John Martin Fischer

Our Stories
“Oh, mother, people get run over by trucks every day. Why can’t something
like that happen to Uncle Elwood?”
—Myrtle Mae Simmons, in Harvey, starring James Stewart
Our Stories
E SSAY S O N LIF E, D EAT H, A ND F R EE WILL
John Martin Fischer
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.
CIP to come
ISBN-13: 978-0-19-537495-7
135798642
Printed in the United States of America

on acid-free paper
Contents
Acknowledgments vii
1. Introduction: Meaning in Life and Death 3

2. Why Is Death Bad? 27
Anthony L. Brueckner and John Martin Fischer
3. Death, Badness, and the Impossibility of Experience 37
John Martin Fischer
4. Death and the Psychological Conception
of Personal Identity 51
John Martin Fischer and Daniel Speak
5. Earlier Birth and Later Death: Symmetry Through
Thick and Thin 63
John Martin Fischer
6. Why Immortality Is Not So Bad 79
John Martin Fischer
Appendix: Philosophical Models of Immortality
in Science Fiction 93
John Martin Fischer and Ruth Curl
7. Epicureanism About Death and Immortality 103
John Martin Fischer
8. Stories 129
John Martin Fischer
9. Free Will, Death, and Immortality: The Role of Narrative 145
John Martin Fischer
10. Stories and the Meaning of Life 165
John Martin Fischer
Index 179
v
This page intentionally left blank
Acknowledgments
I am very grateful to Neal Tognazzini and Patrick Todd for their

excellent assistance in preparing this book for publication.
Permission to reprint the following articles is hereby acknowledged:

John Martin Fischer and Anthony Brueckner, “Why Is Death Bad?”
Philosophical Studies 50 (1986), pp. 213–21.
“Death, Badness, and the Impossibility of Experience,” Journal of
Ethics 1 (1997), pp. 341–53.
John Martin Fischer and Daniel Speak, “Death and the Psychological
Conception of Personal Identity,” Midwest Studies in Philosophy 24 (2000),
pp. 84–93.
John Martin Fischer, “Earlier and Later Birth: Symmetry Through
Thick and Thin,” Richard Feldman, Kris McDaniel, and Jason R. Raibley
(eds.), The Good, the Right, Life and Death (Aldershot, England: Ashgate
Publishing, 2006), pp. 189–202.
John Martin Fischer, “Why Immortality Is Not So Bad,” International
Journal of Philosophical Studies 2 (1994), pp. 257–70.
John Martin Fischer and Ruth Curl, “Philosophical Models of
Immortality in Science Fiction,” George Slusser, Gary Westfahl, and
Eric S Rabkin, eds., Immortal Engines: Life Extension and Immortality in
Science Fiction and Fantasy (Athens, Ga.: University of Georgia Press,
1996), pp. 3–12.
John Martin Fischer, “Epicureanism About Death and Immortality,”
Journal of Ethics 10 (2006), pp. 355–81.
John Martin Fischer, “Stories,” Midwest Studies in Philosophy 20 (1996),
pp. 1–14.
John Martin Fischer, “Free Will, Death, and Immortality: The Role
of Narrative,” Philosophical Papers 34 (2005), pp. 379–404.
John Martin Fischer, “Stories and the Meaning of Life”, revised and
expanded version of “A Reply to Pereboom, Zimmerman, and Smith,”
part of a book symposium on John Martin Fischer, My Way: Essays on
Moral Responsibility, Philosophical Books (2006), pp. 235–44; forthcoming
in Philosophic Exchange.
vii
Permission to reprint the following poetry and song lyrics is hereby
acknowledged:
“Mad World,” words and music by Roland Orzabal. Copyright ©
1983 Roland Orzabal Limited. All rights administered by Chrysalis
Music Limited. All Rights Reserved. Used by Permission.
Six lines from “Lady Lazarus,” in Ariel, by Sylvia Plath. Copyright © 1963
by Ted Hughes. Reprinted by permission of Harper Collins Publishers.
Our Stories
1
Introduction: Meaning
in Life and Death
All around me are familiar faces

Worn out places, worn out faces
Bright and early for their daily races
Going nowhere, going nowhere
Their tears are filling up their glasses
No expression, no expression
Hide my head I want to drown my sorrow
No tomorrow, no tomorrow
And I find it kinda funny
I find it kinda sad
The dreams in which I’m dying
Are the best I’ve ever had
I find it hard to tell you
I find it hard to take
When people run in circles
It’s a very, very mad world mad world
“Mad World,” as sung by Gary Jules
Let us hope, however, that there is a method to the madness—or at least

some philosophical insights that can help to structure the insanity. I shall
begin by laying out in the barest bones my overall view on the badness
of death, the possible goodness of immortality, and the meaning of life.
I believe that the notions of “narrative” (stories) and free will play key
roles in understanding these deep and mysterious phenomena, and I
shall seek to sketch some of the connections between free will, stories,
and the meaning of life. In this introductory essay, I only give the brief-
est overview of issues I explore in more detail in the essays in the book.
Also, I go on to offer some ruminations on the nature and (possible)
implications of these views, including suggestions for future directions
in thinking about this web of fundamental existential issues.
I. AN OVERVIEW
Let me begin this essay by distinguishing the phenomena picked out by
“dying,” “death,” and “being dead.” Dying is part of life—the last part,
and it is not too mysterious how it can be bad for the individual who
3
4 Our Stories: Essays on Life, Death, and Free Will
dies (insofar as it can involve pain and suffering).1 Sometimes death is

taken to be the “transition,” as it were, between the last part of life and
the first part of being dead; metaphorically, death might be thought of
as (e.g.) the crossing of the River Styx. But I shall lump together the ref-
erents of “death” and “being dead,” taking them to be (roughly speak-
ing) the state of permanent nonexistence after being alive.2
Now one can ask how death can be bad for the individual who dies.
Again, as it is relatively clear that pain and suffering can be bad for an
individual, it is not terribly mysterious how the death of one person
can be a bad thing for another (living) person. But how can the death
of an individual be bad for that very individual? If one believes in an
“afterlife” with conscious episodes or something akin to conscious epi-
sodes, then it would be fairly straightforward that death could be bad
for an individual—it could involve the pain and suffering of Hell. And
I certainly don’t know that there isn’t a Hell (i.e., a Hell apart from our
actual lives3). But here we will simply put the possibility of an afterlife
to the side and assume, simply for the sake of the argument, that there
is no “afterlife.” That is, we assume that death is an experiential blank.
Now how can it be that an experiential blank can be bad for an indi-
vidual? Worse yet, how can it be that this sort of state can be bad for the
individual, given that the individual has gone out of existence?
The followers of Epicurus and Lucretius argue that despite the com-
monsense view that death can indeed be bad for the individual who
dies, we can see, upon more careful consideration, that death cannot be
bad for the individual who dies. The Epicurean view is that our ordi-
nary belief death is (or at least can be) bad for the very individual who
dies is based on a set of confusions. After all, the Epicurean reminds us,
badness or misfortune typically involves unpleasant experiences, and
presumably it always requires a subject. But death is an experiential
blank, and it even robs us of a subject. Given that there is no subject
when death takes place, when exactly does the harm, badness, or mis-
fortune of death occur? Given these difficulties, death cannot be bad for
the individual who dies, according to the Epicurean. The commonsense
view is based on a failure to think carefully about the philosophical
issues, and perhaps a conflation of the “body” (that does, lamentably,
molder in the grave or otherwise decay) and the person.
Lucretius adds a fascinating and important argument to the arsenal
provided by Epicurus. Lucretius points out that, on the view according
to which death is nonexistence and an experiential blank, it is the “mir-
ror image” of the time before we were born (or perhaps conceived).
Prenatal and posthumous nonexistence seem to have relevantly simi-
lar features; but since we are indifferent to prenatal nonexistence, we
should be indifferent to posthumous nonexistence. Put in other words,
since we don’t regret that we were born when we were born rather than
significantly earlier, we should not think that it is a bad thing that we
Introduction: Meaning in Life and Death 5
will die when we actually die, rather than later. Lucretius challenges
us to say what the relevant difference is between prenatal and posthu-
mous nonexistence.
I attempt to defend the commonsense view that death can (although
it need not be) a bad thing for the individual who dies.4 First, I take on
the challenges posed by Epicurus and his followers. I claim, following
Thomas Nagel, that death can be a bad thing for the individual who
dies insofar as it is a deprivation of the goods of life; a bit more carefully,
the view is that death is a bad thing for the individual who dies, when
it is in fact a bad thing, insofar as it deprives him of what would be on
balance a good life (or at least a life minimally worth living).5 So the
badness of death (when it is indeed bad) does not consist in anything
“positive,” as it were; it consists entirely in being a deprivation of cer-
tain sorts of goods—goods involved in the continuation of a worthwhile
life. (Of course, not any old deprivation of goods will be a misfortune of
the relevant sort: the fact that I didn’t win the California Lottery today
constitutes a deprivation of certain goods, but it presumably does not
thereby count as a misfortune or bad thing for me.) Insofar as the bad-
ness of death consists entirely in its being a certain sort of deprivation,
it is compatible with there being no unpleasant experiences. I thus need
to defend the thesis that some things can genuinely be bad for an indi-
vidual without involving any unpleasant experiences at all.
In “Why Is Death Bad?” (chapter 2), I (together with my coau-
thor, Anthony Brueckner) begin a defense of the examples provided
by Nagel; that is, we defend his conclusion that some things can be
bad for an individual without involving any unpleasant experiences.6
Nagel’s first example is of a man who is (apparently) “betrayed” by his
so-called friends behind his back. That is, we can suppose that a per-
son’s acquaintances and even relatives regularly get together without
the individual’s ever finding out about these meetings, or experiencing
anything as a result of these meetings.7 At the meetings, the people exco-
riate their “friend,” impugning all his work, contending that he cheats
on his wife, that he is disloyal to his family, and that he is a cowardly
and creepy pervert. Nagel argues that the man is indeed betrayed, and
that he has been wronged; it is not simply that the people have behaved
badly, but a bad thing has happened to the individual himself (although,
admittedly, this is not a bad that is experienced as such).
I agree with Nagel. Consider, also, an example provided by Robert
Nozick.8 Nozick asks us to imagine that someone has secretly placed
a camera in your bedroom and is videotaping all the goings-on. You
(reasonably) believe that you have privacy in your bedroom, but your
activities are beamed by satellite to movie theaters in Outer Mongolia.
We can, as with Nagel’s example, stipulate that you are causally isolated
from Outer Mongolia, and that there are no consequences for you (includ-
ing unpleasant experiences) as a result of the camera in your bedroom.
Nevertheless, Nozick argues that you have been wronged by the place-
ment of the camera in your bedroom; it is not simply that someone has
acted badly; your right to privacy has been violated. I agree with Nozick: a
bad thing has happened to you, even though you are unaware of it.
Now in both Nagel’s case of the man betrayed behind his back and
Nozick’s case of the violation of privacy, it might be thought that at
least it is possible (in some sense) that the individuals find out about the
relevant activities; further, it might be thought that it is this possibility,
with the attendant possibility of unpleasant experiences, that explains
the badness in both cases (if indeed there is badness for the individuals
in question). Nagel, however, has an insightful response:
Loss, betrayal, deception, and ridicule are on this view bad because
people suffer when they learn of them. But it should be asked how
our ideas of human value would have to be constituted to accommo-
date these cases directly instead. One advantage of such an account
might be that it would enable us to explain why the discovery of
these misfortunes causes suffering—in a way that makes it reason-
able. For the natural view is that the discovery of betrayal makes us
unhappy because it is bad to be betrayed—not that betrayal is bad
because its discovery makes us unhappy.9
Given this sort of worry (to which Nagel seeks to respond), however,
it might be helpful to have examples in which we would be inclined to
say that an individual has been harmed (or that a bad thing has hap-
pened to the individual), even though he cannot experience anything
unpleasant as a result of the thing. It seems to me that Nagel has pro-
vided an additional example of precisely the required kind. Nagel asks
us to imagine that an intelligent adult receives a brain injury (perhaps
as a result of an accident or stroke) which leaves him in the “mental
condition of a contented infant, and that such desires as remain to him
can be satisfied by a custodian, so that he is free from care.”10 It would
seem that, after the brain injury, it is both true that the individual does
not have any unpleasant experiences (as a result of the injury) and also
cannot have any such experiences; but, intuitively, he has been harmed
by the stroke or accident—a bad thing has happened to him, although
he has no experiential access to the badness.
Consider, similarly, my revision of Nagel’s betrayal-behind-one’s
back example, which I present in “Death, Badness, and the Impossibility
of Experience,” (chapter 3). Here I employ the resources of “Frankfurt-
cases”—examples provided by Harry Frankfurt which purport to
impugn the Principle of Alternative Possibilities (PAP), according to
which moral responsibility for performing some action requires genu-
ine metaphysical access to an alternative possibility in which one does
otherwise.11 I thus seek to forge a connection between central issues
in the free will/moral responsibility debates and the debates about
the metaphysics of death.12 It will be helpful to have a version of a

