonable span of time);
(2) Either (a) X believes not-p at t, or (b) X denies that he believes p at t;
(3) X recognizes that p and not-p conflict;
(4) X denies that his beliefs conflict, advancing an improbable ad hoc reconciliation,
making no attempt to suspend judgment or to determine which belief is defective.
Since conditions 1 and 2 are parallel to conditions 3 and 4, the attribution of self-
deception is regressive. It is typically justified by an inference to the best explanation, an
account of what X would normally believe, perceive, notice, infer. For more elaborate
formulations of these conditions, see Leon Festinger, Cognitive Dissonance (Stanford:
Stanford University Press, 1957), and B. McLaughlin, “Exploring the Possibility of Self-
deception in Belief,” R. Audi, “Self-deception, Rationalization and Reasons for Acting,” and
A. O. Rorty, “The Deceptive Self: Liars, Layers and Lairs,” in Perspectives on Self-Deception,
ed. B. McLaughlin and A. Rorty (Berkeley: University of California Press, 1988). Please also
see the essay of Al Mele included in this collection.
244
User-Friendly Self-Deception 245
justification.2 But the dangers of self-deception were nevertheless magni-

fied by those who misunderstood the fundamental conviction of the later
Enlightenment that we shall know the truth, and the truth will make us
free. Because the narrow and naive interpretations of that project assigned
a central role to self-consciousness and self-knowledge in the complex
tasks of liberation through knowledge, self-deception seemed threaten-
ing to the primary tasks of rational inquiry. The denial of a systematic
tendency toward various forms of irrationality—toward self-deception,
akrasia, and the conservation of emotions—is, in effect, the Enlighten-
ment’s attack on the epistemological remnants of the doctrine of original
sin. It is finitude—the limits of our epistemological equipment—rather
than constitutional malformation that makes us subject to error. Kant
complicated the Enlightenment story: self-critical rationality can recog-
nize but not prevent its disposition to self-deceptive illusions. Ironically, it
is the fundamental project of rationality—articulating the conditions that
make experience possible—that lures it to treat its postulates as if they
were possible objects of experience.
We cannot avoid self-deception. Even open-eyed ambivalence is sub-
ject to the self-deceived conviction that although it is conflicted, the
appropriate attitude will emerge in the right way at the right time. But
we should not—and cannot—wish to do without the active, self-induced
illusions that sustain us. Nor can we do without second-order denials that
they are illusions, the second-order and regressive strategies that we self-
deceptively believe rationalize our various self-deceptive activities. The
question is: how can we sustain the illusions essential to ordinary life,
without becoming self-damaging idiots? Are there forms of user-friendly
self-deception that do not run the dangers that falsity, irrationality, and
manipulation are usually presumed to bring?
The phenomena of self-deception are extremely various: they encompass

an arbitrarily selected section of a spectrum of closely related activities
of ritualized forms of self-manipulation; their identification presupposes
theories about normal patterns of perceptual, emotional and eviden-
tial salience, norms of rationality and transparency. We distinguish self-
deception from its cousins and clones—compartmentalization, adaptive
denials, repressed conflicts and submerged aggressions, false conscious-
ness, sublimation, wishful thinking, suspiciously systematic errors in self-
reflection—in whatever ways sustain our favorite theories.
2. After having raised the paradox of analysis in the Meno, and come to the brink of
skepticism, Socrates says: “we shall be better, braver and more active if we believe we
should inquire than if we believe we cannot discover what we do not already know. That is
something for which I am ready to fight in word and deed to my utmost ability” (86B).
And there is an evaluative element as well: the hidden politics of the

attribution of self-deception and false consciousness masks their frequency
and advantages. When we deplore what we regard as misplaced loyalty
or highly focused concentration that resists expansion or correction, we
pejoratively classify it as self-deception. But when we admire persistent
and dedicated single-minded attention that systematically resists the dis-
traction of fringe phenomena, we call it courage or purposeful resolution.
The person who does not have our favored reactions is open game for the
charge of self-deception, if not of a more serious form of psychological
abnormality, or worse, a culpable form of political subversion.
To be sure, if the pronouncements of common opinion and ordinary
speech are at all clearly identifiable and reliable, there are constraints
and directions on the analysis of self-deception. Like many of the con-
cepts that concern us as persons and philosophers (“self,” “belief,” “con-
flict” and even “rationality”), self-deception elusively moves between
latitudinarian ordinary speech and a strict, theory- and value-depen-
dent technical vocabulary.3 Enlightenment philosophers attempting to
explain the possibility of knowledge focus on the primacy of cognition
and construe their analyses of other psychological activities in the terms
set by that focus. To be sure, all the phenomena must be accounted for
in one way or another: but the exigencies of elegant theory construc-
tion play a large role in categorizing and describing fringe phenomena
that are not, in the first instance, a philosopher’s primary explanatory
concern. An ambitious philosophy of mind that is designed to conjoin
and support a theory of knowledge does not initially propose a theory
of self-deception or akrasia. Such a philosophy classifies these as deviant
phenomena and explains them in the terms that best suit the directions
of its primary theory.
If we characterize self-deception narrowly, as requiring the strict iden-
tity of deceiver and deceived about beliefs in propositional form, the
phenomena of self-deception seem to evaporate. After all, the conditions
for strict personal identity are so stringent as to cast doubt on the con-
tinued temporal identity of the self, let alone the identity of a self delib-
erately lying to itself. As strict constructionists working with a technical
vocabulary, we may get some understanding of the mind as an epistemic
instrument, but little understanding of its psychosocial functioning and
the popularity of other presumptively deviant activities like weakness of
will and the irrational conservation of emotions. If, on the other hand, we
characterize these phenomena inclusively, with the broad latitudinarian
3. See A. O. Rorty, Mind in Action (Boston: Beacon Press, 1988); and A. Mele,
Irrationality: An Essay on Akrasia, Self-Deception and Self-Control (Oxford: Oxford
University Press, 1987),; D. Pears, Motivated Irrationality (Oxford: Oxford University Press,
1984); M. Martin, ed., Self-Deception and Self-Understanding (Kansas: University of Kansas
Press, 1985); and Jon Elster, Sour Grapes (Cambridge: Cambridge University Press, 1983).
hand that encompasses common practice and common speech, the phe-
nomena that appear on the fringes of our presumptive rationality play a
significant role in virtually all our activities. Beyond the constraints set
by constructing a comprehensive theory of intellectual and psychological
functioning—one that explains extremely diverse cognitive, motivational,
and affective phenomena—there is no fact of the matter about whether
we should be strict or latitudinarian constructionists about the criteria
for the identity of the deceiver and the deceived or about the conditions
that identify cases of deception. Because we typically position ourselves
dialectically, emphasizing the conceptions that have been neglected by
our immediate predecessors, we can expect a continuous (re)cycling of
latitudinarian and strict characterizations of self-deception.
We are in the awkward position of stipulating definitions that will
satisfy our technically exacting colleagues in the cognitive sciences,
while also carrying on with what passes for common sense and ordinary
language. In analyzing and evaluating self-deception, we are engaged
in the method of reflective equilibrium, attempting to balance our
(shared) considered judgments and practices with our principle-laden
theories, as if our ordinary judgments and practices were not already
theory-laden.4
Like deception, self-deception is a species of rhetorical persuasion; and

like all forms of persuasion, it involves a complex, dynamic, and coopera-
tive process. Successful deceivers are acute rhetoricians, astute seducers
who know how to co-opt the psychology of their subjects. They begin
with minute and subtle interactions designed to establish trust, with a
manner of approach, certain gestures and intonation patterns, intimations
of directed and redirected attention. Astute deceivers like Iago engage
the cooperation of their victims. Othello’s psychology—his sensitivity, his
pride, his sense of being a stranger—was a collusive instrument in his
being deceived, and eventually in his being self-deceived. These strate-
gies reveal the political complexity involved in drawing the boundaries
between deception and socially induced self-deception. Deception and
self-deception are not merely detached conclusions of invalid arguments:
they are interactive processes with a complex cognitive and affective
aetiology.
4. Ordinary language is protean in this area: it has incorporated the terminology of

psychoanalysis and popular cognitive science. And as it becomes increasingly cosmopolitan,
it adds “mauvaise foi” and “false consciousness.” We can expect that considered judgments
derived from French (“Je me trompe” for “I made a mistake,” “Je m’en fiche” for “I don’t
care”) would not coincide with those influenced by languages that are less generous with
reflexive pronouns.
3.1 What Self-Deception Is Not

It is illuminating to track self-deception negatively, characterizing its vari-
eties by noting what it is not. By exposing common misconceptions about
self-deception, we shall arrive at a better understanding of its dynamics
and its popularity.
1. Self-deception is typically not episodic: it rarely occurs as a single,

momentary event, a kind of epistemic sneeze. The popularity of self-
deception is not explained by its episodic propositonalized structure but
by its functional activity as a magnetizing disposition. A disposition is
magnetizing or tropic just when “it promotes and even constructs the
occasions that require its exercise.”5 For instance, a person who self-
deceptively denies the estrangement of her affection typically does not
await the occasion to affirm or proclaim it. Her self-deception consists
largely in her active disposition to produce the occasions—the scenarios
and events—that elicit the conventional expressions of affection: a term
of endearment, a caress. Similarly, the Roman Catholic who denies that
she has lost her faith sustains her self-deception by following routine hab-
its, attending Mass, continuing the rituals of religious observance. In both
cases, the evidence for self-deception is a pattern of behavior: the caress is
unconvincing; participation in the ritual of the service is wooden. But one
abstracted caress or absent-minded Credo doesn’t mark a self-deceiver.
Self-deception is characterized by a continued and complex pattern of
perceptual, cognitive, affective, and behavioral dispositions.
2. Self-deception is typically not a solitary activity. Like other inten-
tional activities, it works through sustaining social support. 6 As standard
ordinary beliefs are elicited and reinforced by our fellows, so too are our
primary self-deceptive strategies. The canny self-deceiver puts herself in
situations where her deflected attention will be strongly supported by her
fellows. “How wonderful that you are beginning your Spiritual Retreat
(or going to Lower Slobovia),” the world says to the uncertain and fright-
ened traveler. Though she may be aware that she is more apprehensive
than pleased by the prospect of her journey, she attempts to block her
resistance by using conventional social forms to distract or submerge her
attention.
3. Self-deception need not involve false belief: just as the deceiver
can attempt to produce a belief that is—as it happens—true, so too a self-
deceiver can set herself to believe what is in fact true. A canny self-
deceiver can focus on accurate but irrelevant observations as a way of
denying a truth that is importantly relevant to her immediate projects.
5. See “The Two Faces of Courage,” in Rorty, Mind in Action, 301.

6. See William Ruddick, “Social Self-Deceptions,”, 118–134, and Ron Harre, “The
Social Context of Self-Deception,” 63–97, in McLaughlin and Rorty, Perspectives on Self-
Deception.
Moreover, self-deception need not involve any belief at all: the pro-
cess and the outcome can be protointentional or subdoxastic. 7 When
someone systematically deflects the natural direction of her gaze, ignoring
phenomena that she would normally find salient, her ignorance can be an
instance of self-deception as well as an instrument designed to achieve it.
Systematic, persistent resistance to correction can be internal to the pro-
cesses of believing: it can indicate the functional role of a relatively trivial
belief or a subdoxastic intentional disposition, rather than its epistemic
status.8
Further: stylized or ritualized actions—culturally specific actions that
conventionally express complex attitudes—can deceive. We adopt certain
postures and gestures to show a self-confidence we do not actually pos-
sess. An inclination of the head, a way of gazing, an intonation pattern can
deceptively suggest intimacy.9 Similarly, self-deception can be expressed
in gesture and action: the gestures of an aging coquette—the head at an
angle, the languorous eyes, the flirtatious smile—are not only designed to
help create and sustain an illusion: they can also be its primary expression.
While the beliefs that are implicated in such action—beliefs that such
gestures retain whatever charm they might once have had—are some-
times mistaken without being self-deceived, the coquette’s anxious look
in the mirror as she applies layer after layer of lipstick and rouge indicates
that she also knows better.
4. Self-deception need not focus on important matters: it can range
from the momentous to the minute, from the sublime to the ridiculous.
It can focus on the primary projects of a life (those of a politician or a
parent) . . . or on a new hairdo.
5. Self-deception need not be self-centered. To be sure, self-deception
is—along with other epistemic and psychological attitudes—explained
largely by the deceiver’s system of beliefs, habits, and desires; but although
7. See Annette Baier, “Ignorance and Self-Deception,” in Deception and the Self,
ed. R. Ames and W. Dissanayake (Cambridge: Cambridge University Press, 2003);
and M. Johnston, “Self-deception and the Nature of Mind,” in McLaughlin and Rorty,
Perspectives on Self-Deception, 37–55. Since many preintentional activities can sometimes
function in a fully intentional form, I prefer to speak of protointentional rather than
subintentional activities.
8. Following the model of analyses of justified belief, analyses of self-deception
typically specify necessary and sufficient logically distinct conditions—reified as
independent psychological states—whose conjunctive present themselves as casesof self-
deception. If the conditions of justified belief can be condensed in one activity, so can
those of self-deception. “The same liberty may be permitted to moral, which is allowed
to natural philosophers; and ‘tis very usual with the latter to consider any motion as
compounded and consisting of two parts separate from each other, tho’ at the same time
they acknowledge it to be in itself uncompounded and inseparable”; Hume, A Treatise of
Human Nature, 493.
9. See Bruce Wilshire, “Mimetic Engulfment and Self-Deception,” in McLaughlin
and Rorty, Perceptions on Self-Deception, 144–169.
it is of course always by the self, self-deception is not on that account