Frankfurt-case before us:
Black is a basically nice, but admittedly also slightly naughty, neu-
rosurgeon. He has secretly inserted a chip in Jones’s brain which
enables Black to monitor and control Jones’s brain activities. Black
can exercise this control through a sophisticated computer which
he has programmed so that, among other things, it monitors Jones’s
voting behavior. Black is a committed Democrat, and he has yet
again signed up with the Democratic Party to be of service. (Black
lives in Chicago, where worse things happen, and, as he says, “It
is the least I can do to help.”) If Jones were to show any inclination
to vote for McCain (or, let us say, anyone other than Obama), then
the computer, through the chip in Jones’s brain, would intervene to
assure that he actually decides to vote for Obama and does so vote.
But if Jones decides on his own to vote for Obama (as Black, the old
progressive, would prefer), the computer does nothing but continue
to monitor—without affecting—the goings-on in Jones’s head.
Now suppose that Jones decides to vote for Obama on his own,
just as he would have if Black had not inserted the chip in his head.
It seems, upon first thinking about this case, that Jones can be held
morally responsible for this choice and act of voting for Obama,
although he could not have chosen otherwise and he could not have
done otherwise.13
Return to Nagel’s betrayal-behind-one’s back example, but now
add a Frankfurt-style “counterfactual intervener,” White.14 White is,
like Frankfurt’s Black, a strictly counterfactual intervener. That is, White
plays no role in how the story actually unfolds, because he doesn’t need
to. But we suppose that he can effectively prevent you from ever find-
ing out about the secret meetings and the betrayals that take place at
those unfortunate meetings. So, for example, if someone were about to
call you to inform you of these meetings, White can prevent him from
making the connection; perhaps White can snip the telephone wire, or
White can temporarily paralyze the person, or whatever. We suppose
that White stands by, like Black, disposed to intervene to prevent you
from finding out about the betrayals; given that White is envisaged as
having the power to prevent you from ever finding out anything about
the meetings, you cannot have any unpleasant experiences as a result of
those meetings. And yet it seems that you are harmed by the activities
at those meetings—you have been betrayed, even though you have no
experiential access to the betrayals.
The revised betrayal example and the brain-injury example seem to
be cases in which something bad has happened to an individual and
yet that individual cannot experience anything unpleasant as a result.
Consider, also, an example due to Jeff McMahon.15 In this example,
your daughter is trekking in the Himalayas and tragically dies in an

accident. Before you can possibly find out about this tragedy—say five
minutes after the death—you (independently) die of a sudden heart
attack. (This was not a good day for you.) It seems to me that the death
of your daughter is a bad thing for you; your interests are intertwined
with hers in such a way that her death is a deeply bad thing for you.
And yet, given the circumstances, you cannot (in any relevant sense)
experience anything unpleasant as a result of her death. For the last
five minutes of your life, you have been harmed, although you have no
experiential access to the harm (or its etiology).
So, arguably, there are cases in which something bad happens to an
individual, and yet the individual cannot experience anything unpleas-
ant as a consequence. In this respect, death—construed as a permanent
experiential blank—could be assimilated to these examples. But there
is still an arguably crucial difference between death and the examples
in question: death robs us of a subject, whereas the subject persists in
all of the examples. Thus, we cannot straightforwardly extrapolate from
the apparent badness in the three examples (the brain-injury case, the
revised betrayal case, and the trekking case) to the badness of death.
I nevertheless think that, upon reflection, we should conclude that
death is relevantly similar to the examples under consideration. In devel-
oping my defense of this contention, I discuss the signature structure
of what I have called “Dialectical Stalemates,” and I draw additional
comparisons between the debates about free will and moral responsi-
bility, on the one hand, and the debates about death, on the other (see
chapter 7). I argue that, whereas we have here an argumentative Black
Hole (Dialectical Stalemate) similar to those encountered in the free
will debates (and, indeed, other philosophical contexts such as dis-
cussions of epistemological skepticism), our philosophical response to
the Stalemates here should be different. Although I argue for a restruc-
turing of the debates about free will and moral responsibility, I argue
for a different approach to the debates about the putative badness of
death.16
I have thus sought to reply to the objections raised by Epicurus to the
commonsense view that death can be bad for the individual who dies.
But what about Lucretius’ Mirror-Image Argument? Nagel says that
the asymmetry in our attitudes toward our own past and future nonex-
istence is “the most perplexing feature of our attitude toward death.”17
I (with my coauthor, Anthony Brueckner), have addressed this deeply
puzzling characteristic of our ordinary attitudes.18 Our claim (follow-
ing Parfit) here is that human beings have asymmetric attitudes toward
pleasures and pains; holding other things fixed, and assuming that we
are situated at a particular point in time, we prefer our own pleasures
in the future and pains in the past. The crucial point for us in this con-
text pertains to pleasures (or experiences of a positive nature), not pains;
insofar as death deprives us of future pleasures, there is a reason to be

particularly concerned about death, as opposed to prenatal nonexis-
tence. That is, we argue that the ordinary asymmetry in our attitudes
toward prenatal and posthumous nonexistence is a special case of a
more general asymmetry in our attitudes toward past and future plea-
sures; we prefer our pleasures in the future, holding other things fixed,
and we are relatively indifferent to past pleasures.
Note that this explanation makes some progress insofar as it exhib-
its the asymmetry in our attitudes toward prenatal and posthumous
nonexistence as a special case of a more general asymmetry; but a full
explanation would establish the rationality (or appropriateness) of
the more general asymmetry. I discuss this difficult issue in “Earlier
Birth and Later Death: Symmetry Through Thick and Thin” (chapter
5). Also, there is further discussion of the Brueckner/Fischer approach
to addressing the asymmetry in our attitudes toward prenatal and
posthumous nonexistence, including a defense of this strategy against
alternative strategies in John Martin Fischer and Daniel Speak, “Death
and the Psychological Conception of Personal Identity” (chapter 4) and
“Earlier Birth and Later Death: Symmetry Through Thick and Thin”
(chapter 5).
I have, then, charted out an overall strategy of response to some of
the major concerns about commonsense views about death raised by
Epicurus and Lucretius. I have defended the deprivation account of
death’s badness, and I (together with Brueckner) have shown how our
asymmetric attitudes toward prenatal and posthumous nonexistence
follow from a more general asymmetry (that is arguably defensible).
Now if death can be a bad thing for the individual who dies, it might
seem that we should wish for immortality (that is, eternal life in some
sort of embodied form). Again, it is arguably a part of ordinary think-
ing that immortality would in fact be desirable, given the right circum-
stances. I have again attempted to defend common sense against the
more curmudgeonly view of many philosophers (such as Heidegger
and Bernard Williams) to the effect that immortality would necessarily be
unattractive (if coherent at all). I argue against the immortality curmud-
geons in “Why Immortality Is Not So Bad” (chapter 6), “Epicureanism
About Death and Immortality” (chapter 7), and “Free Will, Death, and
Immortality: The Role of Narrative” (chapter 9).
Return to the Frankfurt-case sketched above. Recall that in the exam-
ple, Jones chooses to vote for Obama and does vote for Obama “on
his own” (and without the intervention of Black). The actual sequence
that issues in Jones’s choice and action flows exactly as it would have,
had there been no Black (or other “counterfactual intervener”). In vari-
ous places, I have argued that we can learn important lessons from the
Frankfurt cases.19 I have argued that the cases are part of an overall
package of considerations that render it plausible that an agent can act
freely and be morally responsible for his actions, even though he lacks
genuine metaphysical access to alternative possibilities. Given that there
can be moral responsibility without the sort of control that involves
access to alternative possibilities, the value of acting so as to be morally
responsible—what we care about in caring about moral responsibil-
ity—cannot be explicated in terms of the value of metaphysical access
to alternative possibilities (“freedom to choose and do otherwise”).
More specifically, I contend that the value of acting so as to be morally
responsible is not the value of “making a difference,” but of “making a
statement” (of a certain sort).20 More specifically, my contention is that
the value of acting so as to be morally responsible is a species of the
value of self-expression.
Of course, I do not think that the value of acting so as to be morally
responsible trumps all other sort of value in all circumstances; nor do I
suppose that the value of self-expression is hegemonic. My contention
is simply that the value of acting so as to be morally responsible—what
we care about in caring about being morally responsible agents—is the
value, whatever that is, of a certain distinctive kind of self-expression.
But what is this kind of self-expression? In a nutshell, my view is that
it is a species of artistic or aesthetic self-expression. When we act freely
(and thus exhibit the kind of control that grounds moral responsibility),
we are, to put it metaphorically, writing a sentence in the stories of our
lives. More specifically, in exhibiting the distinctive control involved in
moral responsibility, we make it the case that our lives have a certain
dimension of value: narrative value. Insofar as having an irreducible
narrative dimension of value presupposes that our lives are indeed nar-
ratives or “stories,” strictly speaking, we are properly thought of as the
authors of the stories of our lives. As such, we are engaging in a species
of aesthetic activity. Note that, although our free actions are a species
of artistic or aesthetic self-expression, the value we place on such self-
expression is not necessarily a specifically aesthetic form of value. My
claim is that the nature of the activity involved in free action is aesthetic
self-expression; further, I contend that its value is the value—whatever
kind of value it is—that we place on this specific sort of aesthetic activ-
ity. Note that the value in question might be “moral” or “ethical” in
a broad sense, or value from the perspective of living a good life; in
addition to leaving the kind of value open, I also mean to leave open the
amount of the value in question. For further discussion, see “Stories and
the Meaning of Life” (chapter 10).
The above claims employ notions that need considerable attention,
even if they cannot be given complete reductive analyses. First, what
exactly is a “narrative” or “story,” strictly speaking, and what is the
putatively irreducible “narrative dimension of value”? Here my work
builds on some of the important work of David Velleman, although I
depart from Velleman in some crucial ways.21
What exactly is a narrative or—in the strict sense—a story? After

all, we can write the “story”—in a loose sense—of just about anything.
We can, for instance, write the story of the table on which my laptop
computer currently sits, or the laptop itself, or the rocks around the
beautiful tree outside my window—or, for that matter, the tree or the
window or . . . But arguably these would not be “stories, strictly speak-
ing”; instead we can call them mere chronicles. What turns a chronicle
into a narrative or story, strictly speaking? To mimic Wittgenstein’s
famous question about what needs to be added to mere behavior to get
action, what ingredient needs to be added to a chronicle to get some-
thing more than a mere chronicle: a full-blooded story?
The answer, according to Velleman, is that a full-blooded story or
narrative is a chronicle that is apt to produce (presumably in a certain
audience) a certain sort of understanding, including a distinctively
affective or emotional component. A story, as opposed to a mere chroni-
cle, is apt (in a certain audience) to produce not just an “aha-moment,”
but an emotionally suffused aha-moment. Arguably, then, a narrative
or story provides a certain sort of understanding or explanation—one
that allows the audience to “comprehend” something in some sense “as
a whole,” and where this comprehension is in part affective (or affec-
tively underwritten). A narrative or story, in a strict sense, is a chronicle
with emotional hooks, as it were.
This view about narrativity has various interesting implications. It
resonates with Martha Nussbaum’s view that reading fiction is neces-
sary in order to gain the kind of insight that is required for deep moral
insight and understanding. On Nussbaum’s view, it is precisely because
stories, strictly speaking, elicit and engage the emotions that they are
crucial to moral understanding. I discuss Nussbaum’s views, and the
role of stories (and hypothetical examples) in moral understanding,
in “Stories” (chapter 8). It is also sometimes claimed that, insofar as
our lives are stories (in a strict sense) or narratives, they must have
a beginning, middle, and (most importantly), an end. That is, some
have claimed that, if we model our lives on narratives, we will have
to say that immortality is incoherent. I explore and (not surprisingly)
reject this suggestion in “Free Will, Death, and Immortality: The Role of
Narrative” (chapter 9).
Given that we have some sort of grasp of what it would be for our
lives to be narratives, what does it mean that our lives have an irre-
ducible narrative dimension of value? This is admittedly a difficult
and delicate issue, but I think it can be given a tolerably clear explica-
tion. Velleman argues that we have a dimension of value that cannot
be reduced to considerations of “welfare” or “utility” experienced at
certain times (or during certain durations); that is, this dimension is
not to be understood in terms of a function that combines such wel-
fare or utility in any way.22 Rather, it is a matter of our lives having
certain structural features, or perhaps it is a matter of the relationships

between various parts of our lives. More specifically, we score higher
on this dimension insofar as the various parts of our lives fit together
in certain ways—ways arguably characteristic of certain (but not all)
deeply resonant stories, and not capturable solely in terms of utility
information.
Unfortunately, Velleman does not give a general account of the rel-
evant sort of relationships between parts of our lives, in virtue of which
we have greater value along the narrative dimension; and I do not think
that I can provide such an account. It is perhaps helpful, however, to
have before us some examples of the sorts of relationship in question.
Velleman offers the following sorts of examples. Arguably, we prefer a
life in which we thrive as a result of learning from our mistakes rather
than from mere accident, or a life in which we flourish as a result of
conscientiousness and hard work, rather than a mere windfall (such as
winning the lottery). A life in which we complete projects or fulfill com-
mitments seems to be preferable to a more fragmented, unconnected
life, even holding fixed the sum total of aggregate utility (welfare) or
even the profile of utility (welfare) over time. Although the examples
are perhaps less plentiful than desirable, and the analysis at best highly
sketchy, I believe that there is something important in Velleman’s
contention: we do seem to care not just about utility (combined and
weighted in various possible ways), but we also care about how the
parts of our lives hang together, as it were. More specifically, we seem
to care about certain relationships that can reasonably be dubbed “nar-
rative” in character, where this rubric groups together a congeries of
different elements of what we can recognize as stories of lives that in
some sense go well.
Despite the somewhat unsatisfactory lack of completeness in Velleman’s
(and my) presentation, I maintain that it is at least plausible to suppose
that our lives have an irreducible dimension of value that can be
described as “narrative.” Note that this notion of narrative value is
crucially different from the notion that is relevant to (say) a reader of
a story or a literary critic (insofar as such critics actually read stories
any longer!); that is, a story can be “good in the telling,” but not score
high along the narrative dimension of value identified above. We can
recognize certain chronicles as stories, strictly speaking, at least in part
in virtue of their tendency to elicit a certain emotional response. So, for
example, if I tell you about a man who tries so hard to succeed that he
thwarts his own efforts, you may well recognize this as a story, strictly
speaking, and not as a mere chronicle. It might not however yet be a
story that’s good in the telling; this depends crucially on how the story
is told. Finally, such a story might pass both the test of being a story,
strictly speaking, and being a story good in the telling, but not score high
along the narrative dimension of value.
Above, I warned against inferring from the nature of our free actions
as aesthetic activity to the conclusion that the value of such activity is
properly considered aesthetic. Additionally, it is extremely important not
to conflate the notion of good in the telling with a higher score along the
narrative dimension of value. A story of one’s life can be rather boring,
from the perspective of (say) a literary critic or potential “reader,” but
score high marks along the narrative dimension of value. It is impor-
tant to note that I do not in any way wish to suggest that we pattern
our lives on good stories (in the sense of good in the telling), or that
we employ primarily aesthetic considerations (in the sense of “good in
the telling”) in our practical reasoning. These views might be defended
by other philosophers, but they are no part of my views.23 For further
discussion, see “Stories and the Meaning of Life” (chapter 10).
I do contend that it is specifically by our acting freely—our exercise
of the distinctive kind of control involved in moral responsibility—that
we make it the case our lives have an irreducibly narrative dimension
of value: “Free Will, Death, and Immortality: The Role of Narrative”
(chapter 9). Although it is hard to prove this thesis, it does seem to me
correct. Imagine, for example, a man who has learned (painfully) from
many years of failure in relationships; perhaps he has engaged in diffi-
cult (and expensive!) psycho-dynamically informed psychotherapy, and
he combines these hard-earned insights with hard work in his new rela-
tionship. Although it is not straightforward or easy for him, he builds
a healthy, satisfying new relationship, and he finds great satisfaction
in it. Contrast this man with one who is equally frustrated in a series
of failed relationships, but engages in no attempt to learn from them
and no subsequent efforts to change his behavior. Rather, his brain is
secretly manipulated (perhaps the neurosurgeon Black, bored with his
role as a purely counterfactual intervener in the Frankfurt-cases, has
run amok) to induce precisely the sorts of change in behavior that come
“the old-fashioned way” for the first man. I think we would all say that
the first man’s life has more value than the second man’s, even though
the utility-information may well be the same in both lives. Further, note
that the behavior is the same in both lives; the only difference is that the
first man’s is free, whereas (after the relevant point) the second man’s is
not. More specifically, I think it is plausible that the first man’s life has
(or scores more highly on) the irreducible narrative dimension of value,
whereas the second man’s lacks this dimension entirely (after the point
of external intervention and manipulation).
More generally, consider scenarios in which individuals are secretly
manipulated in such a way that it would be indisputable that they were
not acting freely. Stipulate that the manipulation is clandestine and inac-
cessible to the conscious awareness. Now, presumably, we can think
about various life-histories involving learning from mistakes, thriving
as a result of hard work and conscientious efforts (rather than sudden
windfalls), and so forth. In all of these cases, I contend that the fact that
the stretches of the lives instantiate the relevant relationships—the rela-
tionships in virtue of which we would (apart form the manipulation)
say that the lives score higher on the narrative dimension of value—is
entirely irrelevant to the overall value of the lives. That is, when the
kind of control associated with moral responsibility is lacking, I do not
have any intuition that it matters that (say) someone has succeeded as
a result of hard work, rather than winning the lottery; that fact does
not matter apart from the underwriting provided by the relevant kind
of free will. And yet in general we do think it matters that our lives (or
stretches of them) instantiate the relationships in question. I conclude
that it is in virtue of our exercising our capacity for free will—more
specifically, our acting freely—that we are the kinds of creatures whose
lives have an irreducible narrative dimension of value. I contend that
by exercising a distinctive kind of control, we endow our lives with this
additional dimension of value, and we thus become the authors of our
life-stories. (Of course, it does not follow, as I have emphasized above,
that in writing the story we should be guided exclusively or primarily
by aesthetic considerations or factors that make for a better story in the
telling. Nor does it follow that I should seek to experience my life as a
narrative, to interpret it as a narrative, and so forth.)
II. CONTINUITY
Followers of Epicurus typically point to the differences between death
and what might be called (perhaps somewhat tendentiously) “standard”
evils (misfortunes or “bad things”). As discussed above, death
(construed as an experiential blank) differs from standard evils insofar
as it does not involve unpleasant experiences. Also, death robs us of
a subject, so that the “existence requirement” (the claim that all evils
require a contemporaneous subject) is not met. Finally, standard evils
are “comparative” in nature; to the extent that something is bad for an
individual, it makes the individual worse off than he otherwise would
be. Typically, then, we would compare an individual’s situation in the
scenario in which he is affected by the purportedly bad thing with the
situation of that very individual in relevant alternative scenarios in
which he is not so affected. But death makes such comparisons impos-
sible. The Epicurean tends to focus on what are deemed paradigmatic
or standard evils, note these salient differences, and conclude that death
cannot be an evil (or bad thing) for the individual who dies.
But there is another way of looking at these matters.24 Common
sense presumably has it that if anything is (typically or at least often)
a bad thing for an individual, it is death; that is, death is arguably a
paradigmatic harm (or bad thing). So whatever other harms there are, our
philosophical theorizing should accommodate death’s being a harm (or
evil or bad thing). Perhaps then we should divide the class of evils (or
bad things) into standard and non-standard subclasses. On this way of
proceeding, the standard evils involve unpleasant experiences, satisfy
the existence requirement, and are comparative in nature. In contrast,
nonstandard evils (including death) would not meet (all of) these cri-
teria. A defender of the common sense view here might insist that the
claim that death can be a bad thing for the individual who dies is on
firmer footing than the alleged requirements that evil requires a sub-
ject, that it involves unpleasant experiences, and that it is essentially
comparative.
How can this debate be resolved? That is, what is the route past
the relevant Dialectical Stalemates? I have suggested what I take to be
a promising path (chapter 7). Here, I simply wish to note that there
might be something to be said for the kind of “continuity” that appears
to underwrite the commonsense view that death can indeed be a bad
thing for the individual who dies. This, roughly speaking, is what I
have in mind. Think of various indisputable and unproblematic harms,
such as pain. If I twist your arm for thirty seconds, thus causing you
pain during this period, I thereby cause you harm. As we ratchet up
the level of pain and/or its duration, presumably we also increase the
level or severity of the harm. If we add tissue damage or damage to the
functioning of the body and/or mind, we again increase the severity
of the harm. Rape, dismemberment, mutilation, and torture are clearly
among the most severe evils one can undergo. Now suppose that (say)
torture is so severe that it issues in the individual’s death. It would
seem natural that this would make the harm of the torture even worse. It
seems to follow that there is a kind of continuity in the nature of harm.
Let me try to say a little more about this notion of continuity. There
are certain features—call them “Level One Features”—that appear to
be arranged on a spectrum of increasing intensity or significance. In
the examples, these features include pain, suffering, and various kinds
of impairment and damage. Further, it is natural to suppose that the
“Level Two Feature” of evil (badness or misfortune) in some sense super-
venes on the Level One Features. At least it seems that as the Level One
Features become more and more significant—that is, as we move along
the spectrum of Level One Features—this is reflected at Level Two in
greater severity of harm. But the Epicurean must say that when we get
to death, there is a radical discontinuity; although death would appear to
be at the far end of the Level One spectrum, this is not reflected continu-
ously at Level Two. Put slightly differently, as the features that intui-
tively underwrite harm-attributions get more and more significant, this
is reflected in greater harm until we get to death, where there is suddenly
no harm at all.
Now I believe that in general we should aim for philosophical
understanding of various phenomena that meets certain desiderata,
including (but, of course, not limited to) elegance and comprehensive-