always for or about the self: a person can be self-deceived about the hon-
esty of her distant political allies or opponents.
Indeed, the individual need not always initiate his self-deception: Like
the members of any sports team, the president’s cabinet can collectively
acquire grandiose attitudes that they could not sustain as individuals.
Affected by one another’s influence, by the luxurious appointments of
the cabinet room, and supported by the army of their secretaries and
assistants, they so collude in magnifying one another’s tendencies to self-
importance that the memoranda on which they consensually agree are
stronger than the views that they would accept individually, in isolation.
And yet it was as distinctive individuals that they participated in the work
of the cabinet.
6. Self-deception need not be motivated by a desire or a wish. A man
who self-deceptively believes that his wife’s professional success far out-
shines his own might be moved by a chronic, painful, envious disposition,
rather by than a desire for her flourishing.
Indeed, self-deception is not always directly motivated. Like many of
psychological activities, it can continue as an entrenched habit long after
its original impetus has been extinguished. The nervous novice teacher
who self-deceptively ignores the boredom of her students can retain
the habit of ignoring their reactions long after she has become a self-
confident and even self-important but still boring teacher. We can also
acquire specific self-deceptive habits in just the same way that we have
imitatively acquired other psychological and intellectual habits. Fearful
about their health, our parents self-deceptively ignored or denied their
ailments. Without the same fears, we may have acquired the same strate-
gies of denial.
7. Even when manipulative deception is morally suspect, its outcome
is not always harmful. Indeed, deception and self-deception are often
benevolently and insightfully motivated. By convincing themselves that
a desired self-transformation is within relatively easy reach, canny self-
improvers can use self-deception as an energizing instrument.
3.2 Strategies of Self-Deception

Clever deceivers rarely tell outright falsehoods. It’s too risky. The art of
deception is closely related to the magician’s craft: it involves knowing
how to draw attention to a harmless place, to deflect it away from the
action. Deeply entrenched patterns of perceptual, emotional, and cogni-
tive dispositions serve as instruments of deception. A skilled deceiver is
an illusionist who knows how to manipulate the normal patterns of what
is salient to their audience. He places salient markers—something red,
something anomalous, something desirable—in the visual field, to draw
attention just where he wants it. The strategy of perceptual self-deception
is identical: the trick is to place oneself where patterns of salience are
likely to deflect attention away from what one does not wish to see. The
best way for a gambler to deceive herself—to avoid noticing her lover’s
roving eye—is to schedule their assignations at the casino or the racetrack
instead of at the disco.
Opacity, vagueness, and overdetermination are the deceiver’s friend.10
Just as we use the ambiguity of polite, ritualized speech to mislead others
(“I had a wonderful time.” “I’ve been hoping to run into you so we could
arrange to have lunch.”), so we fuse the multiple functions of speech acts
when we talk to ourselves. In hopes of levering ourselves to our desks, we
gloss a vague thought as if it were a firm intention; we say “I’ll spend the
weekend finally getting to all those letters I must write.” The more pub-
licly such pronouncements are made, the more force the lever can exert.
Any experience is open to an indefinite number of true and even rel-
atively salient descriptions. To recommend a brash and hostile student,
we call attention to her energetic initiative in discussion. In the interest
of maintaining one’s loyalty to one’s unreliable or treacherous friends, we
praise their originality. While such cases do not involve lying, we typically
do intend to deceive by distraction. Of course, we might well have a sec-
ond-order policy that rationalizes and justifies strategies of this kind. But
they are none the less deceptive for having been rationalized and justified.
Shifting the level of generality of descriptions and explanations is also an
excellent strategy of deception and self-deception. To deflect attention from
the sordid, exasperating, and frustrating details of our major projects—par-
enting, teaching, political action—we move to general abstractions, lump-
ing these details together under the heading “No pain, no gain. It’s all worth
it in the end,” forgetting that when we are making important decisions,
it is often this—whether there is something about the activity that out-
weighs the trouble it brings—that is in question. Or we move in the other
direction: we can deceive someone (including ourselves) into accepting an
undesirable job by focusing on a few genuinely attractive details, drawing
attention away from a general, all-things-considered evaluation.
Second-order policies that legitimate specific self-manipulative strate-
gies are sometimes also canny instruments of self-deception, as follows.
1. We rationalize compartmentalization as a generally efficient and effi-

cacious way of advancing the diversity of our competing and potentially
conflicting projects. (But we are half aware that we don’t—indeed that
we can’t—compartmentalize as thoroughly as our projects require. If the
subsystems don’t actually overlap, they are certainly in close communica-
tion. However great their differences, Dr. Jekyll and Mr. Hyde both knew
their way home. More significantly, if Dr. Jekyll hadn’t been so righteous,
Mr. Hyde might well not have been so venial.)
10. See Iris Murdoch, “The Idea of Perfection,” in The Sovereignty of the Good (London:
Routledge Kegan Paul, 1970).
2. We often justify epistemically dubious cognitive, emotional, and

behavioral habits by policies assigning high priority to the social utility
that such habits are meant to serve. (But in their details, such policies are
often manifestly no more defensible than the strategies they are meant to
support. Moreover, a person’s self-deceptive strategy is a way of specify-
ing his ends rather than a method for achieving them. For instance, Pas-
cal’s wager—the gamble of faith—can express and reinforce rather than
assuage the horror of infinite spaces.)
3. We construct general philosophical theories about human nature,
specifying intrinsically valuable activities or activities that we declare to
be “essential to a fully human life” as a way of helping ourselves through
some of our more difficult and onerous activities. Or we invent something
we call our identity, resting our self-respect on our engaging in its projects,
independently of any other measure of their merits. (In such cases, it is
typically not the theory or the commitment that is self-deceptive, but the
belief that philosophic theories or projects of identity-engagement justify
or ground rather than express our fundamental choices.)11
4. Recognizing the distance between our best intentions and the activ-
ities that actually engage us, between the expected and the actual out-
comes of our activities, we deflect our attention away from the horrors of
contingency, away from the moral luck that attends everything we do. We
characterize what we are pleased to think happens for the most part. (But
we disguise from ourselves the extent to which contingency surrounds
intentional activity, and the extent to which “standard or normal” experi-
ence embeds questionable but self-fulfilling normative claims.)
5. For the sake of promoting cherished ends, we rationalize self-
manipulative strategies designed to produce beliefs, desires, or habits that
we do not initially possess.12 (But we are often self-deceived about the strength
of our commitments; and when responsibility is weighty, we have reason to
magnify or diminish the indeterminacy of the power of our agency.)
Is all this necessarily self-deceptive? Can’t we maintain and indeed

justify tactfully manipulative strategies without actually deceiving our-
selves? We often deliberately mimic confidence and wholeheartedness
in the hope of acquiring them; and indeed we can sometimes succeed
in internalizing an attitude that was initially only mimetically expressed.
But even the most successful of such manipulations often preserve traces
of the original attitude in disguised or repressed ambivalence: a sarcastic
remark, a verbal slips, a taut and guarded manner, submerged hostility.
11. See Sartre, Sketch for a Theory of the Emotions (London: Methuen, 1962), and
Existentialism and Humanism, Anti-Semite and Jew (New York: Grove Press, 1948).
12. See William James, “The Will to Believe,” in; Pascal, Pensees 13; Bas van Fraassen,
“The Peculiar Effects of Love and Desire,” in McLaughlin and Rorty, Perspectives on Self-
Deception, 234–261.
Ambivalence of this kind is not necessarily self-deceptive: but the more

we are intent on achieving a self-transformation, the more likely we are
to deny traces of older attitudes. In any case, since Method acting requires
finding the projected character within oneself, the profoundly diffident
are ill equipped to help themselves to confidence by that method, par-
ticularly when their lines are not provided by a playwright. As a strategy
for self-transformation, relatively contained, temporary self-deception is
often more efficient and effective than Method acting.
Taking a very different tack, we might, in the interests of high-minded
Enlightenment, attempt to persuade our fellows that openly acknowledged
ambivalence may be at least as reliable as forced wholeheartedness. Practi-
cally speaking, however, we’re often better served by self-deceptively under-
taking to be wholehearted (by whatever the going standards are) than by
attempting to persuade our fellows that ambivalence is a mark of reliability.
All things considered, we are probably better served by giving way to the
irrational desire for self-deceptive wholeheartedness than by attempting
the quixotic and self-deceptive project of curing our fellows’ irrationality.
3.3 The Benefits of Self-Deception

Self-deception is sometimes construed as an effective measure against
the despair of global skepticism. To be sure, we have, as Bas van Fraassen
has argued, other ways of dealing with generalized uncertainty about the
worth of our projects, about the reliability of those on whom our welfare
depends.13 Van Fraassen charts the advantages of the voluntarist strategy
of affirming the trust or faith that he argues is implicit in every observa-
tion. Hume omits the voluntarist step: he observes that we just naturally
do believe beyond strict evidence; we trust beyond strict proof of reliabil-
ity; we actively persist in our manifestly questionable projects. Despite our
philosophic doubts about the continued existence of objects or the legiti-
macy of philosophic arguments, hunger guides us out of the study and out
of skeptical philosophy at mealtimes; and after dinner, we are sociable and
even affectionate, despite our clear-eyed assessment of the foibles and fol-
lies of our fellows. Some interpreters take Hume’s solution to mark a final
ironic skeptical turn: the operations of nature are identical with those that
philosophers call self-deception. Others see it as evidence of Hume’s prag-
matic naturalism: nature has so attuned us that what some philosophers
call self-deception is actually a trustworthy sign of the natural health of the
mind. At this point, we’ve returned to the rhetorical politics of philosophi-
cal terminology. The result is the same: some forms of self-deception are by-
products of the standard operations of belief and the imagination. Although
they run serious dangers, we could not do without their contributions to
our intellectual and psychological activities. But it is natural psychology
13. Van Fraassen, “Peculiar Effects of Love and Desire.”

rather than a second-order rational policy that prompts accepting the self-
deceptions that accompany standard modes of imagining and believing.
We’d engage in these activities even if we didn’t approve of our doing so.
The more interesting forms of self-deception are local rather than global.
Without some species of self-deception, our dedications, our friendships,
our work, our causes would collapse. In deciding to have children, we
ignore the travails of parents, obliterating our otherwise keen awareness
of the typical relations among parents and children; in devoting ourselves
to writing philosophy, we conveniently forget how little philosophy we
are willing to read; in the interest of sanity and joy, we sidestep our deep
ambivalences about our kith and kin.
The benefits of individual self-deception are obvious to its practitioners;
the benefits of its socially induced forms are often more compelling.14 The
appearance of earnest and wholehearted conviction about one’s projects—
defending a philosophic position, proposing a curricular reform, raising
funds for a cherished cause—is commonly taken as an indication of trust-
worthy reliability. Disguising and submerging the ambivalence that is natu-
ral to most of our enterprises not only brings us the energy, verve, style, and
ease that successful action requires; it also helps to assure the social coop-
eration that is equally essential to our individual and collective projects. A
good deal of the polite conversation of social life—the public description of
the joys of our social roles and functions (friend, mother, teacher, scholar)—
channels and streams us to play our parts without the mess, confusion, and
upheaval that would occur if we openly expressed our natural and sensible
ambivalence about these roles. It is virtually impossible to imagine any soci-
ety that does not systematically and actively promote the self-deception of
its members, particularly when the requirements of social continuity and
cohesion are subtly at odds with one another and with the standard-issue
psychology of those members.15 Socially induced self-deception is an instru-
ment in the preservation of social cooperation and cohesion.
3.4 The Beneficiaries of Self-Deception