ness. I also have a preference for such features as symmetry and continu-
ity. (Note that, on this approach, the equilibrium or default position is
symmetry, but explained asymmetry [as with our asymmetric attitudes
toward prenatal and posthumous nonexistence] is perfectly accept-
able.) Granted, the preference is somewhat weak, and can be defeated
or outweighed by other considerations. But there is something to be
said for continuity.
Of course, the Epicurean might well point to continuity as one of
the factors that confuses us. That is, he might grant that philosophical
homeostasis tends toward such features as symmetry and continuity,
and we thus get lulled into thinking that death is just straightforwardly
at the end of the Level Two spectrum that supervenes on the Level
One harm-relevant features. But the Epicurean will insist that this is
a mistake, and that upon more careful reflection, we should resist the
underlying picture that favors continuity. I do not have a knockdown
response to the Epicurean here, but I simply commend the beauty
and appeal of the framework that embodies continuity. Given that the
Epicurean presumably has no knockdown argument for the framework
within which we consider death outside the set of harms rather than a
nonstandard harm, I suggest that the appeal of continuity (understood
and applied as above) might constitute at least a small consideration on
behalf of the ordinary view that death can indeed be a bad thing for the
individual who dies.
Perhaps it is also noteworthy that my approach to immortality
embraces a similar kind of continuity. I argue that immortality can
be (although it need not be) a good thing for the individual who is
immortal. Start with an indisputably rich, attractive, and meaning-
ful human life—filled with pleasant experiences, relationships, and
significant accomplishments. Suppose that the individual dies at 70.
Surely we can get an even more attractive life by adding more pleas-
ant experiences, more or deeper relationships, and/or more significant
accomplishments. And presumably, we can get an even more attractive
life by adding (say) two years; wouldn’t it be better if the life ended
at 72 (other things equal) than 70? And wouldn’t it be even better if it
ended at 75? And so forth.
Presumably, there are features at Level One—such as pleasant expe-
riences, relationships, and accomplishments—that intuitively are the
basis of Level Two properties of lives, such as desirability. Again, as
above, it seems natural to suppose that the Level One features are (or
can be) arranged along a spectrum of increasing significance. Further,
it is extremely plausible to think that there is a supervenience relation
such that as we move along the Level One spectrum, this is reflected
at Level Two. That is, it is intuitively appealing to think that as the
Level One features become more and more significant, the relevant
Level Two features, such as desirability or goodness become greater.

(I have argued that we must be careful to think of moving along the
Level One spectrum in a way that is sensitive to various factors, includ-
ing the nature and distribution of the various features.) In contrast, the
skeptics about immortality—the immorality curmudgeons—appear to
be committed to a radical discontinuity; they will concede that our lives
get better and better as they are more pleasant and increasingly filled
with meaningful relationships and accomplishments—but suddenly
they cannot be good at all when they become infinitely long. As with
death, I find this discontinuity at least a bit jarring and disconcerting. I
prefer continuity.
Now, as above, the immortality curmudgeon will insist that our
general preference for continuity has perhaps lulled us into a mistaken
complacency about immortality. It has, according to the curmudgeons,
inclined us not to focus on the fundamental differences between very long
but finite life and immortal life. And, again, I do not have a knockdown
argument against this view. Rather, I simply report that I do not see
exactly why the indisputable differences between very long but finite
life and immortal life would make a difference as regards desirability. I
consider this one of the most difficult and challenging issues surround-
ing immortality. I have argued that, properly understood, immortal life
would involve a mix of activities that could propel us into the future
with genuine engagement; and I do not see exactly why this engage-
ment would disappear as we move from a very long finite life to an
immoral life. But perhaps this simply reflects my cognitive limitations
or lack of imagination.
In any case, I take it that, absent a decisive argument on either side,
there is something to be said for continuity. Indeed, I prefer a picture of
the possible badness of death and the possible goodness of immorality
that employs a framework that embraces the relevant sort of continuity.
Indeed, perhaps it is a consideration in favor of my overall view that
it is symmetrically continuous, as it were; the continuities in the consid-
erations pertinent to death and immortality turn out to be isomorphic.
At the very least, I hope that invocation of the notion of continuity can
help to structure and illuminate key aspects of the debates about the
(possible) badness of death and (possible) goodness of immortality; the
proponents of common sense tend to be driven by certain sorts of con-
tinuities, whereas the skeptics are driven by salient discontinuities.
III. SELF-EXPRESSION: SOME SPECULATIONS
III.1. Close Calls Versus Clear Cases

In this section, I take the liberty of offering a few suggestions that are
(if this is possible!) even more speculative, and less tightly argued for,
than those of the previous sections. I hope that, even if they are not
entirely persuasive, they will be to some extent both illuminating and
suggestive. The ideas all pertain—in some way or another—to the cen-
tral notion of self-expression.
I have argued that the “value” of acting so as to be morally respon-
sible—what we care about in caring about moral responsibility—is
making a statement (of a certain sort), not making a difference (of a
certain sort). Here I rely to some extent on what I take to be the moral
of the Frankfurt stories—that there can be cases of moral responsibility
in which the relevant agent lacks genuine metaphysical access to the
appropriate kind of alternative possibilities.25 Given that what we care
about in caring about moral responsibility is present in such cases, we
cannot construe this as involving the importance of making a differ-
ence. Rather, upon reflection, I believe that it is plausible that the value
of acting so as to be morally responsible can be understood as a species
of the value—whatever that is—of artistic self-expression. After all, in
acting freely, we are writing a sentence in the book of our lives—we
are so acting that our lives have an irreducibly narrative dimension of
value (a dimension that presupposes that our lives are indeed stories,
strictly speaking).
I suggest that this explanation of the value of acting so as to be
morally responsible might illuminate an extremely puzzling disagree-
ment among philosophers who have written about free will and moral
responsibility. Some philosophers anchor free will and moral responsi-
bility in contexts in which “reason” (presumably, practical reason) does
not clearly dictate a correct path. Of course, there are various such con-
texts, but the key is that the agent’s reasons for action do not clearly
or obviously recommend one course of action over others apparently
available to the agent. Perhaps these cases involve conflicts between
morality and prudence, or conflicts internal to morality or prudence;
maybe in these contexts reasons are comparable but “close” or maybe
the reasons are incommensurable. In any case, in Close Calls practical
reason does not commend a unique choice and action. The proponents
of Close Calls as anchoring free will and moral responsibility point to
such contexts as the fundamental or basic locus of free will and moral
responsibility; they sometimes claim that it is only in such contexts that
an agent exhibits “true freedom” and moral responsibility, and they
typically wish to “trace” all genuine instances of free will and moral
responsibility back to Close Calls.
On the other hand, other philosophers find this view highly implau-
sible; they think of Close Calls as contexts characterized by arbitrariness,
which they take to be antithetical to the sort of control that is involved
in free will and moral responsibility. The proponents of Clear Cases
as anchoring free will and moral responsibility contend that it is only
when practical reason clearly dictates a unique path that we are truly
free and morally responsible; typically, such philosophers will claim

that all cases of genuine free will and moral responsibility must be
traceable back to Clear Cases.
It has long been puzzling to me how to explain the debate between
proponents of Close-Call freedom and Clear-Case freedom. It has
seemed to me that the differences in views must be based on a differ-
ence at a deeper, more fundamental level. It might be suggested that
the proponents of Close-Call freedom must be implicitly adopting the
“make-a-difference” view about the value of acting so as to be morally
responsible, whereas the proponents of Clear-Case freedom must be
implicitly adopting the “make-a-statement” view. The “make-a-differ-
ence” view may in fact motivate some proponents of Close-Call free-
dom. But I believe that a proponent of Close-Call freedom would wish
to accommodate the moral of the Frankfurt stories; that is, nothing in
the view that Close-Call freedom is the anchor of free will and moral
responsibility requires that the relevant agent have genuine metaphysi-
cal access to alternative possibilities (and thus the power to make a dif-
ference of the relevant kind). That is, the main point emphasized by
the proponent of Close-Call freedom is that agents are truly free when
they make a choice that is not uniquely dictated by practical reason;
this point is completely compatible with the relevant agent’s not hav-
ing freedom to choose and do otherwise. (Note that it is thus a mistake
to assimilate Close-Call freedom to the traditional doctrine of the liberty
of indifference or to assimilate the debate between the proponents of
Close-Call freedom and Clear-Case freedom to the debate between the
proponents of liberty of indifference and the proponents of liberty of
spontaneity.)
Consider the possibility that the more fundamental idea is that the
value of acting so as to be morally responsible is the value of a cer-
tain distinctive kind of self-expression. If this is correct, we can perhaps
illuminate the debate between the proponents of Close-Call freedom
and the proponents of Clear-Case freedom. Perhaps the proponents of
Close-Call freedom are emphasizing the “creativity” component of self-
expression; here the idea is that nothing external to the agent, including
the configuration of practical reasons—requires any particular decision
or action. Rather, the decision comes entirely from the unconstrained
creative activity of the agent. In contrast, it might be that the proponent of
Clear-Case freedom emphasizes a different aspect of self-expression—
the idea of “endorsement.” Sometimes self-expression involves ratio-
nal endorsement or agreement with what one takes to be the course of
action or policy dictated by reason.
Although I have only given the bare bones of my suggestion here, I
have broached the idea that we can explain the apparently intractable
debates between proponents of Close-Call freedom and proponents
of Clear-Case freedom by invoking a deeper agreement on the value
of self-expression. Both parties to the dispute might be understood as

adopting the fundamental value of self-expression. But the disagree-
ment issues from an emphasis on two different elements or “faces” of
self-expression: pure creativity and reflective endorsement. My own
view is that we should look to neither Close-Call contexts nor Clear-Case
contexts as the anchors of free will and moral responsibility. I believe
that it is a mistake to suppose that either of these sorts of contexts is
the home of true freedom, or that we must trace back to such contexts
to anchor our subsequent attributions of freedom and moral responsi-
bility. Perhaps a consideration in my favor here issues precisely from
acceptance of the very idea under discussion that what we value in
acting so as to be morally responsible is self-expression of a signature
sort. If this is correct, one will find unsatisfactory any view that flows
from only a part or element of the relevant sort of self-expression; and it
is clear that neither pure creativity nor reflective endorsement captures
the entirety of the notion of self-expression.
Further, I believe that, whereas pure creativity and reflective endorse-
ment capture notions that are importantly related to the relevant ele-
ments of self-expression, each actually is a distortion of the basic notion
of self-expression that is involved in moral responsibility. The problem
with pure, unconstrained creativity is that it is hard to see how this dif-
fers from mere arbitrariness, and it is also hard to see how such activ-
ity flows from the agent (the agent’s values and reasons) in the right
way. The problem with reflective endorsement is that we sometimes act
freely without reflective endorsement; that is, we sometimes freely do
what we do not believe to be normatively defensible. Weakness of the
will is a fact of life, and it presents a devastating problem for reflective
endorsement accounts. For example, it seems to me to be the basis of a
decisive objection to the “normative” approach to moral responsibility
pioneered by Gary Watson—an objection I do not believe Watson or his
followers have ever adequately addressed.26
III.2. Authorship, God, and Freedom