Who is served by socially induced self-deception? And who bears the pri-
mary responsibility when an individual’s self-deception depends on social
collusion? It’s no news to post-Hegelian post-Freudian post-Marxists
post-Wittgensteinians that the individual is not always the primary
epistemic agent: like all epistemic activities, self-deception occurs within
a social frame, one that not only defines but actively channels patterns
14. See “Some Social Uses of the Forbidden,” Psychoanalytic Review 12 (Feb.–Mar., 1972).
15. Since they do not involve beliefs in propositional form, such conflicts are not,
of course, technically speaking, contradictions. (See R. Marcus, “Moral Dilemmas and
Consistency,” in C. W. Gowans, Moral Dilemmas [Oxford: Oxford University Press, 1987].)
Other essays in this volume provide a useful background for understanding some of the
motivation for self-deception. See also Festinger, Cognitive Dissonance.
of categorization, salience, and motivation.16 But while we recognize the

social influences on individual belief, we do not have a clear account of
how they occur and where they stop. Locating epistemic responsibility
with the individual—the last in a network of contributory of epistemic
agents—derives from a forced parallel to voluntary behavior. Despite
their repudiating Cartesian philosophical psychology, contemporary epis-
temologists still treat belief as voluntary: the individual is presumed to be
a responsible epistemological agent, capable of identifying—and suspend-
ing assent to—any and all unwarranted beliefs.
Distinctions will help us. We can, to begin with, distinguish (1) the imme-
diate, (2) the primary, and (3) the contributory agents of epistemic attitudes.
(1) The immediate agent of self-deception is the last active link in a causal
chain that generates the work of deception. That work is not always carried
out by individual persons: its agents can be subsystems of the self. Such sub-
systems are extremely various: they can range from subpersonic protointen-
tional perceptual dispositions to internalized idealized group identifications.
Neo-Freudians,17 cognitive psychologists,18 and social theorists19 differ
in their analyses of the components that constitute the self, but they
agree in characterizing it as made up of relatively independent subsys-
tems whose interaction is often only precariously integrated. For them,
the explanation of the phenomena of self-deception lies in our complex
16. See Tyler Burge, “Individualism and Psychology,” Philosophical Review (1986), and
“Intellectual Norms and the Foundations of Mind,” Journal of Philosophy (1987). Burge
argues that the individuations of intentional states essentially refer to social practices.
See also Alvin Goldman, “Varieties of Cognitive Appraisal,” Nous 13 (1979), 22–38, for a
useful discussion of the variety of criteria by which beliefs are assessed.
17. See Freud, “Repression,” and “The Unconscious,” in The Standard Edition of
the Complete Psychological Works of Sigmund Freud (1915), and “Splitting the Ego in
the Service of Defence,” in Standard Edition (1938); R. Schafer, A New Language for
Psychoanalysis (New Haven: Yale University Press, 1976); D. Sachs, “On Freud’s Doctrine
of the Emotions,” in Freud, ed. R. Wollheim (Cambridge: Cambridge University Press,
1974), 101–124; H. Kohut, The Restoration of the Self (New York: International University
Press, 1977); R. Wollheim, The Thread of Life (Cambridge, Mass.: Harvard University Press,
1984).
18. See D. Dennett, “Three Kinds of Intentional Psychology,” in Reduction, Time
and Reality, ed. R. Healy (Cambridge: Cambridge University Press, 1981); D. Davidson,
“Paradoxes of Irrationality,” in Philosophical Essays on Freud, ed. J. Hopkins and R.
Wollheim (Cambridge: Cambridge University Press, 1982), and “Deception and
Division,” reprinted in Action and Events, ed. E. LePore and B. McLaughlin (New York:
Blackwell, 1985); Johnston, “Self-deception and the Nature of Mind”; S. Stich, “Beliefs
and Subdoxastic Systems,” Philosophy of Science (1978), and Fragmentation of Reason
(Cambridge, Mass.: MIT Press, 1990).
19. For an account of the distinctive aspects and features of identity, see Alfred Schutz,
Collected Papers (Amsterdam: Martinus Nojhoff), esp. 16–18, 221–22; G. H. Mead, Mind,
Self and Society (Chicago: University of Chicago Press, 1934), esp. 144–45, 149–52; A.
Rorty and D. Wong, “Aspects of Identity and Agency,” in Identity, Character and Morality,
ed. O. Flanagan and A. Rorty (Cambridge, Mass.: MIT Press, 1990), 121–133.
psychological organization: the immediate agent and presumptive benefi-

ciary of self-deception is a subsystem of the self. Social psychologists join
many neo-Freudians in identifying the subsystems of the self as the inter-
nalized representatives of social personae who have formed—and who
continue to influence—individual psychology.
In the interest of avoiding such regressive homuncular explanations, many
cognitive psychologists have introduced subpersonic subsystems, capable of
nonpurposive but intentional operations. When a subsystem is (by some
measure) central to an aspect of a person’s identity, its strategies are consid-
ered to be self-deceptive, though the self neither is nor has a central panopti-
cal scanner or manipulator. There is no need for reductive zeal here, no need
to determine—as if there were a theory-neutral fact of the matter—whether
the subsystems engaged in self-deception are all homuncular or subhomun-
cular, whether they are all intentionally deceptive or subintentionally mis-
leading, or whether self-deception reduces to subsystem deception.20 When
the deceiving and the deceived subsystems are interdependent extensionally
intersecting “parts” of a psychobiological individual, the problem of whether
self-deception is coherent becomes a verbal puzzle.
(2) The immediate agents of epistemic attitudes need not, of course,
be its primary agents. Typically, the primary agents of deception and self-
deception are its presumptive beneficiaries. As the primary agent of
Othello’s eventual self-deception, Iago orchestrated and directed the
immediate and the contributory agents of the deceptions that generated
Othello’s eventual self-deception. Not only Emilia and Cassi but also
Othello’s own subpersonae—the Moor’s honor, the soldier’s quick reac-
tions—were brought into play as agents of the work of deception.
(3) Like other epistemic attitudes, successful self-deception typically
requires the collusion of contributory agents. Whether it be an individual
or a subpersonic system, the last agent in a causal chain of deception
could not do its work without antecedent and conjoined support. It is
extremely difficult to sustain self-deception without a little help from our
friends, often rendered by observant, but tactful silence. Once begun, the
process of Othello’s self-deception would have been difficult to sustain
without the manipulated contributions of Emilia and Cassio.
Active cooperation in self-deception is more readily assured when it
brings widespread secondary gains. Normal science is, for instance, served
by training scientists to follow a conservative epistemic policy, one that
makes them susceptible to self-deceptive denials of evidence contrary to
dominant theories.21 But collaborative contributions to self-deception
need not always serve larger gains: sometimes self-deception just happens.
20. See Johnston, “Self-Deception and the Nature of Mind,” and MacLaughlin, “Self-
Deception and the Nature of Mind,” for discussions of the presumed incoherence of self-
deception is incoherent and its reduction to other-deception.
21. See Adam Morton, “Partisanship,” in McLaughlin and Rorty, Perspectives on Self-
Deception, 212–235.
A society can systematically contribute to the self-deception of its mem-

bers even when there are few benefits from such patterned misapprehen-
sion. The explanation of self-deception is often global and structural: it
does not lie in its occasions, but in its being an unintended by-product of
functional activities.22
3.5 Why the Best Solution Is Not Available to Us

It would be self-deceptive for us to suppose that we could—by some act
of will, commitment, or cultural enchantment—effectively transcend or
transform the intellectual history that expresses the conflicts arising from
the minutiae of our complexly layered social practices. All of us are, after
all, Socratic inquirers, bred in the muscles and the bones of our minds. But
we are many other things as well: Byronic heroines, aficionadas of Japanese
movies, and Nietzschean self-creators. This does not mean that we are
swimming in the Sargasso Sea of the postmodern condition, equally hos-
pitable to all the historical bits and pieces that constituteus. Our Socratic
personae will—and should—insist on raising stern critical questions about
how it all hangs together, bitterly remembering that Socrates persisted in
asking questions even when he was not confident that they were answer-
able. With the background help of Spinoza and Hegel, it was finally Freud
and (in a surprising way) Marx who gave us the fullest, clearest formula-
tions of what the Enlightenment requires of self-knowledge.
The virtues required for astute self-deception are those required for
astute and righteous lying: deception in the right way at the right time for
the right reason. But what does phronesis about self-deception require?
How do we determine the properly attuned balance between persistence
and fallibility, one that deflects correction as long as closure is benefi-
cial, generating self-deception in love and work but not in self-defense?
In principle, an acute philosophical logician could formulate a context-
sensitive set of policies for determining the cutoff points for beneficial
self-deception, specified for distinctive measures of benefits, distinctive
agents and beneficiaries, appropriate time spans. But while the theorist
can distinguish benign from maladaptive cases of self-deception and
other irrational psychological activities, the practitioner is not, in the very
nature of the case, in a position to do so. If the practitioner always casts
herself as theorist, scanning and testing her psychological activities for
their legitimacy, she would rarely be in a position to benefit from their
exercise. Complex psychological activities best function at a precritical
and prereflective automatic or autonomic level. The utility of many of our
presumptively self-deceptive responses—like those moved by fear and
trust, for example—depends on their being relatively undiscriminating,
operating at a deeply entrenched habitual precritical level.23
22. See Elster, Sour Grapes.

23. See “Fearing Death,” in Rorty, Mind in Action, 202–7.
3.6 Ambivalence in the Service of the Enlightenment

Having argued that self-deception is inevitable and distinguished its layers
and beneficiaries, have we joined the ranks of postmodern social construc-
tionists? Certainly not. Masked as a presumptively egalitarian attitude
to the various personae of the self, a laissez-faire attitude toward self-
deception runs the danger of giving intrapsychic power politics full and
unchecked play: it endorses the actions of the self’s most powerful rather
than those of its most justified personae. Self-deception is only as good as
the person who has it.
If the difference between deception and self-deception is arbitrary, if
the deceived typically collude in their deception and the self-deceived
depend on the complicity of their fellows, the allocation of responsibility
for the harms of deception seems arbitrary. We might well be uneasy that
such an openhanded, latitudinarian way of subscripting the various agents,
benefits, and beneficiaries of self-deception runs the danger of blaming
the victim. Self-deception does not monitor its own use: it doesn’t know
when or where to stop. It is specifically constructed to ignore and resist
correction. The danger of self-deception lies not so much in the irrational-
ity of the occasion as in the ramified consequences of the habits it devel-
ops: its obduracy and its tendency to generalize.
But this is equally true of many of our other, more superficially ratio-
nal intellectual activities.24 Consider the various Platonic recommenda-
tions for dialectical analysis offered in the Sophist and the Statesman: the
method of division is designed to construct a taxonomy of genus, species,
and varieties to “catch the meanings of general terms.” When that method
is astutely used, it charts the geography of a conceptual field. But it is
clear—it was certainly clear to Plato—that when the method of division is
globally or grossly applied, when it is entrenched as a primary and exclu-
sive mode of analysis, it can be deceptive and even self-deceptive. Like
Socratic self-knowledge, the Platonic method of division is only as good
as the mind that uses it.
It was for reasons like these that Descartes wanted to find a method so
simple that any mind could use it, a method that presupposes no ability
or knowledge beyond the capacity to test its ideas for their logical consis-
tency, using reductive proofs, moving only a step at a time. Here again, a
method that is rational, if any method is, brings the fruits of rationality—a
clearly demonstrated knowledge of the world—only when it is supple-
mented by a wide range of other, shadier intellectual and psychological
activities. Without the generous support of suspect nonrational intellectual
and psychological activities, Descartes’s method is sterile and useless.
24. See Roy Sorensen, Thought Experiments (Oxford: Oxford University Press, 1992).
Sorensen remarks that the standard modes of argumentation have their shortcomings as
well as their strengths. He recommends what he calls a “diversified portfolio” of argument
forms.
Even though its authority rarely carries executive power, it is the active,
permanent possibility of asking critically evaluative questions that preserves
us from dangerous folly. “When is self-deception self-defeating? What is really
beneficial and to whom?” There are, to be sure, a variety of context-depen-
dent criteria for such evaluations; and each subsystem has its own claims for
special privilege. Still, at any given level, for any subscripted measure of util-
ity or rationality, intrapsychic might does not make intrapsychic right or even
intrapsychic utility. Socratic inquiry—actively pressing for self-critical evalu-
ation—is the only safeguard against the damaging uses of self-deception, or
indeed of any of our intellectual or psychological devices.
In evaluating the self-deception of our friends and enemies, in retrospec-
tively gauging our own, we are directed by judgments about the merits of
the ends it serves, as well as judgments about whether those ends could
have been better served by other means. In making such evaluations, we
need to think laterally as well linearly, systematically as well as episodically.
We need to consider the global effects of all our epistemic and psychologi-
cal activities—their addictive qualities as well as their immediate benefits.
When they are successful, psychological and intellectual activities typically
tend to become rapidly entrenched, ramified, and generalized.
But we have very little latitude in monitoring our psychological activi-
ties, and still less in forming them. Our epistemological strategies become
habitual before we are aware of their patterns and consequences. As phi-
losophers, the best thing we can do about self-deception is what we should
do about our other psychological and intellectual activities: engage our-
selves in the Stoic task of understanding the minute details of its operations.
Since we are highly susceptible to socially induced self-deception, the wis-
est practical course is to be very careful about the company we keep. But
it is no easy task to determine where our best protection lies. On the one
hand, prudence counsels avoiding the company of charismatic rhetoricians
who might mislead us. On the other hand, it’s not easy to identify epistemic
seducers, particularly when we benefit from hospitality to a wide range of
opinions, each with a distinctive critical perspective on our favorite illusions.
Unfortunately, self-deception is just the thing that prevents us from seeking
its best therapy: it doesn’t know when to expand, and when to limit its epis-
temological company. Fortunately, we have many other kinds of reasons for
being astute about the company we keep. With luck, a canny self-deceiver’s
other psychological and intellectual habits—a taste for astringency and a
distrust of hypocrisy, for instance—can prevent the wild imperialistic ten-
dencies of self-deception from becoming entrenched and ramified.
But that is a matter of luck; and as we know, ambivalence is the best
attitude toward luck.
14
Have I Unmasked Self-Deception