Some of the proponents of irreducible Agent-Causation have con-
tended that our distinctive agency consists (in part) in a Godlike power
to create ex nihilo. But the problems with irreducible agent-causation
(causation by an agent that is not reducible to causation by agent-
involving events) and divine creation ex nihilo can be seen to be similar,
at least insofar as the notions are taken to play a certain theoretical role.
If a God who can create ex nihilo is posited as part of a Cosmological
Argument—an argument according to which ordinary attributions of
causation or perhaps ordinary explanation can only be defensible if
there is an uncaused cause (or unexplained explainer)—it is natural to
respond that such a God is merely the label for a desideratum. That is, it
might seem that the only way we can defend our ordinary attributions
of causation or our ordinary explanations is if there were an entity with

certain properties; but, even if this were true, it would only point to the
desirability of the existence of such an entity. It would not thereby pro-
vide any independent evidence that this sort of entity actually exists.
On this view, then, “God” (construed as purporting to refer to a first
cause or unexplained explainer) is the name of a solution to a problem,
not the solution itself.
Exactly the same sorts of consideration apply to irreducible agent-
causation. It might seem that the only way to make sense of the ideas
that our choices are not causally determined by previous events (together
with the laws of nature) and yet that we are in control of our choices is to
posit irreducible agent-causation. But, as with God as First Cause, this
power is then simply the label for a solution to the problem of explaining
the relevant sort of control, but we do not yet have any independent evi-
dence of its existence (independent, that is, of our wishful thinking).27
Roderick Chisholm contends that our agency can be patterned after
God’s capacity for creation ex nihilo.28 I would suggest that we think
of ourselves as “in God’s image” in a different way. Perhaps it would
be illuminating to think of being similar to God in being an “author,”
but not in the sense of creator ex nihilo. My claim is that the relevant
similarity is authorship, although divine and human authorship (of the
relevant kind) is interestingly different. I have suggested that we think
of ourselves as authors in the sense that we freely endow our lives with
a special sort of meaning: an irreducible narrative dimension of mean-
ing. Recently, Velleman has argued for an even more robust connection
between our free action and authorship. Velleman has argued that we
can fruitfully think of ourselves as the authors of improvisational dra-
mas in which we are the protagonists.29
Here, I wish to focus (very briefly) on God’s side of the equation. If
God is an author, he is an author of a story in which we are characters.
It might be helpful to think of our relationship to God as analogous
to the relationship between an author and the characters in his novel.
More specifically, this model might be at least of some help in concep-
tualizing important features of the great traditional problem of the rela-
tionship between God’s omniscience and human freedom. We wish to
capture God’s sovereignty while at the same time endowing human
beings with some degree of autonomy. Perhaps we can here invoke an
analogy with the relationship between an author and his characters, at
least on a certain interpretation of this relationship.
Although there is perhaps a clear sense in which an author has sov-
ereignty, many authors have emphasized the constraints imposed by
their characters. For example, some authors (most notably, perhaps,
Stephen King) have emphasized that, especially as the stories unfold,
the traits of the characters impose certain limits on the directions in
which the story can plausibly go.30 The issues here are very nuanced
and complex, but perhaps it would be worth considering the idea

that the characters impose certain constraints on the authors that flow
(in some difficult-to-characterize way) from the developing story. At
least I wish to suggest—in a very tentative and provisional way—that
it might be fruitful to explore the contention that our relationship to
an omniscient God (if such a God exists) could be modeled on the
relationship between an author and his characters. After all, arguably
the authors have a certain autonomy that is nevertheless compatible
with the sovereignty of the author. This suggestion is not entirely new,
nor have I developed it in any detail here; I hope to give it further
development in future work. Here I wish simply to highlight an alter-
native conception of our creativity and authorship—a conception that
fits well with my contention that in exercising our freedom, we are
engaging in a distinctive kind of artistic self-expression. This suggests
a way—very different from Chisholm’s suggestion—in which we
are arguably created in God’s image, or at the least, have properties
importantly similar to those attributed to God in the Judeo-Christian
tradition.
NOTES
1. Presumably, pain and suffering is intrinsically bad, but they may not
be the only intrinsic bads; nor is it obvious that states of affairs involving pain
and/or suffering need be all-thing-considered or on-balance bad. These are all
difficult and contentious issues.
2. For an interesting exploration of some of the complexities involved in
defining “death,” see Fred Feldman, Confrontations with the Reaper (New York:
Oxford University Press, 1992).
3. See Bad Religion’s album, How Could Hell Be Any Worse?
4. Death is not necessarily bad for the individual who dies. An individual
in terrible pain or significant disability may welcome death under certain cir-
cumstances, and we might think that death would not be a bad thing for the
individual in such circumstances.
5. For Nagel’s classic treatment of these issues, see Thomas Nagel,
“Death,” in his Mortal Questions (Cambridge, U.K.: Cambridge University
Press, 1979), pp. 1–10; reprinted in John Martin Fischer, ed., The Metaphysics of
Death (Stanford, Calif.: Stanford University Press, 1993), pp. 61–9. (Subsequent
references to the Nagel piece will be to the reprinted version.)
6. For further discussion of these examples, see “Death, Badness, and the
Impossibility of Experience” (chapter 3) and “Epicureanism About Death and
Immortality” (chapter 7).
7. Further, we can assume that the individual does not fail to get some
good he otherwise would have got, but for the meetings.
8. Robert Nozick, “On the Randian Argument,” in Socratic Paradoxes
(Cambridge, Mass.: Harvard University Press, 1999), pp. 249–64.
9. Nagel, “Death,” p. 65.
10. Ibid.
11. The original presentation is in Harry Frankfurt, “Alternate Possibilities

and Moral Responsibility,” Journal of Philosophy 66 (1969), pp. 829–39. The litera-
ture on such examples is huge, but we need not delve into it in any detail here.
For some preliminary discussion, see John Martin Fischer, “Recent Work on
Moral Responsibility,” Ethics 110 (1999), pp. 93–139; also, see David Widerker
and Michael McKenna, Moral Responsibility and Alternative Possibilities: Essays on
the Importance of Alternative Possibilities (Aldershot, U.K.: Ashgate, 2003).
12. For additional connections, see chapter 7.
13. This is a slightly updated version of a Frankfurt-case I presented origi-
nally in John Martin Fischer, “Responsibility and Control,” Journal of Philosophy
79 (1982), pp. 24–40.
14. See chapter 3; for further reflections, see chapter 6.
15. Jeff McMahan, “Death and the Value of Life,” Ethics 99 (1988), pp. 32–61;
reprinted in Fischer, ed., The Metaphysics of Death, pp. 233–66.
16. For my argument that we should restructure the debates about free will
and moral responsibility, see Fischer, The Metaphysics of Free Will.
17. Thomas Nagel, The View from Nowhere (New York: Oxford University
Press, 1989), p. 228.
18. Anthony Brueckner and John Martin Fischer, “Why Is Death Bad?,”
chapter 2; for additional discussions, see Anthony Brueckner and John Martin
Fischer, “The Asymmetry of Early Death and Late Birth,” Philosophical Studies
71 (1993), pp. 327–31; and “Death’s Badness,” Pacific Philosophical Quarterly 74
(1993), pp. 37–45.
19. See, for instance, John Martin Fischer, The Metaphysics of Free Will: An
Essay on Control (Oxford: Blackwell Publishers, 1994); John Martin Fischer and
Mark Ravizza, Responsibility and Control: A Theory of Moral Responsibility (New
York: Cambridge University Press, 1998); and John Martin Fischer, My Way:
Essays on Moral Responsibility (New York: Oxford University Press, 2006).
20. See John Martin Fischer, “Responsibility and Self-Expression,” Journal
of Ethics 3 (1999), pp. 277–97; reprinted in Fischer, My Way: Essays on Moral
Responsibility, chapter 5, pp. 106–23.
21. J. David Velleman, “Well-Being and Time,” Pacific Philosophical Quarterly
72 (1991), pp. 48–77; reprinted in Fischer, ed., The Metaphysics of Death, pp. 327–
57; and “Narrative Explanation,” Philosophical Review 112 (2003), pp. 1–26.
22. Ibid.
23. In his review of the photography exhibit, “Making It New: The Art and
Style of Sara & Gerald Murphy” at the Williams College Museum of Art in The
New Yorker (August 6, 2007), pp. 74–5. Peter Schjeldal says:
The most revelatory and moving item in the show for me is a letter from
Zelda Fitzgerald, following Scott’s death, in 1940. She writes that Scott’s love
of the Murphy’s reflected a “devotion to those that he felt were contributing
to the aesthetic and spiritual purposes of life.” There is a world of excitement
and woe in that conflation of the aesthetic and the spiritual. It’s a madness,
which life will punish. (p. 75)
I have warned of similar conflations. In the epigram to this introductory
essay, Gary Jules reminds us of the world’s madness. We will no doubt be
punished for our various kinds of madness, even if we avoid certain errors of
conflation.
24. For further discussion, see John Martin Fischer, “Introduction: Death,
Metaphysics, and Morality,” in Fischer, ed., The Metaphysics of Death, pp. 3–30,
esp. pp. 23–4.
25. I believe that the Frankfurt-cases are just one “route” to the conclusion
that moral responsibility does not require the kind of control that involves free-
dom to choose and do otherwise. There are also paths that follow the work
of Peter Strawson and Daniel Dennett to get to the same conclusion; Peter
Strawson, “Freedom and Resentment,” Proceedings of the Aristotelian Society 48
(1962), pp. 1–25; Daniel Dennett, Elbow Room (Cambridge: MIT Press, 1984); and
Freedom Evolves. There is a sustained development of a Strawsonian critique of
the Principle of Alternative Possibilities (PAP) in R. Jay Wallace, Responsibility
and the Moral Sentiments (Cambridge, Mass.: Harvard University Press, 1994)
So there is more than one way to reject PAP. Further, they seem to me to
be entirely compatible and mutually reinforcing. The situation here thus may
be a bit different from that of a man described by my wife. The man came to
her Buddhist temple near Pasadena, California. He was elderly and concerned
about death, and he was engaging in the preparations for death prescribed by
various of the major religions: Christianity, Judaism, Hinduism, Buddhism, and
so forth. Better safe than sorry, I suppose, but . . .
26. Gary Watson, “Free Agency,” Journal of Philosophy 72 (1975), pp. 205–20.
For additional important work by Watson, see Gary Watson, Agency and
Answerability (Oxford; Clarendon Press, 2004).
27. Gary Watson makes this point in “Free Will and Free Action,” Mind 96
(1987), pp. 145–72.
28. Roderick Chilshom, “Human Freedom and the Self,” reprinted in Gary
Watson, ed., Oxford Readings in Philosophy: Free Will (Second Edition), (New
York: Oxford University Press, 2003), pp. 26–37.
29. David Velleman, How We Get Along, forthcoming, Cambridge University
Press; a model for what Velleman has in mind might be Larry David’s HBO
television show, Curb Your Enthusiasm, in which Larry David plays himself (and
the script is not fully written out in advance, so the actors improvise, at least
within certain constraints).
30. Stephen King develops an account of authorship which emphasizes
the autonomy of his characters in On Writing (New York: Pocket Books, 2002).
The mystery writer, Barbara Seranella, the sister of a close friend of mine, once
told me that she always starts with a corpse and a bunch of characters. She
then figures out “who dunnit” (personal conversation, 2006) This is clearly an
even more “extreme” version of the model described by Stephen King. Sadly,
Seranella died at a tragically young age in 2007, awaiting a liver transplant that
never came.
Seranella was the author of the “Munch Mancini” mystery series, culminat-
ing in An Unacceptable Death. The “About Barbara” part of “Barabara Seranella’s
Webpage,” accessed Monday, January 22, 2007 says:
Seranella was born in Santa Monica, California and grew up in Pacific
Palisades. After a restless childhood that included running away from home
at 14, joining a hippie commune in the Haight, and riding with outlaw
motorcycle clubs, she decided to settle down and do something normal so
she became an auto mechanic.
She worked at an Arco station in Sherman Oaks for five years and then
a Texaco station in Brentwood for another twelve. At the Texaco station she
rose to the rank of service manager and then married her boss. Figuring
she had taken her automotive career as far as it was going to carry her, she
retired in 1993 to pursue the writing life.
Seranella’s books have been hailed for their “gritty realism, smart plot-
ting, taut suspense, and [their] highly original heroine.”
A Booklist review of An Unacceptable Death says:
Mancini herself has crawled up from the streets. As an ex-abuse victim, ex-
prostitute, ex-biker old lady, ex-drug addict, she is both forever conscious of
how lucky she is to be one of the few to escape and how unlucky the many
others are who never do; this perspective, plus street smarts, enables her to
go undercover convincingly.
In Fall 2006, Barbara Seranella won the Dennis Lynch Memorial Award for
Social Consciousness in Crime Fiction.
2
Why Is Death Bad?
Anthony L. Brueckner and John Martin Fischer
I find this in my diary, written twenty and more years ago:

People say of death, ’There’s nothing to be frightened of. ’They say it quickly, casually.
Now let’s say it again, slowly, with re-emphasis. ’There’s NOTHING to be frightened
of. ’Jules Renard. ’The word that is most true, most exact, most filled with meaning,
is the word “nothing”.
Julian Barnes, Nothing To Be Frightened Of
I. WHY IS DEATH BAD?

It seems that, whereas a person’s death needn’t be a bad thing for
him, it can be. In some circumstances, death isn’t a “bad thing” or
an “evil” for a person. For instance, if a person has a terminal and
very painful disease, he might rationally regard his own death as a
good thing for him, or at least, he may regard it as something whose
prospective occurrence shouldn’t be regretted. But the attitude of
a “normal” and healthy human being—adult or child—toward the
prospect of his death is different; it is not unreasonable in certain
cases to regard one’s own death as a bad thing for oneself.1 If this is
so, then the question arises as to why death is bad, in those cases in
which it is bad.
If one believes in an afterlife, one could explain how death (con-
ceived of roughly as the cessation of bodily functioning) can be bad
insofar as it can involve eternal torment—an indefinitely long sequence
of (highly) unpleasant experiences. Of course, on this sort of account,
death needn’t be bad, even for a normal and healthy human being, since
he may experience eternal bliss in the afterlife. If there is an afterlife, and
for some it includes unpleasant experiences, then this would explain
how death can be a bad thing, but it is controversial whether there is an
afterlife. Since it is quite possible to deny the controversial assumption
that there is an afterlife and yet regard death as a bad thing, it would be
desirable to produce an explanation of death’s badness which doesn’t
27
presuppose that there are experiences after death. Many have thought
that such an explanation can be given.
If death can be a bad thing for a person, though not in virtue
of including unpleasant experiences of that person, then death is
a bad thing for a person in a way that is different from the way in
which, say, pain is a bad thing for a person. That is, some things
which are bad (or evil) for a person (such as pain) are “experienced
as bad by the person,” whereas other things which are bad for a
person (such as death) are not (ever) experienced as bad by the per-
son.2 Death, then, is assimilated to such bads as betrayal by a friend
behind one’s back, which, though never experienced as bad (one
never finds out and suffers no bad consequences), are nevertheless
bad for a person.3
Let’s suppose that some things which are never experienced as
bad by a person are nevertheless bad for the person. Death could
then be an experiential blank and still be a bad thing for an individ-
ual. And one plausible explanation of why this is so is that death
(though an experiential blank) is a deprivation of the good things of
life. That is, when life is, on balance, good, then death is bad insofar
as it robs one of this good: if one had died later than one actually
did, then one would have had more of the good things in life. This
is the sort of explanation of death’s badness which is adopted by
Thomas Nagel.4
But a problem emerges. We intuitively think that it is appropriate
to have asymmetric attitudes toward prenatal nonexistence and death.
We think that it is reasonable to regard death as a bad thing in a way in
which prenatal nonexistence is not. If death involves bad experiences in
an afterlife, then this asymmetry could be explained. But we are assum-
ing here that death’s badness is not experienced as bad by the individ-
ual who dies. If this is so, how can we explain the intuitive asymmetry
between prenatal and posthumous nonexistence? Both periods are,
after all, experiential blanks. And it seems that prenatal nonexistence
constitutes a deprivation in a sense analogous to that in which death
is a deprivation: if a person had been born earlier than he actually was
born, then he would have had more of the good things in life. (When it
is supposed that one is born earlier here, we hold fixed the date of one’s
death. Similarly, when it is supposed above that one dies later, we hold
fixed the date of one’s birth.) Being born at the time at which one was
born (rather than earlier) is a deprivation in the same sense as dying at
the time when one dies (rather than later). Both Epicurus and Lucretius
argued that our ordinary asymmetric attitudes are irrational and since
we don’t regret prenatal nonexistence, we ought not regard death as a
bad thing. If death is a bad insofar as it is a deprivation, the challenge
posed by Epicurus and Lucretius is pressing: why should we treat pre-
natal and posthumous nonexistence asymmetrically?
Why Is Death Bad? 29
One way to respond to the challenge (and thus defend the Nagelian
explanation of death’s badness) is to say that, whereas one could (logi-
cally) have lived longer, it is logically impossible that one should have
been born much earlier. Further, the claim is that it is irrational (or impos-
sible) to regret that a proposition which is necessarily false isn’t true.5
This response is unsatisfying. It is not clear that it is logically impossible
that an individual should have been born substantially earlier than he
actually was. It is not at all clear, for instance, that Socrates—the very
same Socrates—couldn’t (logically) have come into being ten years ear-
lier than he in fact did. Why exactly should (roughly) the actual time of
one’s birth be an essential property of a person? Given that the essenti-
ality of the actual time of birth is a controversial metaphysical claim, it is
unsatisfying to use it as part of an explanation of the intuitive asymme-
try.6 The explanation will not be acceptable to anyone who denies the
assumption.7 If it is at least logically possible that one should have been
born much earlier (and no reason has been offered to rule this out), then
we still need to develop a response to the challenge raised by Epicurus
and Lucretius (insofar as we cling to the explanation of death’s badness
in terms of deprivation).
Recently, Derek Parfit has suggested another response.8 His position
could be put as follows:
We have a (not irrational) bias toward the future to the extent that
there are cases where we are indifferent toward (or care substantially
less about) our own past suffering but not indifferent toward our own
future suffering. Since there are such cases, and the attitudes therein
seem rational, the general principle that it is always rational to have
symmetric attitudes toward (comparable) past and future bads is
false, and so it might be true that it isn’t irrational to have asymmet-
ric attitudes toward our own past and future nonexistence (where
such periods of nonexistence are taken to be bads). Thus, death could
be considered a bad thing for us, and yet we needn’t assume sym-
metric attitudes toward death and prenatal nonexistence.
Consider Parfit’s example:
I am in some hospital, to have some kind of surgery. This kind of

surgery is completely safe, and always successful. Since I know this,
I have no fears about the effects. The surgery may be brief, or it may
instead take a long time. Because I have to co-operate with the sur-
geon, I cannot have anaesthetics. I have had this surgery once before,
and I can remember how painful it is. Under a new policy, because
the operation is so painful, patients are now afterwards made to for-
get it. Some drug removes their memories of the last few hours.
I have just woken up. I cannot remember going to sleep. I ask my
nurse if it has been decided when my operation is to be, and how
long it must take. She says that she knows the facts about both me
and another patient, but that she cannot remember which facts apply
to whom. She can tell me only that the following is true. I may be the
patient who had his operation yesterday. In that case, my operation
was the longest ever performed, lasting ten hours. I may instead be
the patient who is to have a short operation later today. It is either true
that I did suffer for ten hours, or true that I shall suffer for one hour.
I ask the nurse to find out which is true. While she is away, it is
clear to me which I prefer to be true. If I learn that the first is true, I
shall be greatly relieved.9
Parfit’s claim is that it seems to be a deep-seated feature of us that

we regard our own past and future sufferings asymmetrically. He
doesn’t explicitly defend the rationality of this sort of asymmetry, but
he has pointed to a class of examples involving bads other than death in
which it doesn’t appear obviously unreasonable to hold asymmetric
attitudes.10
Let us grant, for the sake of argument, that Parfit is correct about his
example. The problem is that it cannot be extended to the case of death.
The reason is that Parfit’s case involves a bad for a person which is expe-
rienced as bad by the person. One’s own pain is perhaps paradigmatic of
such bads. But death is not a bad of this kind; indeed, the entire problem
of justifying our intuitive asymmetric attitudes arises precisely because
death is a bad for a person which is not experienced as bad by the per-
son. Further, it seems that it is plausible to suppose that Parfit’s conclu-
sion will only apply to cases involving bads experienced as bad by the
person. Cases which are structurally similar to Parfit’s except involving
bads not experienced as bad by the person yield symmetric attitudes.
Suppose, for instance, that you know that either some friends of
yours have betrayed you behind your back nine times in the past or
some friend will betray you behind your back once in the future. Here,
it seems that you should prefer the one betrayal in the future (given that
the betrayals are comparable, etc.). It also appears that, given a choice
between being mocked once behind your back in the past and being
similarly treated once in the future, you should be indifferent. (Of course,
we assume here that you know that you can have no effect on the future
events).11 These cases suggest that Parfit’s point only applies to the class
of bads experienced as bad by the person, and not to the class of bads
(like death) which are not experienced as bad by the person.
Note that there are two different kinds of cases within the class of
things which a particular person might reasonably regret (or wish
wouldn’t happen or take to be bad), but which he himself doesn’t
experience as bad. One kind contains things which no person experi-
ences as bad (such as death). Another kind contains things which are
experienced as bad by another person (such as another’s pain). If it is
reasonable to take temporally symmetric attitudes toward regrettable

things which we don’t experience as bad and which no one experiences
as bad, then it shouldn’t be surprising that we take temporally symmet-
ric attitudes toward regrettable things which are experienced as bad by
others. And Parfit has produced just such an example:
I am an exile from some country, where I have left my widowed
mother. Though I am deeply concerned about her, I very seldom get
news. I have known for some time that she is fatally ill, and cannot
live long. I am now told something new. My mother’s illness has
become very painful, in a way that drugs cannot relieve. For the next
few months, before she dies, she faces a terrible ordeal. That she will
soon die I already knew. But I am deeply distressed to learn of the
suffering that she must endure.
A day later I am told that I had been partly misinformed. The facts
were right, but not the timing. My mother did have many months of
suffering, but she is now dead.12
Parfit claims, about this example, that the new piece of information—
that my mother’s suffering is in the past—should not have a crucial
impact on my attitude. Concerning the suffering of others it is rational to
have temporally symmetric attitudes. This is precisely what one should
expect in the light of the foregoing discussion of the appropriateness
of temporally symmetric attitudes toward certain bads not experienced
as bad by the person—those not experienced by anyone. The difference
between our symmetric attitudes toward another’s past and future suf-
fering and our asymmetric attitudes toward our own past and future
suffering is a special case of the difference between our attitudes toward
bads not experienced by us and bads experienced by us. If this is cor-
rect, it is appropriate to have temporally symmetric attitudes toward the
class of regrettable things experienced by others, even if it is appropriate
to have temporally asymmetric attitudes toward the class of regrettable
things experienced by us.13 Thus Parfit’s own example highlights the
inadequacy of the present response to the challenge posed by Epicurus
and Lucretius, namely the response suggested by Parfit’s examples of
temporally asymmetric attitudes toward experienced bads.
It might seem appealing to suggest that what makes death a bad
thing for a person is that it is the deprivation of good things already
had by the person. On this account, the asymmetry between our atti-
tudes toward prenatal and posthumous nonexistence is due to the fact
that the time before our birth cannot be conceived as a deprivation
of good things we have already had, whereas the time after our death
clearly can be so conceived. But why exactly should we care especially
about the lack of good things we already have had, in comparison
with the lack of good things which we could have had, had we been
born earlier?
The plausibility of the suggestion may come from a psychological

truth which says that, in general, if a person has experienced a good
thing and then been deprived of it, he tends to lament its absence (to
“miss it”) in a way in which a person who has never experienced the
good doesn’t. If a person has regularly drunk fine wines with dinner, he
regrets the lack of a fine wine at tonight’s dinner more than someone
who has never had a fine wine with dinner.
But why would one regret the absence of something good to which
one has grown accustomed? Presumably, because one tends to be frus-
trated by the lack of such goods—their absence causes unpleasant experi-
ences. When a person accustomed to fine wines must do without, he is
likely to have unpleasant experiences caused by the (partially invol-
untary) comparison of his present quite ordinary wine with his past
delightful wines. In general, it is true that, when one is accustomed to a
good thing, its absence causes unpleasant experiences and is therefore
especially regrettable.
But clearly this principle is not applicable to death, since death
deprives a person of goods without causing any experiences at all (accord-
ing to our supposition). The psychological principle may apply to bads
which are experienced as bad by a person (or which cause unpleasant
experiences had by the person), but it doesn’t apply to death, since it is
not such a bad. So this explanation of our asymmetric attitudes suffers
from the same problem as the above strategy. Suppose, on the other
hand, that we do not appeal to the psychological principle and instead
conceive of death as a bad which is not experienced. Then, insofar as
it is held that in regretting the prospect of death we regret the future
deprivation of goods we have already had, it would be equally reason-
able to regret the prenatal deprivation of such goods, goods which, we
now know, could have graced our life had it begun earlier.
If death is taken to be a bad thing for a person, and it is appropriate to
take symmetric attitudes toward past and future bads that are not expe-
rienced as bad by the person, then either we ought radically to revise
our attitudes toward prenatal nonexistence, or we haven’t explained
why death is a bad thing for a person. In Annie Hall, Woody Allen says,
“We have two complaints about life. First, life is terrible. And second,
life is too short.” If life is terrible, it is—in the typical case—because of
bad experiences. But if life is too short, why?
II. WHY DEATH IS BAD