or Am I Self-Deceived?
Alfred R. Mele
What is self-deception? And how does garden-variety self-deception

happen? Some philosophers believe that the first question—a conceptual
one—is answerable by a full-blown conceptual analysis, something that
requires a statement of individually necessary and jointly sufficient con-
ditions for self-deception. Although, since the early 1980s (Mele 1982),
I have written more than a few articles on self-deception and a small book,
I have never offered such an analysis of self-deception. (The title of that
book, Self-Deception Unmasked [Mele 2001], helps to explain the title of
this article.) This is not because I am opposed to conceptual analysis: I am
not. Rather, I find the second question—an explanatory one—much more
interesting; it can be answered in the absence of a conceptual analysis of
self-deception; and defending an analysis of any complicated concept is a
painstaking process that tends to put off many readers.
Some philosophers who seem pretty happy with the answer I develop
in Self-Deception Unmasked to the explanatory question are worried about
the conceptual question and about what I count as self-deception. Some
of their worries are the focus of this article. In sections 1 and 2, I set the
stage for an examination of these worries. In section 3, I examine them.
1 HOW DOES SELF-DECEPTION HAPPEN?
According to a traditional view, self-deception is an intrapersonal ana-

logue of stereotypical interpersonal deception.1 In the latter case, deceiv-
ers intentionally deceive others into believing something, p, and there is
a time at which the deceivers believe that p is false while their victims
falsely believe that p is true. If self-deception is properly understood on
1. For citations of this tradition in philosophy, psychology, psychiatry, and biology, see
Mele 2001, 125 n. 1. Stereotypical interpersonal deception does not exhaust interpersonal
deception.
260
Have I Unmasked Self-Deception or Am I Self-Deceived? 261
this model, self-deceivers intentionally deceive themselves into believing

something, p, and there is a time at which they believe that p is false while
also believing that p is true.
In Mele 2001 (and in earlier work, beginning with Mele 1983), I criti-
cize this view and defend an alternative, deflationary view, according to
which self-deception does not entail any of the following: intentionally
deceiving oneself; intending (or trying) to deceive oneself; intending (or
trying) to make it easier for oneself to believe something; concurrently
believing each of two explicitly contradictory propositions. I also argue
that, in fact, ordinary instances of self-deception do not include any of
these things. My data include widespread agreement that various vignettes
count as cases of self-deception. I argue that explaining how the protago-
nists come to believe what they do in these vignettes does not require
appealing to any of the items in the preceding list and that an alternative
style of explanation is more plausible and much more firmly grounded in
relevant empirical work.
Obviously, falsely believing that p in the absence of deception by
anyone else is not sufficient for self-deception. If it were, we would be
self-deceived whenever we make, for example, unmotivated arithmetical
mistakes. That is why motivation figures prominently in the literature on
self-deception.
Elsewhere, I have distinguished between what I call straight and twisted
cases of self-deception (Mele 1999, 2001). In straight cases, which have
dominated the literature, people are self-deceived in believing something
that they want to be true—for example, that their children are not using
illegal drugs. In twisted cases, people are self-deceived in believing some-
thing that they want to be false (and do not also want to be true). For
example, an insecure, jealous husband may believe that his wife is having
an affair despite having only thin evidence of infidelity and despite his
wanting it to be false that she is so engaged (and not also wanting it to be
true that she is). In cases of both kinds, as I have explained in Mele 2001
and will briefly explain here, self-deceivers have motivationally biased
beliefs.
Some illustrations of ways our desiring that p can contribute to our
believing that p in instances of straight self-deception will be useful (see
Mele 2001, 26–27). Often, two or more of the phenomena I describe are
involved in an instance of self-deception.
1. Negative misinterpretation. Our desiring that p may lead us to mis-

interpret as not counting (or not counting strongly) against p data that
we would easily recognize to count (or count strongly) against p in the
desire’s absence. For example, Rex just received a rejection notice on a
journal submission. He hopes that the rejection was unwarranted, and
he reads through the referees’ comments. Rex decides that the referees
misunderstood two important but complex points and that their objec-
tions consequently do not justify the rejection. However, the referees’
criticisms are correct, and a few days later, when Rex rereads his article
and the comments in a more impartial frame of mind, it is clear to him
that this is so.
2. Positive misinterpretation. Our desiring that p may lead us to inter-
pret as supporting p data that we would easily recognize to count against
p in the desire’s absence. For example, Sid is very fond of Roz, a college
classmate with whom he often studies. Because he wants it to be true
that Roz loves him, he may interpret her declining his invitations to
various social events and reminding him that she has a steady boyfriend
as an effort on her part to “play hard to get” in order to encourage Sid
to continue to pursue her and prove that his love for her approximates
hers for him. As Sid interprets Roz’s behavior, not only does it fail to
count against the hypothesis that she loves him, it is evidence that
she does love him. This contributes to his believing, falsely, that Roz
loves him.
3. Selective focusing/attending. Our desiring that p may lead us to fail
to focus attention on evidence that counts against p and to focus instead
on evidence suggestive of p. Beth is a twelve-year-old whose father died
recently. Owing partly to her desire to have been her father’s favorite,
she finds it comforting to attend to memories and photographs that place
her in the spotlight of her father’s affection and unpleasant to attend to
memories and photographs that place a sibling in that spotlight. Accord-
ingly, she focuses her attention on the former and is inattentive to the
latter. This contributes to Beth’s coming to believe—falsely—that she
was her father’s favorite child. In fact, Beth’s father much preferred the
company of her brothers, a fact that the family photo albums amply
substantiate.
4. Selective evidence-gathering. Our desiring that p may lead us both
to overlook easily obtainable evidence for not-p and to find evidence for
p that is much less accessible. Betty, a political campaign staffer who
thinks the world of her candidate, has heard rumors from the opposi-
tion that he is sexist, but she hopes he is not. That hope motivates her
to scour his past voting record for evidence of his political correctness
on gender issues and to consult people in her own campaign office about
his personal behavior. Betty may miss some obvious, weighty evidence
that her boss is sexist—which he in fact is—even though she succeeds
in finding less obvious and less weighty evidence for her favored view.
As a result, she may come to believe that her boss is not sexist. Selective
evidence-gathering may be analyzed as a combination of hypersensitivity
to evidence (and sources of evidence) for the desired state of affairs and
blindness—of which there are, of course, degrees—to contrary evidence
(and sources thereof ).
In none of these examples does the person hold the true belief that
not-p and then intentionally bring it about that he or she believes that p.
Yet, assuming that these people acquire relevant false, unwarranted beliefs
in the ways described, these are garden-variety instances of self-deception
(or so I claim).2 Rex is self-deceived in believing that his article was
wrongly rejected, Sid is self-deceived in believing certain things about
Roz, and so on.
We can understand why, owing to her desire to have been the one her
father’ loved most, Beth finds it pleasant to attend to photographs and
memories featuring her as the object of her father’s affection and painful
to attend to photographs and memories that put others in the place she
prizes. But how do desires that p trigger and sustain the two kinds of mis-
interpretation and selective evidence-gathering? It is not as though these
activities are intrinsically pleasant, as attending to pleasant memories, for
example, is intrinsically pleasant.
Attention to some sources of unmotivated biased belief sheds light
on this issue. Several such sources have been identified (see Mele 2001,
28–31), including the following two:
1. Vividness of information. A datum’s vividness for us often is
a function of such things as its concreteness and its sensory,
temporal, or spatial proximity. Vivid data are more likely
to be recognized, attended to, and recalled than pallid data.
Consequently, vivid data tend to have a disproportional influence
on the formation and retention of beliefs.
2. The confirmation bias. People testing a hypothesis tend to search
(in memory and the world) more often for confirming than for
disconfirming instances and to recognize the former more readily
(Baron 1988, 259–65). This is true even when the hypothesis is
only a tentative one (and not a belief one has). People also tend to
interpret relatively neutral data as supporting a hypothesis they are
testing (Trope et al. 1997, 115).
Although sources of biased belief apparently can function inde-
pendently of motivation, they also may be triggered and sustained by
desires in the production of motivationally biased beliefs.3 For example,
desires can enhance the vividness or salience of data. Data that count
in favor of the truth of a proposition that one hopes is true may be
rendered more vivid or salient by one’s recognition that they so count.
2. If, in the way I described, Betty acquires or retains the false belief that her
boss is not sexist, it is natural to count her as self-deceived. This is so even if, owing
to her motivationally biased evidence-gathering, the evidence that she actually has
does not weigh more heavily in support of the proposition that her boss is sexist than
against it.
3. I develop this idea in Mele 1987, ch. 10, and 2001. Kunda 1990 develops the same
theme, concentrating on evidence that motivation sometimes primes the confirmation bias.
Also see Kunda 1999, ch. 6.
Similarly, desires can influence which hypotheses occur to one and

affect the salience of available hypotheses, thereby setting the stage
for the confirmation bias.4 Owing to a desire that p, one may test the
hypothesis that p is true rather than the contrary hypothesis. In these
ways and others, a desire that p may help produce an unwarranted
belief that p.
An interesting recent theory of lay hypothesis testing is designed, in
part, to accommodate self-deception. I explore it in Mele 2001, where
I offer grounds for caution and moderation and argue that a qualified
version is plausible.5 I call it the FTL theory, after the authors of the two
articles on which I primarily drew, Friedrich 1993 and Trope and Liberman
1996. Here, I offer a thumbnail sketch.
The basic idea of the FTL theory is that a concern to minimize costly
errors drives lay hypothesis testing. The errors on which the theory
focuses are false beliefs. The cost of a false belief is the cost, including
missed opportunities for gains, that it would be reasonable for the per-
son to expect the belief—if false—to have, given his desires and beliefs,
if he were to have expectations about such things. A central element of
the FTL theory is a “confidence threshold”—or a “threshold,” for short.
The lower the threshold, the thinner the evidence sufficient for reach-
ing it. Two thresholds are relevant to each hypothesis: “The acceptance
threshold is the minimum confidence in the truth of a hypothesis,” p,
sufficient for acquiring a belief that p “rather than continuing to test [the
hypothesis], and the rejection threshold is the minimum confidence in
the untruth of a hypothesis,” p, sufficient for acquiring a belief that not-p
“and discontinuing the test” (Trope and Liberman 1996, 253). The two
thresholds often are not equally demanding, and acceptance and rejection
thresholds, respectively, depend “primarily” on “the cost of false accep-
tance relative to the cost of information” and “the cost of false rejection
relative to the cost of information.” The “cost of information” is simply the
“resources and effort” required for gathering and processing “hypothesis-
relevant information” (252).
Confidence thresholds are determined by the strength of aversions to
specific costly errors together with information costs. Setting aside the lat-
ter, the stronger one’s aversion to falsely believing that p, the higher one’s
threshold for belief that p. These aversions influence belief in a pair of
related ways. First, because, other things being equal, lower thresholds are
easier to reach than higher ones, belief that not-p is a more likely outcome
than belief that p, other things being equal, in a hypothesis tester who has
a higher acceptance threshold for p than for not-p. Second, the aversions
influence how we test hypotheses—for example, whether we exhibit the
4. For motivational interpretations of the confirmation bias, see Friedrich 1993 and
Trope and Liberman 1996, 252–65.
5. See Mele 2001, 31–49, 63–70, 90–91, 96–98, 112–18.
confirmation bias—and when we stop testing them (owing to our having