Imagine that you are in some hospital to test a drug. The drug induces
intense pleasure for an hour followed by amnesia. You awaken and ask
the nurse about your situation. She says that either you tried the drug
yesterday (and had an hour of pleasure) or you will try the drug tomor-
row (and will have an hour of pleasure). While she checks on your
status, it is clear that you prefer to have the pleasure tomorrow. There is
a temporal asymmetry in our attitudes to “experienced goods” which is
parallel to the asymmetry in our attitudes to experienced bads: we are
indifferent to past pleasures and look forward to future pleasures.
Perhaps it is this temporal asymmetry in our attitudes toward
certain goods, and not the asymmetry in our attitudes toward bads,
which explains our asymmetric attitudes toward prenatal and posthu-
mous nonexistence. Death is a bad insofar as it is a deprivation of the
good things in life (some of which, let us suppose, are “experienced
as good” by the individual). If death occurs in the future, then it is a
deprivation of something to which we look forward and about which
we care—future experienced goods. But prenatal nonexistence is a
deprivation of past experienced goods, goods to which we are indiffer-
ent. Death deprives us of something we care about, whereas prenatal
nonexistence deprives us of something to which we are indifferent.
Thus we can defend Nagel’s account of the badness of death by
explaining the asymmetry in our attitudes toward prenatal and posthu-
mous nonexistence. This explanation makes use of a principle clearly
related to (but different from) Parfit’s principle concerning the asym-
metry in our attitudes toward past and future experienced bads. If we
have asymmetric attitudes toward past and future experienced goods,
then death is a bad thing in a way in which prenatal nonexistence is
not.14
Let us end with a fanciful example that illustrates the present point.
It is now 1985 and you will live eighty years in any case. Suppose you
are given the following choice. Either you were born in 1915 and will die
in 1995, or you were born in 1925 and will die in 2005. In each case, we
will suppose, your life contains the same amount of pleasure and pain,
distributed evenly through time. It is quite clear that you would prefer
the second option—you want your good experiences in the future. Note
that the periods before 1915 and after 2005 involve “experiential blanks”
in any case. However, on the first option there is an “extra” blank between
1995 and 2005, and on the second option this extra blank is placed
between 1915 and 1925. If one focuses simply on this experiential blank
of ten years and asks whether it would be better to have the blank in the
past or the future, it seems that one shouldn’t care. That is, as argued
above, it is rational for a person to have temporally symmetric attitudes
toward bads not experienced by him. Thus, our preference for the second
option—living more in the future—cannot be explained directly by an
alleged asymmetry in our attitudes toward experiential blanks. Rather, it
is crucial that the placement of the “extra” experiential blank of ten years
determines the temporal distribution of experienced goods, since we do
have temporally asymmetric attitudes toward experienced goods.
Nagel is correct to assimilate death to a bad such as betrayal by a
friend behind one’s back—both bads do not involve unpleasant
experiences. But the two sorts of bads are interestingly different. If death
occurs later than it actually does, we will have a stream of good experi-
ences in the future. The alternative to death is good experiences, whereas
(in the typical case, at least) the alternative to a future betrayal behind
one’s back is not good experiences. Thus prenatal and posthumous non-
existence deprives us of things to which we have temporally asymmet-
ric attitudes, whereas past and future betrayals do not. Death’s badness
is similar to the badness of betrayal behind one’s back, but different in
a way which explains why death is rationally regarded as worse than
prenatal nonexistence.15
NOTES
1. This does not imply that it is rational to preoccupy oneself with one’s own
death or to focus one’s attention upon it constantly, etc.
2. Something is “experienced as bad by a person” roughly speaking inso-
far as that thing causes unpleasant experiential episodes in the person (and
perhaps, the person believes that the thing is causing such experiences).
3. Thomas Nagel discusses such bads in: “Death,” reprinted in Thomas
Nagel, Mortal Questions (Cambridge: Cambridge University Press, 1979),
pp. 1–10. Also, Robert Nozick discusses similar examples in: “On the Randian
Argument,” in Jeffrey Paul (ed.), Reading Nozick (Totowa, N.J.: Rowman &
Littlefield, 1981), pp. 218–222.
4. Nagel, Ibid.
5. Ibid., pp. 7–8.
6. Even if one—controversially—held that generation from such and such
gametes is an essential property of an individual, this would not commit one to
the further essentialist claim in the text.
7. Nagel himself is unsatisfied with this response. (Nagel, Ibid. fn. 3, pp. 8–9).
He points out that “it is too sophisticated to explain the simple difference between
our attitudes toward prenatal and posthumous nonexistence.” (Ibid.) To explain his
doubts, he presents an example (attributed to Robert Nozick) in which it is granted
that it is logically possible that an individual be born years before he is actually
born (by prematurely “hatching” the spore from which one develops), and yet it
seems that even here the intuitive asymmetry is justified. Thus, the logical impos-
sibility of being born earlier cannot explain the asymmetry in our attitudes.
8. Derek Parfit, Reasons and Persons (Oxford: Oxford University Press,
1984), pp. 165–185, esp. p. 175.
9. Ibid., pp. 165–166.
10. Nagel seems to have been aware of some version of Parfit’s claim. Given
his worries about the view that it is logically impossible that one should have
been born much earlier than one actually was, Nagel admits that “Lucretius’
argument still awaits an answer.” He continues (Ibid., fn. 3, p. 9): “I suspect that
it might require a general treatment of the difference between past and future
in our attitudes toward our own lives. Our attitudes toward past and future
pain are very different, for example. Derek Parfit’s unpublished writings on
this topic have revealed its difficulty to me.”
11. So a symmetric attitude towards past and future betrayals involves preference
for one betrayal over several comparable ones regardless of when they occur and
indifference between two comparable betrayals regardless of when they occur.
12. Ibid., p. 181.
13. Parfit (Ibid., p. 182), says: “My own examples reveal a surprising asym-
metry in our concern about our own and other people’s pasts. I would not be
distressed at all if I was reminded that I myself once had to endure several
months of suffering. But I would be greatly distressed if I learnt that, before she
died, my mother had to endure such an ordeal.”
This asymmetry is not the same as the asymmetry between my attitudes
toward my own past and my own future, yet the two asymmetries are connected
as follows. The first asymmetry consists in my indifference to my own past suf-
fering paired with my concern for another’s past suffering. Given my concern for
my own future suffering, it follows that I have asymmetric attitudes toward my
own past suffering and my own future suffering. Given my concern for another’s
future suffering, it follows that I have symmetric attitudes toward another’s past
suffering and another’s future suffering. Thus the contrast between temporally
asymmetric attitudes regarding my own suffering and temporally symmetric
attitudes regarding another’s suffering stems from the “surprising” asymmetry
Parfit notes in the above-quoted passage. But the contrast in question, which arises
from the “surprising” asymmetry, is precisely what one should expect given the
discussion in the text: the contrast matches up with the contrast between bads
which one experiences and bads which one does not.
14. Though Parfit focuses upon examples involving temporally asymmetric
attitudes towards pain, he speaks of our “bias toward the future” with respect
to experienced goods such as pleasure as well. So he would endorse the prin-
ciple about temporally asymmetric attitudes toward experienced goods, which
grounds the foregoing explanation of the asymmetry in our attitudes toward
prenatal and posthumous nonexistence. Though this explanation is consistent
with Parfit’s remarks in the passages surrounding his discussion of Epicurus
on death, that discussion itself does not indicate that he had the explanation in
mind: “Epicurus’s argument fails for a different reason: we are biased towards
the future. Because we have this bias, the bare knowledge that we once suffered
may not now disturb us. But our equanimity does not show that our past suf-
fering was not bad. The same could be true of our past nonexistence. Epicurus’s
argument therefore has force only for those people who lack the bias towards
the future, and do not regret their past non-existence. There are no such people.
So the argument has force for no one.” (Ibid., p. 175)
In any case, it is crucial to see that only the principle about temporally asym-
metric attitudes toward experienced goods such as pleasure will afford an explana-
tion of why death is bad. The principle about experienced bads which is suggested
by Parfit’s examples, it has been argued, will not generate such an explanation.
15. We would like to thank Phillip Bricker for helping us to arrive at the
foregoing explanation of why death is bad.
3
Death, Badness, and the

Impossibility of Experience
John Martin Fischer
Dying
Is an art, like everything else.
I do it exceptionally well.
I do it so it feels like hell.
I do it so it feels real.
I guess you could say I’ve a call, . . .
Sylvia Plath, “Lady Lazarus”
Few Fallacies depressed me more than the line, “I don’t mind being dead; it’s just like
being asleep. It’s the dying I can’t face.” Nothing seemed clearer to me in my nocturnal
terrors than that death bore no resemblance to sleep. I wouldn’t mind dying at all,
I thought, as long as I didn’t end up Dead at the end of it.
Julian Barnes, Metroland1
I. NAGEL AND THE CRITIQUE OF EPICURUS

It is a perennial philosophical puzzle how death can be bad for the
individual who dies. Insofar as death is construed as an experiential
blank, some have argued (following Epicurus) that one’s own death (as
opposed to one’s dying) cannot be bad for one; after all, one does not
experience death as bad or have any unpleasant experiences as a result
of it. As the saying goes, “What you don’t know can’t hurt you.”
But is the view expressed by this saying correct? Thomas Nagel has
argued that it is not.2 Nagel argues that an individual can be harmed by
something which does not result in any unpleasant experiences for that
individual.3 He employs the following example in support of his position:
It (the view that what you don’t know can’t hurt you) means that
even if a man is betrayed by his friends, ridiculed behind his back,
and despised by people who treat him politely to his face, none of it
can be counted as a misfortune for him so long as he does not suffer
as a result.4
Let us call Nagel’s example the “betrayal-behind-one’s-back” exam-
ple (or the “betrayal” example). Other philosophers have presented
37
similar examples in support of the view that what one does not know
can indeed hurt (or at least harm or be bad for) one.5
But various philosophers have replied to Nagel. In order to develop the
approach of Nagel’s critics, it is helpful to distinguish two principles:
Experience Requirement I (ER I): An individual can be harmed by

something only if he has an unpleasant experience as a result of it
(either directly or indirectly).
Experience Requirement II (ER II): An individual can be harmed by
something only if it is possible for him to have an unpleasant experi-
ence as a result of it (either directly or indirectly).
The critics have noted that Nagel’s betrayal-behind-one’s back exam-
ple only impugns ER I, and not ER II. After all, it is presumably possible
(in some appropriate sense) for the individual in Nagel’s example to
find out about the betrayal and thus have unpleasant experiences, even
if he actually does not. Further, these philosophers have emphasized
that ER II appears to imply that death—construed as an experiential
blank—is not bad for the individual who dies: death rules out even
the possibility of experience. Thus, it is alleged that Nagel’s example
falls short of establishing that death can be bad for an individual. What
is needed (according to these philosophers), and what has not been
provided, is a counterexample to ER II rather than simply to ER I; that
is, what is required is a non-question-begging example in which an
individual cannot have unpleasant experiences as a result of something
and yet we would say that that thing is bad for the individual.6
It will be useful to have some examples of this criticism of Nagel
before us. Harry Silverstein says:
. . . Nagel’s argument has force only against the strongest and least
plausible version of VCF [the “Values Connect with Feelings” view],
the version that requires that the value-recipient actually have the
appropriate feeling. It has no force against more plausible versions,
e.g., a version according to which x can intelligibly be said to have a
certain A-relative value provided merely that it be possible, or possible
under certain conditions, for A to have the appropriate feeling as a
result of x. For A’s suffering from, e.g., undetected betrayal is possible
in the sense that he may later discover the betrayal and suffer as a
result—indeed, he may then suffer, not merely from the fact that
he was betrayed but from the fact that the betrayal was undetected
until that time. Thus, Nagel’s examples are quite consistent with,
and therefore constitute no argument against, this weaker version of
VCF . . .
Hence, Nagel’s argument by counterexample is insufficient.7
Following Silverstein on this point, Stephen Rosenbaum says:
Death, Badness, and the Impossibility of Experience 39
Thomas Nagel argues that what a person does not know may well
be bad for the person. Nagel seems thereby to object to premise (A)
[A state of affairs is bad for person P only if P can experience it at
some time]. He gives plausible cases in which something can be bad
for a person even if the person is unaware of it. Unknown betrayal
by friends and destruction of one’s reputation by vile, false rumors
of which one is unaware are examples of evils which a person might
not consciously experience. Strictly, however, such cases are logically
compatible with (A) [ER II] and hence do not refute (A) [ER II], since
all (A) [ER II] requires for something to be bad for a person is that
the person can experience it (perhaps not consciously) at some time,
not that he actually experience it consciously. We can grant that what
one does not consciously experience can hurt one without granting
that what one cannot experience can hurt one.8
II. A DEFENSE OF NAGEL: NAGEL MEETS FRANKFURT

I concede that Nagel’s betrayal-behind-one’s-back example does not, as
it stands, decisively establish the falsity of ER II. But I believe it can be
modified so that it can indeed establish the falsity of ER II. The problem
with the example is that, although you do not actually have bad experi-
ences as a result of the betrayal, you nevertheless can (in some suitable
sense) have such experiences. (It is thus open to the proponent of ER
II to admit that you are harmed, but say that this is in virtue of your
ability to have unpleasant experiences.) Now it seems to me that there
are various ways in which Nagel’s example could be modified so that
you cannot (in some natural sense) have any bad experiences as a result
of the betrayal. One such way employs the idea of a “counterfactual
intervener”—an agent who does not play any role in the actual course
of events, but stands ready to intervene under certain (counterfactual)
circumstances.9
Imagine first that the example is as described by Nagel. You are
betrayed behind your back by people who you thought were good
friends, and you never actually find out about this or have any bad
experiences as a result of the betrayal. But now suppose that these
friends were (very) worried that you might find out about the betrayal.
In order to guard against this possibility, they arrange for White to
watch over you. His task is to prevent you ever from finding out about
the betrayal. So, for example, if one of the individuals who betrayed
you should decide to tell you about it, White can prevent him from
succeeding: White can do whatever is required to prevent the informa-
tion from getting to you. Or if you should begin to seek out one of the
friends, White could prevent you from succeeding in making contact.
I simply stipulate that White is in a position to thwart any attempt by
you or your friends to inform you of what happened.10
In this Frankfurt-style version of Nagel’s betrayal example, I further

stipulate that everything (plausibly thought to be relevant) that actually
happens among your friends and to you (and your family) is exactly
the same as in the original version of the example; we could “subtract”
the existence of White and this would make no relevant difference to what
actually happens among your friends and to you (and your family) for
the rest of your life. The only difference between the original case and
the modified case is that your friends have so arranged things that
White is poised to intervene at any point in your life where there would
be a chance that you would discover what happened; it turns out that
intervention is never actually necessary, and thus the actual sequence
of events in the modified example is in relevant respects precisely like
that of the original example. White serves as a fail-safe mechanism; his
intervention is never triggered, but his presence ensures that you will
never find out about the betrayal.
I claim that this modified version of Nagel’s example is one in which
it is plausible to say that something happens that is bad for you—the
betrayal—and yet it is not possible for you to have any bad experiences
as a result of it. If this is correct, then we do indeed have a counter-
example to ER II.
I suppose that someone could concede that in the modified example
you cannot have any bad experience as a result of the betrayal but insist
that precisely this makes it the case that the betrayal is not bad for you.
Whereas this position is certainly open to one, it results in the extremely
implausible differentiation of the original betrayal case and the modi-
fied betrayal case: one must say that you are harmed by the betrayal
in the original case but not in the modified case. But although you are
harmed (according to this view) only in the original case, everything
that happens among your friends and to you (and your family) is in all
relevant respects the same in both cases.11 It then seems very implau-
sible to say that you are harmed in the original case but not in the modi-
fied case.
I thus claim that the modified example is precisely the sort of exam-
ple demanded by philosophers such as Silverstein and Rosenbaum
in order to establish that ER II is to be rejected: an example in which
something is bad for an individual and yet it is not even possible
for him to have unpleasant experiences as a result of it. The existence
of such an example should not be surprising, given certain insights
of Nagel. Consider again the original betrayal-behind-one’s-back
example. Reflecting on this example, it is natural to think that it is
not merely the actual lack of unpleasant experiences that is relevant;
the example suggests that even the possibility of unpleasant experi-
ences is not what makes the betrayal bad. And Nagel provides some
theoretical resources that could be developed to bring out this point
more explicitly.
After laying out the original version of the example, Nagel says:
Someone who holds that all goods and evils must be temporally
assignable states of the person may of course try to bring difficult
cases into line by pointing to the pleasure or pain that more compli-
cated goods and evils cause. Loss, betrayal, deception, and ridicule
are on this view bad because people suffer when they learn of them.
But it should be asked how our ideas of human value would have
to be constituted to accommodate these cases directly instead. One
advantage of such an account might be that it would enable us to
explain why the discovery of these misfortunes causes suffering—in a
way that makes it reasonable. For the natural view is that the discov-
ery of betrayal makes us unhappy because it is bad to be betrayed—
not that betrayal is bad because its discovery makes us unhappy.12
Nagel is here discussing whether future actual discovery of the betrayal,
with its attendant unpleasant experiences, is what would make the
betrayal bad. But his Euthyphro-type point could be adapted to the
issue of whether the possibility of discovery and attendant unpleas-
ant experiences is what makes the betrayal bad. For it is natural to say
that it is not the possibility of bad experiences that makes betrayal bad,
but rather the badness of betrayal that explains why one would have
unpleasant experiences given the possible circumstance of discovery of
the betrayal. If this sort of analysis of the order of explanation is correct,
then it is not surprising that there should be examples in which some-
thing is bad for an individual and yet there is not even the possibility
that the individual have bad experiences as a result.
III. CLARIFICATION: TWO KINDS OF POSSIBILITY

Notoriously, the notion of “possibility” is vague. And of course this
notion plays a crucial role in ER II and my counterexample to it. It will
be useful to make a distinction between a broader and a narrower notion
of possibility. The broad notion is “metaphysical possibility in the broad
sense.” This is the sort of possibility that is (very roughly) compatibility
with the laws of logic, the analytic or conceptual truths, and the proposi-
tions entailed by basic metaphysical truths (including truths about the
essences of things). According to this broad notion of possibility, it would
be possible for you to have bad experiences as a result of the betrayal
even in the modified betrayal example. This is because it is compatible
with the laws of logic (and the basic metaphysical truths) that White not
succeed in performing his task (for whatever reason).
But broad possibility is very broad indeed. Suppose that you are
chained to your chair by very heavy chains which you cannot break, and
also imagine that there is no way that you can get anyone else to help
you remove the chains (within, say, an hour). Given certain plausible
ancillary assumptions, it follows that you cannot get out of the chair
within an hour, in the sense of “can” typically thought to be relevant
(in some way or another) to moral responsibility. Note, however, that
your breaking the chains is compatible with the laws of logic (and the
relevant metaphysical truths), and thus your getting out of the chair is
possible in the broad sense.
Let us say that the sort of possibility that corresponds to the free-
dom typically associated (in some way or another) with moral respon-
sibility is “narrow possibility.” This sort of possibility implies that the
relevant agent have a general ability to do the thing in question and
also the opportunity to exercise the ability. Of course, it would be too
daunting a task for me to attempt to provide an analysis of the relevant
sort of freedom. Here I simply associate the narrower sort of possibil-
ity with this freedom (however it is analyzed, if it is analyzed at all).
Possibility in the narrow sense implies that a certain course of action is
genuinely accessible to or open to an agent. It is not genuinely open to
you in the modified betrayal example to discover that you have been
betrayed.13
Note that, if the broad notion of possibility is employed in ER II, then
the modified betrayal example is no counterexample. But I would con-
tend that it is the narrow notion of possibility that is relevant to ER II.
Surely, if one is concerned to connect badness (or harm) with the possi-
bility of experience, it is not plausible to employ the broad notion of pos-
sibility. Consider, as an example, an individual who has been reduced
to a persistent vegetative state as a result of a stroke. Physicians reliably
diagnose this person as terminally comatose. Presumably, in the sense
of possibility relevant to the issue of whether this individual can be
harmed by (say) a betrayal, it is impossible for the individual to have
unpleasant experiences. But if this is correct, then the relevant notion
of possibility cannot be the broad notion, for it is possible in the broad
sense for the individual to have unpleasant experiences (as a result,
say, of a miraculous recovery of the capacity for consciousness). I con-
tend then that it is the narrow notion of possibility that is appropriately
employed in ER II, and that the modified betrayal example constitutes
a counterexample to ER II thus interpreted.14
IV. IMPLICATION: DEATH’S BADNESS

One implication of this result is to vindicate the Nagelian critique of
Epicurus’ argument that death is not bad for the individual who dies.
In the paper cited above (and others), Rosenbaum has sought to give
a reconstruction of Epicurus’ argument which withstands Nagel’s
assault.15 His main point is that Epicurus relies upon ER II rather than
ER I, and thus Nagel’s original case does not undermine Epicurus’
argument. But if I am correct and the modified betrayal example shows
the inadequacy of ER II, it follows that Rosenbaum’s attempt to provide

a defense of Epicurus must fail.
It is interesting to note how Rosenbaum attempts to argue for ER II:
Suppose that a person P cannot hear and never will hear. Then
the egregious performance of a Mozart symphony cannot causally
affect P at any time, supposing that what makes the performance
bad is merely awful sound, detectable only through normal hear-
ing, and supposing further that the performance does not initiate
uncommon causal sequences that can affect the person. It is clear
that the person cannot experience the bad performance, auditorily
or otherwise. Furthermore, it seems clear that the performance can-
not be bad for the person in any way. It cannot affect the person
in any way. The reason why it is not bad for him is that he is not
able to experience it. . . . Similarly, a person born without a sense of
smell cannot be causally affected by, and thus cannot experience,
the stench of a smoldering cheroot. The stench cannot be an olfac-
tory negativity for her. We could imagine indefinitely many more
such cases.
Since I see nothing eccentric about these cases, I believe that we
are entitled to generalize and claim that our judgments about these
cases are explained by the principle that if a person cannot experi-
ence a state of affairs at some time, then the state of affairs is not bad
for the person.16
Certainly, there is nothing so “eccentric” as a counterfactual intervener
(such as White) in the cases envisaged by Rosenbaum. Nevertheless,
I contend that his examples are indeed eccentric in the sense of being
an inappropriate sample on the basis of which to generalize. To
explain, the question at issue in this dialectic is whether there are some
things that are bad for someone and yet cannot result in unpleasant
experiences for the individual. Now Rosenbaum invokes a number of
examples in which the thing in question typically causes unpleasant
experiences: an “egregious performance of a Mozart symphony,” and
“the stench of a smoldering cheroot.” Such things, by their very nature,
would be bad for an individual precisely by causally affecting him and
producing unpleasant experiences. This is because in these cases the
badness could only be (or be the result of) sensory unpleasantness. It is
surely uncontroversial that in this class of cases an individual cannot
be harmed by the thing in question, if he cannot be causally affected
by it.
But the critic of Epicurus will point out that this class is only a proper
subclass of all the relevant cases. There are other cases in which it is
alleged that there is something which is bad for an individual even
though it does not (and cannot) causally affect the individual. In these
cases, the alleged badness would not be a sensory badness (i.e., an
unpleasant experience coming from some offensive sensory stimulus).

Now of course it is contentious whether this allegation is true; but all
I wish to point out is that it surely is dialectically unfair to point to
the “sensory subclass” of cases and then generalize on the basis of this
subclass to the conclusion that badness requires the possibility of expe-
rience. And of course, the betrayal examples (both the original version
and the modified version) are precisely examples which are not in the
sensory subclass. Here the badness is supposed to be something like
the undetected ruining of one’s reputation; if the example works as it
is supposed to, the mechanism of harm would not be sensory. It is an
example of a very different sort.
Let me highlight the—admittedly quite delicate—dialectical situa-
tion. The class of putative bad things can be partitioned into various
proper subclasses (in various different ways). On one way of making
the partition, one proper subclass contains putative bad things which
involve—by their very nature—unpleasant sensory stimuli; in virtue of
this fact, it seems to be true of such putative bads that they could not be
bad for an individual who is unable to experience the sensory stimuli. A
quite different proper subclass of putative bads does not contain mem-
bers which involve unpleasant sensory stimuli; if these are indeed bad
for individuals, they could be bad for individuals who are incapable
of experiencing them as bad. Given the existence of this second sub-
class, it is clearly inappropriate to generalize from the first subclass to
a conclusion such as ER II. Of course, this does not imply that the mere
existence of the second subclass (of putative bads) decisively establishes
that there are indeed things that are bad for individuals who cannot
experience them as bad (or have unpleasant experiences as a result of
them); rather, it simply shows that one cannot precipitously generalize
from the existence of the first subclass to ER II.
Finally, a defender of Epicurus might wish to take a slightly differ-
ent tack. Rather than focusing on the experience requirement, he may
put forward some sort of existence requirement. According to this sort
of requirement, an individual cannot be harmed by the occurrence of
something at a time T if the individual does not exist at T. Clearly, my
argument above does not in itself refute the existence requirement. But
it is unclear why one would wish to insist on an existence requirement
apart from an experience requirement. This is because the most natural rea-
son (or at least a very salient reason) to require existence is that exis-
tence is necessary for experience. That is, if one gives up the experience
requirement for badness (or harm), why exactly would one cling to the
existence requirement? Perhaps there is reason to maintain an existence
requirement without an experience requirement, and if so, my argu-
ment does not address this position. But my argument clearly has force
against the proponent of an existence requirement who bases it upon an
experience requirement. And it seems to shift the argumentative burden
to a proponent of the existence requirement to explain its basis, if it is

not the experience requirement.17
V. CONCLUSION
Nagel’s “betrayal” example is a case in which it seems that the betrayal
is directly bad for someone—bad for someone quite apart from the indi-
vidual’s experiencing anything unpleasant as a result of the betrayal.
It seems intuitively that actual or even possible experience is not what
makes the betrayal bad. Further, Nagel’s “Euthyphro-type” point (about
the order of explanation of the badness of betrayal) provides some the-
oretical backing for the intuitive notion that experiential considerations
are not the basis of the view that the betrayal is bad. Nagel’s example
and theoretical insight then show that in principle, there should be
examples in which something is bad for an agent even though he can-
not have bad experiences as a result. The “Frankfurt-type” version of
Nagel’s original betrayal case is precisely this sort of example. And if
so, then ER II, together with the most powerful reconstruction of the
Epicurean argument against death’s badness, must be rejected.
Note, also, that if ER II is rejected, then so also must an intriguing
recent argument by Walter Glannon.18 Lucretius argued that posthu-
mous and prenatal nonexistence should be treated symmetrically, and
that, since we do not think of prenatal nonexistence as a bad thing, we
should also not think of death as a bad thing. Various philosophers
have responded, insisting on the commonsense asymmetry: prenatal
nonexistence is not a bad thing, but death is.19 Glannon has employed
ER II to argue for precisely the opposite asymmetry: since we can (while
we are alive) have unpleasant experiences as a result of things that hap-
pen in the prenatal environment, but we cannot have bad experiences
after we are dead, Glannon argues that prenatal nonexistence can be
bad for an individual, but death is not. However, it is clear that if ER II
is rejected, then there is no basis for Glannon’s asymmetry.
I wish to end with an even more fanciful example and some specula-
tive reflections on its implications. Imagine that your spouse and your
best friend are on a space colony orbiting Mars, which is now on the
opposite side of the Sun from Earth. Hence, it will take a few minutes
for light waves to travel from Mars to the village on the Alaskan coast
where you reside. They betray you. It turns out, however, that a gargan-
tuan earthquake-induced tidal wave is going to kill you in fewer than
the number of minutes it takes for light to travel from Mars to Earth.20
Here it is impossible (in a very strong sense) for you to experience
something bad as a result of the betrayal. And yet it seems that you
have been harmed. Certainly, if you are harmed by the betrayal in the
Frankfurt-type betrayal example, then you are harmed by the betrayal
in the Mars example. But it is impossible (short of a violation of the
laws of physics) that you experience anything bad as a result of the

betrayal in the Mars example.
It seems to me that it is correct to say that there is an interesting anal-
ogy between space and time with respect to death’s badness. Just as one
can be harmed by a spatially distant event, one can also be harmed by
a temporally distant event.21 The Mars example drives this point home:
one can be harmed by something that is spatially remote and from
which one is causally isolated (barring violations of the laws of phys-
ics). But if this is so, then it is plausible to say that one can be harmed
by temporally distant events (from which one is causally isolated). If
you are harmed by the betrayal in the Mars case, why not also say you
can be harmed by your death, even though the death occurs after you
cease to exist?
These reflections, then, suggest simple answers to some of the most
perplexing puzzles pertaining to the badness of death. Death is bad in
virtue of being a deprivation of the good things in life. The subject of the
misfortune of death is you (the individual who dies). The time of the
misfortune is the time during which you are dead.22 Just as some misfor-
tunes occur at a spatial distance and are causally isolated from you, other
misfortunes occur at a temporal distance (and are causally isolated from
you).23
NOTES
1. I am indebted to G. Dworkin for bringing this passage to my attention.
In “Death,” T. Nagel expresses a similar view: “It is sometimes suggested that
what we really mind is the process of dying. But I should not really object to
dying if it were not followed by death” [T. Nagel, Mortal Questions (Cambridge,
UK: Cambridge University Press, 1979), pp. 1–10; reprinted in J.M. Fischer
(ed.), The Metaphysics of Death (Stanford: Stanford University Press, 1993),
pp. 61–69].
2. T. Nagel, Mortal Questions (1979), pp. 1–10.
3. If the individual has unpleasant experiences simply as a result of
discovering or recognizing that a certain event has taken place, this is a rel-
atively “direct” way in which that event results in unpleasant experiences.
If, however, the event has consequences (other than mere recognition by
the individual) that then cause unpleasant experiences in the individual,
this would be a relatively “indirect” way in which the event results in
unpleasant experiences. For a similar distinction, see J. McMahan, “Death
and the Value of Life,” Ethics 99 (1988), pp. 32–61, esp. pp. 32–34; this essay
is reprinted in Fischer (ed.), pp. 233–266. I mean to include both direct and
indirect ways of resulting in unpleasant experiences in my discussion in
this essay.
4. Nagel, p. 4.
5. See J. Feinberg, Harm to Others (Oxford: Oxford University Press, 1984),
pp. 86–87; and R. Nozick, “On the Randian Argument,” in J. Paul (ed.), Reading
Nozick (Totowa, N.J.: Rowman & Littlefield, 1981), p. 221.
6. For the purposes of this discussion, I will not distinguish between some-
thing’s being bad for an individual and that thing’s harming the individual.
7. H.S. Silverstein, “The Evil of Death,” Journal of Philosophy 77 (1980),
pp. 414–415; reprinted in Fischer (ed.), pp. 95–110.
8. S.E. Rosenbaum, “How to Be Dead and Not Care: A Defense of Epicurus,”
American Philosophical Quarterly 23 (1986), p. 221; the essay reprinted in Fischer
(ed.), 119–134. Another proponent of ER II is W. Glannon, who appears to
base much of his view of death’s badness on something like it. In his article,
“Temporal Asymmetry, Life, and Death,” American Philosophical Quarterly 31
(1994), pp. 235–244, Glannon is not always careful to distinguish ER I from ER
II. For example, the following suggests that Glannon is a proponent of ER I:
. . . it is irrational to care now about the goods of which we allegedly will be
deprived by death. For it is rational to be concerned about the pleasure and
pain, the happiness and suffering, that we actually experience as persons. Yet
we cannot experience anything after we die (p. 238) [emphasis added].
But I believe that Glannon’s considered view is ER II. He says:
We care about future experienced goods to the extent that we can anticipate
actually experiencing them in the lived future. By contrast, in the postmor-
tem future there are no goods that we can actually experience, and so there
is no reason to be concerned now about the non-actual goods of which death
purportedly deprives us (p. 238) [emphasis added].
Also, Glannon says:
On the intuitively plausible assumption that the value of our lives is a func-
tion of what we can experience, something is intrinsically good or bad for us
only if it is possible for us actually to experience it as such (p. 238). . . . Even
if death is bad in the extrinsic sense of depriving the deceased of the goods
they would have experienced if they had continued to live, it does not fol-
low that it is rational to be concerned about death. For what makes our con-
cern about a state of affairs rational is the possibility of our experiencing it as
intrinsically good or bad, and we cannot experience anything in the state of
postmortem nonexistence (p. 241) [emphasis added].
9. The “Principle of Alternative Possibilities” states that an agent can be
morally responsible for what he has done only if he could have done other-
wise. In “Frankfurt-style” counterexamples to this principle, an agent acts “on
his own” in just the way we believe an agent typically acts when we hold him
morally responsible, and yet some counterfactual intervener is associated with
the agent in such a manner as to render it plausible that the agent cannot do
other than what he actually does. The version of Nagel’s example I develop
in the text takes its cue from Frankfurt-type counterexamples to the Principle
of Alternative Possibilities: H.G. Frankfurt, “Alternate Possibilities and Moral
Responsibility,” Journal of Philosophy 66 (1969), pp. 828–839; and “Freedom of
the Will and the Concept of a Person,” Journal of Philosophy 68 (1971), pp. 5–20.
10. The presence of White, so described, appears to rule out the possibil-
ity of the betrayal’s directly resulting in unpleasant experiences. Although this
makes the example a bit less elegant, I also stipulate that White is in a position
to prevent indirect unpleasant experiences.
11. Of course, in the modified case, your friends arrange for White’s pres-
ence, but I assume that this in itself (and absent any interventions by White)
cannot be a relevant difference.
12. Nagel, p. 5.
13. I do not here suppose that there is some interesting connection between
the issues related to moral responsibility and those related to death; rather,
I am simply attempting to identify the notion of possibility that is relevant to ER
II. Alternatively, one could simply say that it corresponds to Austin’s “all-in”
notion of possibility (or “can”): J.L. Austin, “Ifs and Cans,” in his Philosophical
Papers (London: Oxford University Press, 1961). That is, “narrow possibility”—
having a pathway genuinely accessible to one—is picked out by Austin’s “all-
in” sense of “can.”
14. I do not deny that someone could dig in his heels and simply insist
that I have not “proved” that narrow possibility is the relevant notion of
possibility. I concede this, but I think it is clear that if one bases value on
the possibility of experience, it is not plausible to adopt the broader notion
of possibility. The intuitive motivation for connecting value with the pos-
sibility of experience does not sit well with employing the broad notion of
possibility.
15. Rosenbaum, “How To Be Dead and Not Care: A Defense of Epicurus,”
and “Epicurus and Annihilation,” The Philosophical Quarterly 39 (1989), pp. 81–90.
The latter essay is reprinted in Fischer (ed.), pp. 293–304.
16. Rosenbaum, “How To Be Dead and Not Care: A Defense of Epicurus,”
p. 219.
17. Suppose one sought to defend the existence requirement by basing
it on some sort of requirement of the possibility of “being affected.” On this
approach, something could not be bad for one, if it were impossible for one to be
affected by it (quite apart from experiencing it). It seems to me that the Frankfurt-
style counterexample employed above against ER II would also work against
this sort of approach.
18. Glannon, “Temporal Asymmetry, Life, and Death.”
19. See, for example, Brueckner and Fischer, chapter 2. See also, F. Kamm,
Morality, Mortality, Vol. One (New York: Oxford University Press, 1993).
20. I am grateful to M. Otsuka for this example.
21. For these points, see Silverstein. Silverstein believed that in order to
sustain his analogy between space and time, he had to argue that the future
“exists atemporally.” But I wish to employ the analogy with space and time
without making this further argument. I do not believe that one needs to estab-
lish the additional (highly contentious) fact about the future, in order to employ
the analogy. For criticism of Silverstein here, and one alternative picture of the
ontological status of future events, see: P. Yourgrau, “The Dead,” Journal of
Philosophy 86 (1987), pp. 84–101; reprinted in Fischer (ed.), pp. 137–156. T. Nagel
also suggests the analogy between time and space in the context of death, say-
ing “For certain purposes it is possible to regard time as just another type of
distance.” (Nagel, p. 6)
22. Thus, the conclusion I draw from the analogy between space and time
in the context of death is different from the respective conclusions of Silverstein
and Nagel. Silverstein believes that one is atemporally harmed by one’s death.
I believe that Nagel holds that the time of the harm of death is indeterminate.
In contrast, I believe that the time of the harm is the time during which one is
dead. (Of course, I am thereby committed to the view that one can be harmed
during a time at which one does not exist.)
23. I thank M. Otsuka, D. Copp, H. Silverstein, A. Brueckner, and
D. Zimmerman for their help. My work on this essay has been supported by a
Fellowship for Independent Study and Research from the National Endowment
for the Humanities (let us hope that the NEH does not die!). I read a version of
this essay at the Western Washington State University Philosophy Conference,
Bellingham, Washington in March, 1996. I benefited greatly from the stimulat-
ing discussion at this conference.
4
Death and the Psychological