reached a relevant threshold).6
Friedrich claims that desires to avoid specific errors can trigger and
sustain “automatic test strategies” (1993, 313), which supposedly hap-
pens in roughly the nonintentional way in which a desire that p results in
the enhanced vividness of evidence for p. I argue elsewhere (Mele 2001,
41–49, 61–67) that a person’s being more strongly averse to falsely believ-
ing that not-p than to falsely believing that p may have the effect that he
primarily seeks evidence for p, is more attentive to such evidence than
to evidence for not-p, and interprets relatively neutral data as supporting
p, without this effect’s being mediated by a belief that such behavior is
conducive to avoiding the former error. The stronger aversion may simply
frame the topic in a way that triggers and sustains these manifestations of
the confirmation bias without the assistance of a belief that behavior of
this kind is a means of avoiding particular errors. Similarly, having a stron-
ger aversion that runs in the opposite direction may result in a skeptical
approach to hypothesis testing that in no way depends on a belief to the
effect that an approach of this kind will increase the probability of avoid-
ing the costlier error. Given the aversion, skeptical testing is predictable
independently of the agent’s believing that a particular testing style will
decrease the probability of making a certain error.
The FTL theory applies straightforwardly to both straight and twisted
self-deception. Friedrich writes:
a prime candidate for primary error of concern is believing as true some-
thing that leads [one] to mistakenly criticize [oneself] or lower [one’s] self-
esteem. Such costs are generally highly salient and are paid for immediately
in terms of psychological discomfort. When there are few costs associated
with errors of self-deception (incorrectly preserving or enhancing one’s self-
image), mistakenly revising one’s self-image downward or failing to boost it
appropriately should be the focal error. (1993, 314)
Here, he plainly has straight self-deception in mind, but he should not stop
there.Whereas for many people it may be more important to avoid acquiring
the false belief that their spouses are having affairs than to avoid acquiring the
false belief that they are not so engaged, the converse may well be true of
some insecure, jealous people. The belief that one’s spouse is unfaithful
tends to cause significant psychological discomfort. Even so, avoiding falsely
believing that their spouses are faithful may be so important to some peo-
ple that they test relevant hypotheses in ways that, other things being equal,
are less likely to lead to a false belief in their spouses’ fidelity than to a false
belief in their spouses’ infidelity. Furthermore, data suggestive of infidelity
may be especially salient for these people and contrary data quite pallid by
6. Whether and to what extent subjects display the confirmation bias depends on such
factors as whether they are given a neutral perspective on a hypothesis or, instead, the
perspective of someone whose job it is to detect cheaters. See Gigerenzer and Hug 1992.
comparison. Don Sharpsteen and Lee Kirkpatrick observe that the “jealousy
complex”—that is, the “thoughts, feelings, and behavior typically associated
with jealousy episodes”—is interpretable as a mechanism “for maintaining
close relationships” and appears to be “triggered by separation, or the threat
of separation, from attachment figures” (1997, 627). It certainly is conceiv-
able that, given a certain psychological profile, a strong desire to maintain
one’s relationship with one’s spouse plays a role in rendering the potential
error of falsely believing one’s spouse to be innocent of infidelity a “costly”
error, in the FTL sense, and more costly than the error of falsely believing
one’s spouse to be guilty. After all, the former error may reduce the prob-
ability that one takes steps to protect the relationship against an intruder.
The FTL theory provides a basis for an account of both straight and twisted
self-deception (Mele 2001, ch. 5).
2. A PROTOANALYSIS AND PROPOSED SUFFICIENT

CONDITIONS
Although I have never offered a conceptual analysis of self-deception,

I have suggested the following protoanalysis: people enter self-deception
in acquiring a belief that p if and only if p is false and they acquire the
belief in a suitably biased way (Mele 2001, 120). The suitability at issue
is a matter of kind of bias, degree of bias, and the nondeviance of causal
connections between biasing processes (or events) and the acquisition of
the belief that p. My suggestion is that someone interested in constructing
a conceptual analysis of entering self-deception in acquiring a belief that p
can start here and try to tease out an account of suitable bias. Some philos-
ophers have tried to do something along these lines, and I will discuss some
relevant work in section 3. Of course, an analysis of entering self-deception
in acquiring a belief that p will not be a complete analysis of self-deception
if, for example, there are other ways of entering self-deception; and, as
I have explained elsewhere, people may also enter self-deception in retain-
ing a belief that p (56–59). Someone who impeccably acquires the belief
that p may later enter self-deception in persisting in that belief. It may be
suggested that if a complete analysis of self-deception is constructable, it is
constructable out of analyses of these two ways of entering self-deception.
Some theorists would definitely reject this suggestion. Robert Audi, for
example, contends that no one who is self-deceived about p has a false
belief that p; rather, self-deceived people have an unconscious true belief
that not-p and—in the absence of a belief that p—sincerely avow that p
(1982, 1985, 1997). (I criticize Audi’s attempted analysis of self-deception
in Mele 1982; I will not do so again here.)
I have also proposed a set of conceptually sufficient conditions for self-
deception. Some philosophers have argued that these conditions are not
sufficient—that a person can satisfy them without being self-deceived.
I discuss that issue in section 3 and reproduce the conditions now.
S enters self-deception in acquiring a belief that p if:

1. The belief that p which S acquires is false,
2. S treats data relevant, or at least seemingly relevant, to the truth value of p
in a motivationally biased way,
3. This biased treatment is a nondeviant cause of S’s acquiring the belief that
p, and
4. The body of data possessed by S at the time provides greater warrant for
not-p than for p. (Mele 2001, 50–51; see Mele 1997, 95)
Brief commentary on these conditions is in order. Condition 1 cap-
tures a purely lexical point. A person is, by definition, deceived in believ-
ing that p only if p is false; the same is true of being self-deceived in
believing that p. The condition in no way implies that the falsity of p has
special importance for the dynamics of self-deception. Motivationally
biased treatment of data may sometimes result in someone’s believing an
improbable proposition, p, that, as it happens, is true. There may be self-
deception in such a case, but the person is not self-deceived in believing
that p or in acquiring the belief that p.7 My discussion in section 1 of
motivated bias and various ways of entering self-deception puts some
flesh on the bones of condition 2. My inclusion of the term “nondeviant”
in condition 3 is motivated by a familiar problem for causal character-
izations of phenomena in any sphere. Specifying the precise nature of
nondeviant causation of a belief by motivationally biased treatment of
data is a difficult technical task; I have provided guidance on the issue
elsewhere (Mele 2001).
The thrust of condition 4 is that self-deceivers believe against the
weight of the evidence they possess. I do not view condition 4 as a nec-
essary condition of self-deception. In some instances of motivationally
biased evidence-gathering, for example, people may bring it about that
they believe a falsehood, p, when not-p is much better supported by evi-
dence readily available to them, even though, owing to the selectivity
of the evidence-gathering process, the evidence that they themselves
actually possess at the time favors p over not-p. In my view, such people
are naturally deemed self-deceived, other things being equal. However,
some philosophers require that a condition like condition 4 be satisfied
(Davidson 1985; McLaughlin 1988; Szabados 1985), and I have no objec-
tion to including condition 4 in a list of jointly sufficient conditions. Natu-
rally, in some cases, whether the weight of a person’s evidence lies on the
side of p or of not-p (or equally supports each) is subject to legitimate
disagreement.
7. People may be deceived into believing something that they are not deceived
in believing (see Mele 1987, 127–28). A might execute a tricky plan for deceiving B
into believing something that, unbeknownst to A, is true. And A might thereby cause B
to believe this proposition, p. Since p is true, B is not deceived in believing it. Even so, it
is plausible that A deceived B into believing it, if A caused B to believe that p partly by
deceiving him into believing some false propositions suggestive of p.
3 SELF-DECEPTION, DESIRE, AND SELF-KNOWLEDGE
If it were to turn out both that my proposed sufficient conditions for self-
deception are not sufficient and that my answer to the explanatory ques-
tion applies only or primarily to phenomena that satisfy my conditions
but fall short of being self-deception, that would definitely be a prob-
lem. But this is not how the conceptual critics of Self-Deception Unmasked
(Mele 2001), whose work is to be discussed here, see things. They are
willing to allow that my answer to the explanatory question applies to
what they count as self-deception. Even so, if the conditions I proposed
are not sufficient for self-deception, some thought should be given to how
to beef them up.
Dana Nelkin argues that “a necessary condition of self-deception is the
desire to believe that p is true, and that this desire causes the self-deceiver
to treat her evidence in a biased way” (2002, 393). She incorporates
this condition into my proposed sufficient conditions for entering self-
deception in acquiring a belief that p in an attempt to provide individu-
ally necessary and jointly sufficient conditions for the phenomenon (394).
Whereas I see an appropriately cautious version of the FTL theory as
providing a basis for a unified account of how both straight and twisted
self-deception happen (Mele 2001, 96–98), Nelkin argues for the need
to postulate a desire to believe that p in this connection. She argues, for
example, that “the jealous husband, who, by hypothesis, does not desire
that his wife is having an affair, does nevertheless want to believe that
she is,” if he is self-deceived in believing that she is so engaged (Nelkin
2002, 395). Nelkin makes it clear both that “the desire to believe is to be
thought of as a real, causally efficacious mental state, and not merely as
an abstraction from our attributions of self-deception” (394) and that it
“need not be conscious” (395).
Imagine two jealous husbands with very similar evidence in very simi-
lar circumstances. Each acquires the false, unwarranted belief that his wife
is having an affair—the belief that a, for short. Impartial observers judge
that the men are clearly and equally unwarranted in believing that a.
The men deal with the evidence in very similar, biased ways; data sugges-
tive of infidelity are especially salient for both; and they spend the same
considerable amount of time and energy mulling the matter over. One of
the men, Jack, has an unconscious desire to believe that a. This desire is
“a real, causally efficacious mental state.” It causes Jack to treat pertinent
evidence in a biased way, and that treatment results in Jack’s acquiring the
belief that a. The other man, John, lacks a desire to believe that a, but he
does have desires that contribute to his having acceptance and rejection
thresholds for a that are just like Jack’s. Suppose, for good measure, that
John has a desire not to acquire a false belief that his wife is innocent of
infidelity and that this desire causes him to treat pertinent evidence in a
biased way, which treatment results in his acquiring the belief that a. John
may also have a desire not to acquire the false belief that his wife is guilty;
but if he does, it is much weaker than the desire just mentioned. (A desire
not to acquire a false belief that p is a desire not to believe that p if p is false;
it is not simply a desire not to believe that p.)
On Nelkin’s view (assuming the absence of relevant causal deviance),
the facts about Jack and John entail that Jack enters self-deception in
acquiring the belief that his wife is having an affair, whereas John does not.
The role attributed to the desire to believe that p in Nelkin’s attempted
analysis is a causal one: “this desire causes S to treat evidence concerning
the truth value of p in a biased way” (2002, 394). If desires to believe
that p are well suited for this causal work, so are desires not to acquire a
false belief that not-p. So her claim that S’s having a desire specifically to
believe that p is a conceptually necessary condition of S’s entering self-
deception in acquiring a belief that p is implausible. And given the simi-
larities between the cases of Jack and John, it is very plausible that if Jack
is self-deceived, so is John.
Several philosophers who, like Nelkin, are willing to take (much of) the
explanatory portion of my view on board have argued that self-deception
necessarily involves a failure of self-knowledge and that a person may
satisfy my proposed sufficient conditions for self-deception in the absence
of a failure of the required kind (Holton 2001; Funkhouser 2005; Scott-
Kakures 2002). According to Richard Holton, the following is a case in
point:
Catherine has applied for several jobs recently, and has been unsuccessful
each time. She has also been horribly disappointed each time. She puts
her disappointment down to too much thought. On each occasion she had
spent a great deal of time thinking about the job, had, as a result, imagined
just what it would be like to get it, and so had been devastated when she
didn’t. She has just decided to apply for another job. She thinks that it is
clearly better than her current job; otherwise she wouldn’t be putting in for
it. But she has resolved not to think too deeply about what it is like, or to
examine the evidence that she has; at least not until or unless she gets an
offer. She knows that, were she to think more about the job, there is some
chance that her opinion of its merits would change; but she thinks that the
possibility of error here is worth risking to maintain her equanimity. As it
happens she is radically wrong about the job. It is a terrible job, far worse
than her current one, as a little more reflection would have shown her.
(2001, 60–61)
Holton asks whether Catherine is self-deceived and answers “Surely not”

(61).
In Holton’s story, it is not clear that my third condition is satisfied. As
I read the story, Catherine pretty quickly acquires the belief that the job
in question “is clearly better than her current job” (the belief that j, for
short), and then, with that belief in place, she decides to put the issue of
the job’s relative merits out of mind “until or unless she gets an offer.” As
far as I can tell, Catherine may acquire her belief that j in an unbiased
way, even if her retaining it owes significantly to motivational bias. Even
so, I believe, as I have mentioned, that people can enter self-deception in

retaining a belief (Mele 2001, 56–59), and a variant of my proposed suf-
ficient conditions applies straightforwardly to self-deception of the latter
kind. Although Holton is confident that Catherine is not self-deceived,
my impression is that she is self-deceived—if not at first, then later—in
believing that j. Of course, my impression may be a product of bias: I may
even be self-deceived in believing that Catherine is self-deceived. Given
that the conditions at issue are alleged sufficient conditions (and not indi-
vidually necessary and jointly sufficient conditions), I can beef them up
in case I am in the grip of a watered-down conception of self-deception
without having to worry that, in the process, I will commit myself to
a necessary condition that is not, in fact, necessary. It is interesting, in
this connection, that Holton’s proposed sufficient conditions (63–65) for
being “self-deceived about a subject matter a” are composed of a failure-
of-self-knowledge condition together with analogues of my conditions
2, 3, and 4 that are appropriate for an analysis of being self-deceived about
something (as opposed to entering self-deception in acquiring a belief
that p); and he does not dispute the explanatory portion of my view.
If I were to beef up my supposed sufficient conditions, how should
I do it? Dion Scott-Kakures argues that “reflective, critical reasoning is
essential to the process of self-deception” (2002, 577) and that “the error
of self-knowledge that makes . . . self-deception possible is a misconcep-
tion about what animates [the believer’s] doxastic or cognitive activities.
Like any reflective reasoner, she will regard her investigations as directed
by . . . her grasp upon what reason recommends,” but she is wrong about
this (599). “Her investigations are directionally driven by desire or inter-
est” (599), in ways featured in my account of how self-deception hap-
pens. If Scott-Kakures is right in requiring these things for self-deception,
something like the following condition should be added to my proposed
sufficient conditions for S’s entering self-deception in acquiring (or, alter-
natively, retaining) a belief that p:
5. S’s acquiring the belief that p (or, alternatively, retaining the belief that
p) is a product of “reflective, critical reasoning,” and S is wrong in regarding
that reasoning as properly directed.
Applying this condition to Holton’s story about Catherine is more com-

plicated than one may think. Again, Catherine might have acquired her
belief that j in an unbiased way. Her acquiring it might also have been a
product of relatively quick reflective, critical reasoning. If her acquiring it
was a product of this, then her retaining it was at least a partial product—
that is, effect—of this.8 Furthermore, Catherine might have regarded her
quick reasoning as properly directed, and she might not have been wrong
8. Here I am assuming that, when a person acquires and retains an x, his acquiring it is
among the causes of his retaining it and that the causes of his acquiring it are among the
indirect causes of his retaining it.
about that. If there is a mistaken “regarding” here along Scott-Kakures’

lines, the best candidate for it would seem to be Catherine’s wrongly
regarding her retaining the belief in question as supportable by “reflective,
critical reasoning.” Condition 5 can be split into two conditions to avoid
the problem identified—one for belief acquisition and the other for belief
retention—as follows:
5a. S’s acquiring the belief that p is a product of “reflective, critical
reasoning,” and S is wrong in regarding that reasoning as properly directed.
5b. S’s acquiring the belief that p is a product of “reflective, critical
reasoning,” and S is wrong in regarding her retaining the belief in question
as supportable by “reflective, critical reasoning.”
Does Catherine satisfy condition 5b? According to Holton, the failure

of self-knowledge required for self-deception has two components. First,
the required mistakes “about the self” do not “result from the application
to oneself of an erroneous belief that is not about oneself” (2001, 65).
Second, people make these “mistakes about themselves in ways that, had
they got it right, would have been self-knowledge” (65). If Catherine does
satisfy condition 5b, she would seem to satisfy the conditions that Holton
contends are sufficient for self-deception. And Holton denies that she is
self-deceived. So, presumably, as he understands Catherine’s story, she
does not acquire the belief at issue by means of “reflective, critical reason-
ing” (no matter how quick) or she does not regard her retaining the belief
as supportable by such reasoning (or both, of course). The story is his,
and he can build these details into it. Although, even when these details
are made explicit, my impression is that in retaining the belief Catherine
is self-deceived, I have no objection to including condition 5a in a list
of jointly sufficient conditions for entering self-deception in acquiring a
belief that p and including condition 5b in a parallel list of sufficient con-
ditions for entering self-deception in retaining a belief that p.
The second and third conditions in my proposed set of sufficient con-
ditions include the expressions “S treats data” and “This biased treatment.”
I intended my discussion (in Mele 1997 and 2001) of various ways of
entering self-deception in acquiring a belief that p to provide guidance
on the interpretation of “treats” and “treatment” in these conditions. But
if, strictly speaking, relatively simple motivationally biased mispercep-
tion counts as motivationally biased treatment of data (given the standard
meaning of “treats data”), trouble is brewing. Imagine that a hungry cat
misperceives a noise as the sound of her food being shaken into a bowl
and runs into the room from which the noise is emanating (Scott-Kakures
2002, 578–80). Those who are happy to attribute beliefs to cats may be
happy to say that the cat has a belief to the effect that food is available,
and that belief may be a relatively direct product or a constituent of her
motivationally biased misperception of the noise. If feline self-decep-
tion is out of the question and if “treats data” has a broader sense than
I intended, then something should be done about “treats” in condition 2
or a condition should be added that beefs up the proposed set of sufficient

conditions in a useful way. Condition 5a would handle the case of the cat;
so would its first conjunct alone.
Like Holton and Scott-Kakures, Eric Funkhouser defends the view that
self-deception requires “a failure of self-knowledge” (2005, 296), but his
position is more similar to Audi’s (mentioned in section 2) than to theirs.
In Funkhouser’s view, “in cases of self-deception, the agent has a desire to
believe that p, and this motivates her to engage in biased reasoning, avoid-
ance behavior, and similar deceptive measures that have been extremely
well characterized by Mele and other theorists.. . . This desire does not
result in a belief that p” (303). Instead, it results in the person’s having a
false second-order belief—the belief that she believes that p (309). Fur-
thermore, in addition to having “sufficient evidence to warrant a belief”
that not-p, the self-deceived person actually believes that not-p (308); but
she does not believe that she believes that not-p (309). As Funkhouser
notes (300, 307–8), his view resembles Audi’s, according to which the
self-deceiver unconsciously believes the truth, not-p, while sincerely
avowing p in the absence of a belief that p.
Funkhouser contends that “the presence of behavior that points against
the avowed belief” is conceptually required for self-deception (2005, 304).
I have argued against contentions of this kind elsewhere (Mele 2001, 52–
56), and I will not do so again here. I believe that some instances of self-
deception involve behavior of the sort Funkhouser has in mind (see, e.g.,
70–73). Also, I have no wish to claim that people are never in conditions
of the kind that Funkhouser and Audi specify in their attempted analyses
of self-deception; and one might stipulatively reserve the term “self-
deception” for those conditions. A more interesting issue is raised by
Audi’s and Funkhouser’s idea that the behavior “that points against the
avowed belief” that p should lead us to attribute to the agent a belief that
not-p. Although Funkhouser suggests that, in my examples of self-
deception, behavior of this kind is “invariably” missing (2005, 303–4), my
discussion of the following apparent example of self-deception in Self-
Deception Unmasked (2001, 70–73, from which I borrow in the following
four paragraphs) tackles this very issue.
Amelie Rorty (1988, 11) offers a putative example of self-deception that
may seem to speak strongly in favor of the presence of unconscious true
beliefs in some cases of self-deception. Dr. Androvna, a cancer specialist who
denies that she has cancer, “has begun to misdescribe and ignore symptoms
[of hers] that the most junior premedical student would recognize as the
unmistakable symptoms of the late stages of a currently incurable form of
cancer.” She has been neither a particularly private person nor a financial
planner, but now she “deflects [her friends’] attempts to discuss her condi-
tion [and] though young and by no means affluent, she is drawing up a
detailed will.” What is more, “never a serious correspondent, reticent about
matters of affection, she has taken to writing effusive letters to distant friends
and relatives, intimating farewells, and urging them to visit her soon.”
If I had read Rorty’s vignette out of context, I would have been con-
fident that Androvna knew—consciously—that she had cancer but did
not want to reveal that to others. That hypothesis certainly makes good
sense of the details offered. Even so, it is conceivable that Androvna is
self-deceived. If she is, what explains the detailed will and the effusive
letters? Some will suggest that, “deep down” or unconsciously, Androvna
knows or believes that she is dying and that this accounts for these activi-
ties. Assuming that it is conceivable that Androvna does not consciously
believe that she has cancer in the circumstances Rorty describes, is it also
conceivable that she does not unconsciously believe this either?
Yes, it is. Androvna’s not believing, unconsciously or otherwise, that she
has the disease is consistent with her consciously believing that there is a
significant chance that she has it, and that belief, in conjunction with rel-
evant desires, can lead her to make out a will, write the letters, and deflect
questions. (Notice that she may be self-deceived in believing that there is
only a significant chance that she has cancer.) Given Rorty’s description
of the case and the assumption that Androvna lacks the conscious belief
that she has cancer, is it more likely (1) that she believes unconsciously
that she has the disease (has a “type 1” cancer-belief ), or (2) that she
consciously believes that there is a significant chance that she has cancer
without also believing, unconsciously or otherwise, that she has it (has a
“type 2” cancer-belief )? Base rate information is relevant here. My stu-
dents know that there are a great many more blue-collar workers than
lawyers. Yet when I ask them whether a man wearing a nice suit and a tie
is more likely to be a lawyer or a blue-collar worker, most of them answer,
“a lawyer”—at least until the relevance of base rates is made salient. What
are the relative frequencies of type 1 and type 2 beliefs (i.e., unconscious
beliefs of the sort that Funkhouser may want to attribute to Androvna
and conscious beliefs that there is a significant chance that p that fall
short of being beliefs that p)?9 Until one has at least a partial basis for an
answer to this question that would help underwrite the judgment that
Androvna unconsciously believes that she has cancer, one is not entitled
to be confident that she has such a belief. Plainly, we have and act on a
great many type 2 beliefs. For many of us, such beliefs help to explain why
we purchase home insurance, for example, or take an umbrella to work
when we read in the morning paper that there is a 30 percent chance of
rain. If there is anything approaching comparably weighty evidence of
frequent type 1 beliefs, I am not aware of it.10
9. Those who prefer to think in terms of degree of belief should read such expressions
of mine as “S believes that p” as shorthand for “S believes that p to a degree greater than
0.5 (on a scale from 0 to 1).”
10. The concept of unconscious belief in play here does not include so-called “standing
beliefs”—dispositions of a being to have occurrent beliefs that p, which dispositions have a
partial source in occurrent beliefs that p that the being had earlier.
One may ask why, if Androvna believes that there is a significant

chance that she is stricken with cancer, she does not seek medical atten-
tion. Recall that she knows the type of cancer at issue to be incurable; she
may see little point in consulting fellow cancer specialists. Setting that
detail aside, procrastination about seeking medical attention is, unfortu-
nately, a familiar phenomenon, and it does not require type 1 beliefs.
People often wait too long to act on their type 2 beliefs in this sphere.
Even a story like Androvna’s—one designed to make it very plausible that
a crucial unconscious true belief is at work—can be accommodated by
the view of self-deception I have sketched.
Funkhouser suggests that we should “limit the term ‘self-deception’ ”
to cases “in which people sincerely avow one thing while behaving
otherwise” and “reserve ‘self-delusion’ for Mele-style examples” (2005,
304), such as the examples of Rex, Sid, Beth, and Betty in section 1. The
expression “behaving otherwise” oversimplifies matters. Funkhouser’s
expression “behavior that points against the avowed belief” is more
apt. Some of the agent’s (nonverbal) behavior may point toward the
avowed belief, as Androvna’s refraining from seeking medical atten-
tion does. And even if we do limit our use of the term “self-deception”
in the way Funkhouser suggests—that is, to cases in which there is a
significant tension between some of the agent’s behavior and certain
of her sincere avowals—it may be that most cases of self-deception do
not fit his model. Perhaps, even on the assumption that self-deception
requires tension of this kind, in most cases of self-deception agents
who sincerely avow that p also believe that p and do not unconsciously
believe that not-p.
Suppose that in some cases like Androvna’s, the person who sincerely
avows that p has the false, unwarranted belief that p and that in others,
although the person sincerely avows that p, she does not believe that p but
instead falsely believes that she believes that p. Funkhouser suggests that
an explanation of the sort I offer for the production of the false beliefs in
cases of the former kind applies nicely to the false second-order beliefs
in cases of the latter kind. Moreover, in cases of the latter kind, the per-
son has a directly relevant, motivationally biased false belief—the belief
that she believes that p. This fits the protoanalysis I mentioned earlier,
according to which people enter self-deception in acquiring a belief that
p if and only if p is false and they acquire the belief in a suitably biased
way (Mele 2001, 120). This protoanalysis does not restrict the beliefs
at issue to first-order beliefs. So, regarding Androvna-style cases—that
is, cases involving significant behavior-assertion tension—Funkhouser
and I agree about quite a bit. Of course, he would say that if Androvna
believes that she does not have cancer, then she is not self-deceived and
that self-deception on the matter requires her believing that she does
have cancer (while not believing that she believes this). My view is that
each of the following inconsistent hypotheses about Androvna’s story is
consistent with her being self-deceived: (Hm) Androvna believes that
she does not have cancer; (Hf ) Androvna does not believe that she does
not have cancer, but she does believe that she believes that she does not
have cancer. Funkhouser can stipulate that if Hm is true, then Androvna is
not self-deceived in believing what she does, but stipulation is uninterest-
ing. What might be interesting is an argument that in no Androvna-style
case does the protagonist have the pertinent motivated false first-order
belief. Unless I am deceived, a convincing argument for this will not be
forthcoming.
Funkhouser and I also agree about how false beliefs are produced in
cases of what he calls “self-delusion” ( but not “self-deception”)—namely,
in the same ways they are produced in the Androvna-style cases. In light
of the fact that we agree about this and in light of the points made in the
preceding two paragraphs, a debate about whether or not we should say
that “self-delusion,” in this sense, is a species of self-deception does not
promise to be very interesting.
So have I unmasked self-deception, or am I self-deceived? In Self-
Deception Unmasked and in earlier work, I tried to show that self-deception
is masked by traditional models of the phenomenon that treat it as an
intrapersonal analogue of stereotypical interpersonal deception. Accord-
ing to these models, self-deceivers intentionally deceive themselves into
believing that p, and there is a time at which they believe that p is false
while also believing that p is true. I offered an alternative model of self-
deception and a detailed explanation of how garden-variety self-deception
happens. The critics whose work I have discussed in this article are happy
enough with the explanatory portion of my position. Their worries are
about what I count as self-deception. Someone might claim that if I have
directly unmasked anything, it is not self-deception. Funkhouser comes
close to claiming that; but for the reasons I have offered, I am unper-
suaded. The other critics whose work I have discussed here defend posi-
tions on self-deception that are more similar to mine. Nelkin attempts to
augment my proposed sufficient conditions for entering self-deception
in acquiring a belief that p in such a way as to arrive at individually nec-
essary and jointly sufficient conditions, and Holton and Scott-Kakures,
focusing on self-knowledge, attempt to improve on my proposed suffi-
cient conditions for self-deception. As I explained, I am happy to add a
failure-of-self-knowledge condition to my list of jointly sufficient condi-
tions, and so doing is consistent with my having unmasked self-deception.
Of course, in this article, I have discussed the work only of some critics;
some readers may worry that I am guilty of one-sided evidence-gathering.
However, I have responded to a great many critics elsewhere (especially
in Mele 1997 and 2001), and I wanted to examine interesting objections
to my position on self-deception that I had not yet examined.11
11. For helpful feedback, I am grateful to Eric Funkhouser, Dana Nelkin, Dion
Scott-Kakures, and Neil Van Leeuwen.
References
Audi, R. 1982. “Believing and Affirming.” Mind 91, 115–20.
——— . 1985. “Self-Deception and Rationality.” In Self-Deception and Self-
Understanding, ed. M. Martin. Lawrence: University of Kansas Press, 169–94.
——— . 1997. “Self-Deception vs. Self-Caused Deception: A Comment on Profes-
sor Mele.” Behavioral and Brain Sciences 20, 104.
Baron, J. 1988. Thinking and Deciding. Cambridge: Cambridge University Press.
Davidson, D. 1985. “Deception and Division.” In Actions and Events, ed. E. LePore
and B. McLaughlin. Oxford: Blackwell, 138–48.
Friedrich, J. 1993. “Primary Error Detection and Minimization (PEDMIN) Strate-
gies in Social Cognition: A Reinterpretation of Confirmation Bias Phenomena.”
Psychological Review 100, 298–319.
Funkhouser, E. 2005. “Do the Self-Deceived Get What They Want?” Pacific Philo-
sophical Quarterly 86, 295–312.
Gigerenzer, G., and K. Hug. 1992. “Domain-Specific Reasoning: Social Contracts,
Cheating, and Perspective Change.” Cognition 43, 127–71.
Holton, R. 2001. “What Is the Role of the Self in Self-Deception?” Proceedings of
the Aristotelian Society 101, 53–69.
Kunda, Z. 1987. “Motivated Inference: Self-Serving Generation and Evaluation of
Causal Theories.” Journal of Personality and Social Psychology 53, 636–47.
——— . 1990. “The Case for Motivated Reasoning.” Psychological Bulletin 108,
480–98.
——— . 1999. Social Cognition. Cambridge, Mass.: MIT Press.
McLaughlin, B. 1988. “Exploring the Possibility of Self-Deception in Belief.” In
Perspectives on Self-Deception, ed. B. McLaughlin and A. Rorty. Berkeley: Uni-
versity of California Press, 29–62.
Mele, A. 1982. “Self-Deception, Action, and Will: Comments.” Erkenntnis 18,
159–64.
——— . 1983. “Self-Deception.” Philosophical Quarterly 33, 365–77.
——— . 1987. Irrationality. New York: Oxford University Press.
——— . 1997. “Real Self-Deception.” Behavioral and Brain Sciences 20, 91–102.
——— . 1999. “Twisted Self-Deception.” Philosophical Psychology 12: 117–37.
——— . 2001. Self-Deception Unmasked. Princeton: Princeton University Press.
Nelkin, D. 2002. “Self-Deception, Motivation, and the Desire to Believe.” Pacific
Philosophical Quarterly 83, 384– 406.
Rorty, A. 1988. “The Deceptive Self: Liars, Layers, and Lairs.” In Perspectives
on Self-Deception, ed., B. McLaughlin and A. Rorty. Berkeley: University of
California Press, 11–28.
Scott-Kakures, D. 2002. “At ‘Permanent Risk’: Reasoning and Self-Knowledge in
Self-Deception.” Philosophy and Phenomenological Research 65, 576–603.
Sharpsteen, D., and L. Kirkpatrick. 1997. “Romantic Jealousy and Adult Romantic
Attachment.” Journal of Personality and Social Psychology 72, 627–40.
Szabados, B. 1985. “The Self, Its Passions, and Self-Deception.” In Self-Deception and
Self-Understanding, ed. M. Martin. Lawrence: University of Kansas Press, 143–68.
Trope, Y., and A. Liberman. 1996. “Social Hypothesis Testing: Cognitive and Moti-
vational Mechanisms.” In Social Psychology: Handbook of Basic Principles, ed.
E. Higgins and A. Kruglanski. New York: Guilford Press, 239–70.
Trope, Y., B. Gervey, and N. Liberman. 1997. “Wishful Thinking from a Pragmatic
Hypothesis-Testing Perspective.” In The Mythomanias: The Nature of Deception
and Self-Deception, ed. M. Myslobodsky. Mahwah, N.J.: Erlbaum, 105–31.
Index
Achilles, guile of, 50, 57 Bok, Sissela, 26, 119

actions, 227 consequenses of lies, 28
blameworthy, 196 excuses, manufacture of, 28
wrongful, 242 Bond and DePaulo, 120, 121
activities, 251 Braude, Steve, 21
Adorno, Theodor, 93 breach of trust, 150, 151
Aeneid, 55 Buddhism, 17
Age of Chivalry, 51 Bugental et al., 126
Akrasia. See self-deception bullshit, 43, 154, 181
altruistic lies, 125 enemy of truth, 48
Ames, Roger, 34 attention to detail, 43
animals, 114 construction of, 47
deceptions of, 105 lying, difference from, 46
experiments on, 105, 108, 109, 116 types of, 183, 184
inferiority to man, 112
practices of deception, 23 Caesar, 55
Aristotle, 20, 29, 40, 41, 63, 207 Camus, Albert, 19
truthfulness, 20 anti-hero, 21
assertion, concept of, 173 Meursault, 19
Audi, Robert, 266, 272 The Fall, 27, 32
Augustine, 17, 47 Carter, Stephen, 20
authenticity, 96 Cassandra, 55
autonomy, 150 Chamberlain, Neville, 127
avoidability principle, 144 character, 24
Chaucer, 52
bad faith, 54. See also Sartre, Jean-Paul, cheating, 125. See also deceit, trade
bad faith Chisholm and Feehan, 162
Barnes, J. A., 175 lying, definition of, 174
beliefs, 187 Christianity, 63
first-order, 7, 83, 88, 90, 91, 92, 115, Coady, Tony, 15, 17
116, 274, 275 conceptuality, sociohistorical
ideology of, 92 conditions, 92
second-order, 7, 83, 88, 89, 90, 91, confidence, thresholds, 264
92, 93, 94, 99, 100, 102, 115, Constant, Benjamin, 229, 231, 235,
116, 245, 250, 253, 272, 274 242, 243
second-order, reformations of, 93 Cosmides and Tooby, 125
third-order, 99, 100, 102 courage, as virtue, 61
Benton, Robert, 230 Crooks-Seshadri, Kalpana, 113
Bernstein, J. M., 93 cunning, definition of, 52
277
Index 279
Dasein, 95, 96 unavoidability of, 146

David (Goliath’s enemy), 51 as vice, 18
deceit wrongness of, 141
behavioral clues, 133 DePaulo and Pfeifer, 122
contemporary examples of, 49, 65 Derrida, Jacques, 15, 113
historical examples of, 55, 60 criticism of Lacan, 112, 113, 116
trade, 61, 62, 63 L’animal que donce je suis, 112
war, 58 Descartes, Rene, 257
deception Diogenes, 18
allowable forms, 119 Diomedes, 53, 54
animals, 114 Dogen, 18
avoidance of, 198 Dolon, as coward, 53
belief, 178 Dostoevski, Fyodor, The Idiot, 21
bullshit, 182 duplicity, 108
Cartesian variety, 193 duties
characterizations of, 246 ethical, 218
cheating, 120, 121, 122 juristic, 218
concealment, 119, 120 perfect and imperfect, 215
concepts of, 176, 154 Right Not to Lie, 221
criteria of, 118 social maintenence, 221
defensive acts of, 139, 144
definitions of, 154, 176, 178 Eisenstadt v. Centel, 147
ethical problems of, 30 Ekman and Friesen, 120
examples of, 18, 74, 79 Ekman and O’Sullivan, 120, 122
expected, 119 Ekman, Paul, Telling Lies, 132
experience of, 71 Enlightenment, 245
experiments about, 120, 121, 122 Epictetus, 19
falsification, 120 existential-phenomenological
identification of, 128 tradition, 67, 70
and ignorance, 21 descriptions of, 71
intention of, 118, 176
interpersonal relationships, 127 facial movements. See deception,
vs keeping someone in the dark, 179 experiments about
linguistic intuitions, 177 facticity, 100
manipulation, 143 faith. See Sartre, Jean-Paul,
moral relavance, 152 good faith
moral status, 231 false beliefs, 148, 154, 265
motivations for, 32, 120 false consciousness, 245
nature of, 188 false statements, 154, 156, 160, 166
negotiation, 145 fools, as victims, 66
penalties of, 28 Foucault, Michel, 88, 89
perception of, 76 Frankfurt, Harry, 154, 181, 186
social consequences of, 126 Frege, Gottlob, 164
as social construct, 22, 88, 124, 188 Freud, Sigmund, 19, 22, 33, 101, 104,
sociohistorical context of, 85, 87, 96 113, 255
temptations of, 76 Friesen, Wally, 129
thought experiments, 140, 142, 143, FTL Theory, 264, 266, 268
144, 147, 150, 177, 179, 180, Funkhouser, Eric, 272, 274, 275
189, 197 unconscious beliefs, 273
278 Index
Geller, Uri, 118 illusions. See self-deception,

Ginton, Daie, Elaad, and Ben-Shakhar, experience of
121 insincerity, 205
Glaukos (enemy of Diomedes), 54 intent to deceive, 159
Gnostics, 65 intentional acts, 161
Goffman, Erving, 66, 119, 128, 134 intentions, 251
Goliath, 51 irrationality, motivations of, 95
Grafen, Alan, 125 Islam, 63
guard-dropping. See deceit, historical
examples of Johnson, Samuel, 26
guile jokes, 155
necessity, 58 Jones, Ernst, 22
games, 56 Jung, Karl, 28
war, 56
Kant, Immanuel, 4, 5, 7, 9, 17, 19,
Haggard and Isaacs. See deception, 20, 29, 30, 38–39, 86–87, 96,
experiments about 158, 201, 203, 208–210, 218,
half-truths, 154, 186, 187 224–225, 227, 236–238, 244
hallucinations. See deception, a priori morality, 86
experience of government, 236–238
Hegel, G. W. F., 7, 18, 86, 87, 88, 91, political power, 241
92, 95, 99, 255 Anthropology from a Pragmatic Point
“absolute knowing”, 89 of View, 204, 219
beliefs, accounts of, 91 believed-true statements, 206
Phenomenology of Spirit, 89 believed-false statements, 206
Philosophy of Right, 93, 94 civil condition, 232
pursuit of truth, 89 Critique of Practical Reason, 217
sociohistorical concepts, 90 deception, sociohistorical terms,
Heidegger, Martin, 67 85, 86
das Man, 95 Doctrine of Right, 218, 223, 240
deception, account of, 72, 75 duties, examples of, 215, 223
errors, distinctions of, 74 Groundwork of the Metaphysics of
hammer example, 73 Morals, 214, 216, 219, 225
ontology of, 96 Hegel’s criticism of, 86, 87, 89
Sein und Zeit, 70, 71 intentions, 210
self, 95 internal lie, 205
structural conditions, 73, 74 Lectures on Ethics, 38
“high-minded lie”, 16 lying, examples of, 201, 202, 204, 207,
Hitler, Adolph, 127 213, 214, 217, 219, 231, 233, 239
Holton, Richard, 269, 270, 271, 275 Metaphysics of Morals, 210, 211
Homer, heroic ethics in, 53 moral obligations, 232, 235
honesty moral purity, 226
perfect duty, 20 moral viewpoint, 212
reflection of popular morality, 19 morality requirements, 231
virtue, 18 natural law, 229
Husserl, Edmund, 84 self-defense, 240, 242
On the Supposed Right to Lie from
Icelandic sagas, 54 Philanthropy, 201, 206, 222, 226,
Iliad, deceptive practices in, 51 229, 234
Index 281
Kant, Immanuel (continued ) contextual clues, 170

political nature, 230 deception, differences between, 154
practical philosophy, 230 definition of, 153, 154, 155, 156,
responsibility, 226 162, 171, 173, 174
rightful relations, 236 duty not to, 214
The Doctrine of Virtue, 218 intentions of, 189
The Right to Lie, 221 Kantian three senses, 203, 205,
Theory and Practice, 227 207, 208
truthfulness, 234 language of, 158, 174
violation of rights, 233 linguistic intuitions, 174
Kantian condemnations, 229 moral questions about, 174
Kierkegaard, Soren, 95, 96 necessary conditions of, 160
Kraut and Poe, 122 public performance, 30
self-defense, 229
Lacan, Jacques, 7, 104–106, 108, thought experiments, 158, 159, 169
110–115 trust, 175
behaviorism, criticism of, 109 under oath, 217 (see also under
consciousness, 115 truth)
criticisms, 106, 107 unintentional accounts of, 132
experiments, 108 violations to duty, 228
language, 107, 109, 113
sexual behaviour, 111 macroexpressions, 132
speech, 114 manipulation. See deception, acts of
Laclos, Chodoros, Liasons Marcius, guile of, 58
dangereuses, 27 Marx, Karl, 255
language, 108, 110 Masserman, Jules, 109
Laocoön, 55 Mathemes, Lacan, 115
La Rochefoucauld, François, 66 Mele, Alfred, Self-Deception Unmasked,
Leibniz, Gottfried, monadic theory, 83 268, 272
Levine and McCornack, 122 Merleau-Ponty, Maurice, 67, 71
Lewinsky, Monica, 154 deception, 72, 76
liars, faults of, 40 perception, 76, 80, 81
lie detection, 118, 132. See also under metastable. See Sartre, Jean-Paul,
deception “metastable”
direct, 24 Micro Expression Training Tool, 131
ethical sense, 208 microfacial expressions, 118, 129, 131
juristic sense, 209 contextual issues of, 130
perpetration of, 118 Mill, J. S., 195
possiblities of, 37 misinterpretation, 261
sense of right, 211 misleading statements, 126, 143,
as skill, 126 149, 177
training, 131 moral acceptability of, 148
Longfellow, 43 Montaigne, Michel, 5, 6, 38, 39, 60
lying. See also under deception Essays, 57, 58
artful forms, 29 Of Giving the Lie, 38
bullshit, 182, 185 treachery, 60
casuistical questions about, 227 Virgil, 59
cheating, 161, 168 moral ambiguity, 206
concept of, 154, 166, 174 moral purism, 149
consequences of, 29 moral wrongness, 195
280 Index
motivated irrational beliefs. See beliefs, to know the truth, 158

first-order romantic love, 119
motivations, first-order, 83 Rorty, Amelie, 26, 33, 272, 273
myth, 28, 119 Rorty, Richard, 15, 191
Ross, W. D., 163
Nelkin, Dana, 268, 269, 275
Neuro-Linguistic Programming, 129 Sapientia, 54, 55, 56
Nietzsche, Friedrich, 15, 67, 84, 87, Sartre, Jean-Paul, 19, 34, 67
95, 114 bad faith, 96, 97, 100
appearances, values of, 67 Being and Nothingness, 96, 100, 103
Beyond Good and Evil, 67 Being-for-Others, 25, 33
Kritische Gesamtausgabe, 67 choice, 99
untruths, 17 “divided mind”, 101
“The Noble Lie”, 16 facticity-transcendence, 97
freedom, 97
obligations, 220 good faith, 98
Odysseus, guile of, 50, 53, 61 “metastable”, 25, 100
O’Sullivan and Ekman, 123 self, 95
the Other, 114 transparency of consciousness, 19
scams. See deceit, contemporary
paranoid fantasies, 58 examples of
Pascal, Fania, 44 Scott-Kakures, Dion, 270, 271, 275
Paton, H. J., 229 second-order motivations, 83
perceptions, 78, 79, 80 Sedgwick, Sally, 234, 235
perceptual deception selective focus, 262
errors of judgment, 68 self
examples of, 69, 77 social construction, 34, 35
existential-phenomenonological complexities of, 35
account of, 70 nature of, 34
phenomenon of, 68 sociohistorical components of, 98
temptations of, 69, 70 self-betrayal, 41
phenomenology, of deception, 68, 77 self-deception, 244
philosophical truth, 18 about the self, 23
Pincoffs, Edmund, dishonesty as vice, 19 beneficiaries of, 253
Pinocchio, 133 benefits of, 252
Plato, 16, 17, 18, 21, 257 biased beliefs, 261
The Republic, 18 conditions of, 266, 267, 268, 270, 271
politeness, pretenses of, 9, 17, 66, desire to, 269
119, 207 epistemological problems of, 33
Posner, Richard, 147 epistimic attitudes, 254
President Clinton, 154 ethical problems of, 27
promise-breaking, 163 externalist models, 33, 34
psychoanalysis, 104 “false consciousness”, 16
generalizations of, 256
Ramsey, Frank, 164 intentionalist approach to, 96, 101
rationality, 40 internalist models, 33, 34
respect. See trust lexical approach, 82
Rich, Adrienne, 39, 40 motivational bias of, 269
right myth, 16
of self-defense, 149 notions of, 247
Index 281
self-deception (continued ) antirealist theories of, 193

paradoxes of, 82, 96 bullshit, 182, 185
performance of, 33 condition of (see Heidegger, Martin,
phenomena of, 245 das Man)
practitioners of, 253 deceived about, 196
protoanalysis of, 266 harm, 22
self-defense, 256 nature of, 188, 194
social construct, 102 nonepistemic theories, 192
social phenomenon of, 24 objective account of, 192
strategies of, 250 pluralist theory, 195
thought experiments, 263, 268, sociohistorical, 96
269, 272 under oath, 132, 143, 160, 167
traditional view of, 260 value of, 197
varieties of, 18, 94 vulnerability of, 191
self-defense, 145 warranting of, 165, 167, 168, 169,
semantemes, lexical distribution of, 110 170, 171, 172
sexual behaviour, 110
Shakespeare, 138, 41, 42 untruthful statements, 212
Sidgwick, Henry, 16, 17 untruthfulness, 205, 207, 222
Sillars and Scott, 122
social constructivism, 83, 99 victims
society betrayals of, 40
conventions of, 24 moral failings of, 53
ethical values of, 23 Vikings, war practices of, 56
sociohistorical truths, 90 Virgil, 55
Socrates, 18, 19, 21, 255
speech, and lies, 105 Walker, John, 132
spin, 154, 186 war
Spinoza, Baruch, 255 game of skill, 50
St. Exupery, 34 moral problems of, 51
stages of inquiry, 193 stakes of, 56
Starr Grand Jury, 154 weapons of mass destruction, 178
Sticklebacks. See sexual behaviour willfulness, 190
Strawson, 164, 175 William of Malmesbury (William the
suicide, 228 Conquerer), 59
Sun Tzu, Art of War, 50 Williams, Bernard, 89
superwarrant, 194 lying, definition of, 173
Truth and Truthfulness, 139
teleological claims, 228 wisdom, etymology of, 56, 57
Transparency Thesis, 164 Wittgenstein, Ludwig, 6, 17, 18, 43,
Thrasymachus. See Plato, The Republic 44, 45, 46
trade bullshit distinctions, 45, 46
buyers’ expectations of, 62, 63 Wood, Allen, 94, 97, 100, 101
sellers’ expectations of, 62 wrongs
treachery, and treason, 52 formal, 237
tricksters, in trade, 53, 65 material, 237
Trojans, 53
trust, 42, 139, 141 Zeitgeist, 19, 87
truth Zeus, tricks of, 49
accurate evaluations of, 133 Zöllner’s illusion, 68, 69, 77

Clancy Martin - The Philosophy of Deception-Oxford University Press (2009)

Uploaded by

Copyright:

Available Formats

Clancy Martin - The Philosophy of Deception-Oxford University Press (2009)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Clancy Martin - The Philosophy of Deception-Oxford University Press (2009)

Uploaded by

Copyright:

Available Formats

The Philosophy of Deception

This page intentionally left blank

Oxford New York

Copyright © 2009 by Oxford University Press, Inc.

Published by Oxford University Press, Inc.

Library of Congress Cataloging-in-Publication Data

Introduction: The Intersections of Deception

PART I. THE PRACTICES OF DECEPTION

1 Self, Deception, and Self-Deception in Philosophy, 15

2 On Truth, Lies, and Bullshit, 37

3 Deceit in War and Trade, 49

4 On the “Existential Positivity of Our Ability to Be Deceived”, 67

5 Self-Deception, Deception, and the Way of the World, 82

6 Duplicity Makes the Man, Or, Can Animals Lie?, 104

7 Lie Catching and Microexpressions, 118

PART II. TRUTH, LIES, AND SELF-DECEPTION: THE THEORY

8 Deception and Trust, 139

9 Lying, Deception, and Related Concepts, 153

10 Deception and the Nature of Truth, 188

11 The Truth about Kant on Lies, 201

12 On the Supposed Duty of Truthfulness: Kant on Lying

13 User-Friendly Self-Deception: A Traveler’s Manual, 244

14 Have I Unmasked Self-Deception or Am I Self-Deceived?, 260

THOMAS L. CARSON is the author of The Status of Morality (1984),

PAUL EKMAN is a pioneer in the study of emotions and their relation

HARRY FRANKFURT is the author of the bestselling On Bullshit (2005)

MICHAEL P. LYNCH is professor of philosophy at the University of

JAMES EDWIN MAHON is associate professor and chair of the Philosophy

ALFRED R. MELE is William H. and Lucyle T. Werkmeister Professor of

WILLIAM IAN MILLER has research interests in medieval history, especially

KELLY OLIVER is W. Alton Jones Chair of Philosophy and professor of

AMELIE RORTY is visiting professor of philosophy at Boston University and

DAVID SHERMAN is associate professor of philosophy at the University

ROBERT C. SOLOMON (1942–2007) was professor of philosophy at the

ALAN STRUDLER is professor of legal studies and coordinator of the

DAVID SUSSMAN is associate professor of philosophy at the University of

MARK A. WRATHALL is associate professor of philosophy at the

The suspicion that motivated me in gathering together this unusual and

1. Friedrich Nietzsche, Human, All-Too-Human, trans. R. J. Hollingdale (New York:

since the literature on self-deception is much more complete than that

varieties of self-deception are ineradicable and useful” (244), it is not good

believing each of two explicitly contrary propositions” (261). This is

Mele then canvasses the opinions of various critics—Dana Nelkin,

This page intentionally left blank

Self, Deception, and Self-Deception

“I have done that,” says my memory. “I cannot have done that,”

sneaking suspicion that there might be a number of contexts in which lying

Call me a truth-seeker, and I will be satisﬁed.

When it comes to the larger questions of philosophy—the meaning of life,

6. Amelie Rorty, “Adaptivity and Self-Knowledge,” in Mind in Action (Boston: Beacon,

of nature, it is by no means clear what it means to seek “the truth.”7 Philo-

7. Wittgenstein’s statement in the epigraph is quoted in Ray Monk, Ludwig

10. Nietzsche, Beyond Good and Evil.

18. Plato, Republic, trans. G. M. A. Grube (Indianapolis: Hackett, 1974) bk. 7.

DECEPTION, SELF-DECEPTION, AND THE SELF

It is too often assumed that deception is a peculiarly linguistic

But, as I suggested earlier, I can effectively deceive someone by driving

need to fool ourselves coupled with the strong temptation to deceive

perhaps, what Sartre calls “metastable”)27 and a continuous effort of enor-

with difﬁcult social situations. Within the limited realm of self-knowledge,