Conception of Personal Identity
John Martin Fischer and Daniel Speak
The way I see it, being dead is not terribly far off from being on a cruise ship. Most of
your time is spent lying on your back. The brain has shut down. The flesh begins to soften.
Nothing much new happens, and nothing is expected of you.
Mary Roach, Stiff; The Curious Lives of Human Cadavers
I. THE PROBLEM: OUR ASYMMETRIC ATTITUDES

It is frequently claimed in the literature on the metaphysics of death
that human beings have asymmetric attitudes toward death and pre-
natal nonexistence. The claim is that we tend to consider the prospects
of our future deaths as bad or unfortunate for us, whereas we do not
regret the fact that we were born when we actually were born, rather
than earlier. This is not to say that we would think of any future death
as bad; if we are asked to consider that we might be in the end stages of
a very painful, debilitating, and terminal disease, we might well look
at death in such circumstances as a welcome relief. However, when it is
assumed that life is still good, we tend to consider death (but not failure
to be born earlier) a bad thing for the individual in question.
This alleged datum—the asymmetry in our attitudes toward prena-
tal and posthumous nonexistence—poses a problem for an appealing
account of death’s badness. On the assumption that death is a period
of nonexistence that is an experiential blank, then it is tempting to say
that death is bad for the individual who dies insofar as it deprives the
individual of the goods (whatever they are thought to be) of life. We find
this “deprivation account” of death’s badness highly plausible.1 But if the
deprivation account of death’s badness is correct, then it would appear to
imply that we should also consider the fact that we were born when we
actually were, rather than earlier, a bad thing. It seems that the time of our
birth deprives us of time during which we could enjoy the goods of life in
just the way in which the time of our death deprives us of such goods.
51
One could of course address this apparent problem in various ways.

One approach would be to stick with the deprivation account and accept
the conclusion that we should indeed have symmetric attitudes toward
prenatal and posthumous nonexistence. There are various different ver-
sions of this approach. One possibility would be to contend that we
should regret both the fact that we do not die later and the fact that we
are not born earlier. Another possibility would be to regret neither fact.
An alternative way of responding to the problem is to continue to
embrace the deprivation account of death’s badness but to argue that
(despite the initial appearance) it does not entail that we should have
symmetric attitudes toward prenatal and posthumous nonexistence. A
salient version of this strategy contends that whereas a person could
live longer than he actually lives, he could not have been born earlier
than he actually was born. Thus, since deprivation requires the rele-
vant possibility, death can deprive the person of the goods of life, but
the time of his birth—the fact that he was born when he actually was
born—cannot deprive him of such goods.
Thomas Nagel presented this way of defending the deprivation
account in “Death,” although he confessed to some doubts about its
adequacy.2 Why exactly is it not possible for an individual person to
have been born considerably earlier than he actually was born? It is not
clear why this would be impossible (in some broad metaphysical sense).
If it is suggested that it is a necessary condition of an individual’s being
the individual he actually is that he come from a particular sperm and
egg, then why is it impossible that that sperm and egg have existed con-
siderably earlier than they actually existed? Of course, the particular
sperm and egg came from the individual’s parents, but why exactly is
it impossible (in some appropriately broad metaphysical sense) for the
parents to have lived considerably earlier?3
II. THE PSYCHOLOGICAL CONCEPTION OF THE PERSON

Recently Frederik Kaufman has proposed a different way of arguing that
an individual could not have been born earlier.4 In brief, Kaufman argues
that when an individual is concerned about his own death, he is concerned
about the death of a “psychologically thick” person—an individual with
a particular history and thus a particular set of memories, desires, beliefs,
values, personality traits, and so forth. But, Kaufman argues, such an indi-
vidual could not have existed earlier; an individual existing (considerably
earlier) would necessarily have been some other particular individual (con-
strued in the “thick” way). Thus the deprivation account can be defended.
It is not clear to us that it is impossible (in the relevant way—in some
broadly metaphysical sense) for the same thick individual to have been
born considerably earlier than he actually was born. But we grant that
this sort of possibility is at best very far-fetched; it would appear to
Death and the Psychological Conception of Personal Identity 53
require wholesale changes in the past: various persons existing and

events taking place considerably earlier than they actually do. About
this Kaufman says:
So even if it is not logically impossible for me, thickly conceived, to
have existed slightly earlier, it is so extremely unlikely that an earlier
existing person would turn out to be qualitatively identical (or even
similar) to me as I currently am as to make it virtually impossible
that I could have existed earlier.5
We will grant, then, for the sake of the argument, that human persons,
thickly conceived, could not (in the relevant sense) have been born
much earlier than they actually were.6
It will be helpful to lay out Kaufman’s reasoning in some detail.
Kaufman is concerned to distinguish between thick and thin concep-
tions of personhood. The thin conception involves nothing more than
“metaphysical essences.” Kaufman says:
A person, on this view, is simply a particular essence, and that person
exists in all possible worlds which contain that essence. The details
of one’s actual life are wholly contingent features of an individual.
On this understanding of ‘person,’ since the features of one’s actual
life are not in any way constitutive of the person one is, it is possible
for one to be shorn of all the attributes of one’s actual life and remain
the same person throughout the changes.7
Kaufman concedes that one’s metaphysical essence could have been born
earlier—could have been associated with a different thick self. But he con-
siders this irrelevant to the issues about death and earlier birth. He says:
[I]nsofar as concern about death is driven by concerns that one’s con-
scious personal existence will be extinguished forever, the fact that
one’s metaphysical essence might occur in different times and places
seems beside the point. This is why certain possible occurrences that
leave my metaphysical essence intact but which nevertheless extin-
guish my subjective sense of myself as myself are things which, like
death, I could not survive, such as brain zaps, philosophical amne-
sia, permanent coma, some versions of reincarnation, or ‘merging
with the infinite.’8
In defending his view, Kaufman presents the following intriguing
example:
My metaphysical essence—whatever it is—might have led a very
different life. I could have lived as an Eskimo, for example, had my
parents given ‘me’ to an Eskimo tribe upon my birth. I would speak
Aleut, live in igloos and hunt seals. That person—metaphysically
me—would be otherwise unrecognizable to the person I currently
am, the one who grew up in middle-class America. The conscious
personal life of the Eskimo ‘me’ would consist of completely differ-

ent memories, projects, beliefs and commitments. Whatever point of
view the Eskimo ‘me’ would have on his life and circumstance, it will
be vastly different from my current subjective awareness. . . . Were the
‘thin’ metaphysical me to be raised by an Eskimo tribe, the conscious
personal entity that I currently am would regard him as a complete
stranger. I wish him well, but I am no more concerned about his
death than I am about the death of any other stranger.9
Thus, on Kaufman’s view, it is important to distinguish between the
thin person (the metaphysical essence) and the thick person (the per-
son with a particular set of memories, histories, projects under way,
beliefs, and values). Kaufman holds that what we worry about, when
we consider our own deaths, is the elimination of the thick person.
But this sort of person cannot have been born earlier than he or she
actually was born, and thus the time of one’s birth cannot deprive
the thick person of goods he or she otherwise would have had. Thus,
the deprivation account of death’s badness is perfectly compatible
with our asymmetric attitudes toward prenatal and posthumous
nonexistence.
III. CRITIQUE OF KAUFMAN

We are willing to concede that the actual time of one’s birth cannot
deprive the thick person of goods he or she otherwise would have had.
But why exactly should this be the only relevant issue? That is, why
should we agree with Kaufman’s contention that it is irrelevant that the
thin person—the metaphysical essence—could indeed have been born
earlier (and thus could have been associated with a different thick per-
son)? We will contend that persons often make judgments that presup-
pose that it is coherent to care—and perhaps care deeply—about which
thick person one’s metaphysical essence is, as it were, attached to. (For
the purposes of this discussion, we will adopt Kaufman’s terminology.)
Imagine that you read in the newspaper that a certain hospital has
had various problems with “baby-switching” cases: babies have been
mixed up in the maternity ward, and couples have been given babies
who are not biologically theirs to take home. Assuming that you had
a generally pleasant and favorable set of childhood experiences, you
might say to yourself, “I’m relieved that sort of thing did not happen
to me! I’m glad that my biological parents brought me home from the
hospital and raised me.” This seems to be a perfectly natural and not
uncommon kind of thought, and it seems to presuppose that one can
coherently make judgments about which thick person one’s thin self
is attached to. If one’s family life was particularly pleasant, one might
explicitly contrast, in one’s mind, one’s early childhood experiences
with (as one would naturally assume) the experiences one would have
had, had one gone home from the hospital with another couple.
Suppose, however, that one’s early childhood experiences were
particularly horrid, involving significant poverty and physical and
psychological abuse. Under such circumstances, it would not at all be
unreasonable, upon reading the newspaper article, to wish that one had
in fact gone home from the hospital with another family. One might
wistfully think about having gone home with a happy, loving, and
financially secure family; it would be natural to say to oneself, “My life
would have been very different. . . . ” Again, these sorts of judgments
appear to presuppose precisely the coherence of stepping outside one’s
thick self in a certain way, contrary to Kaufman’s supposition.
Return to Kaufman’s Eskimo example. Recall that Kaufman con-
cludes, “Were the ‘thin’ metaphysical me to be raised by an Eskimo
tribe, the conscious personal entity that I currently am would regard
him as a complete stranger. I wish him well, but I am no more concerned
about his death than I am about the death of any other stranger.”10 But
let us now fill in the details of the story a bit differently. Imagine that
you were adopted when you were very young, so that you have never
known who your biological parents are. It happens that you have been
raised in a middle-class community in Anchorage, Alaska. One day
—when you are (say) forty years old—you get a telephone call from an
elderly couple who explain that they are your biological parents. They
would like to see you, and you arrange to meet them at a restaurant in
Anchorage. When they arrive at the restaurant, you are surprised to
learn that they are from an Eskimo tribe who live some distance from
the city. They apologize profusely for having had to give you up for
adoption at birth, but they already had eight children, and they just
could not manage another. They are a lovely, warm, generous couple
who have learned to speak English in their later years, in anticipation
of this meeting with you.
The meeting is very emotional, and your feelings and thoughts are cha-
otic for some time afterward. But as you think about your biological par-
ents and the story they told, you develop a strong feeling of sadness that
you were not in fact raised by them in the Eskimo tribal community. You
do love your adoptive parents, and you are grateful for all the love they
have given you. But you can’t help wondering what your life would have
been like had you been raised by your biological parents among your bio-
logical brothers and sisters in the Eskimo community. You wonder about
it constantly, and although you know there is nothing now that can be
done to change the fact that you were raised in a middle-class community
in Anchorage, you nevertheless wish that it had been different.
Of course, the story need not be told in this manner; upon meeting your
biological parents, you might have little or no regrets about not having
been raised by them. But the point is that the story can be told in this way.
It seems perfectly natural for human beings to prefer to have been raised
by their biological parents, even when they have had relatively favorable
circumstances. It would seem even more reasonable to have such a pref-
erence, if the adoptive circumstances are unpleasant and difficult.
When Kaufman concludes that the “conscious personal entity that
I currently am” would regard the individual raised by the Eskimos
as a “complete stranger” and that he would be no more concerned
about this individual’s death than about that of any other stranger,
this is a complex and puzzling claim. But it is interesting to ask why
Kaufman focuses on whether the individual as raised in the middle-
class community would be concerned about the death of the already-
mature individual raised in the Eskimo community. This does not
seem to be the relevant question. Rather, the question would appear
to be whether the individual raised in the middle-class community
could coherently form preferences about having been raised in the
Eskimo community. If it is possible for the middle-class individual
to prefer having been raised in the Eskimo community, then his not
having been so raised could deprive him of significant goods. That
is all that is required; the middle-class individual’s alleged attitudes
toward the death of the already-mature Eskimo individual seem
beside the point.
If the judgments and preferences in the cases discussed above are
coherent, then it is plausible that we can in general form judgments and
preferences about which thick persons our metaphysical essences—or
thin selves—are associated with. Put in simpler language, it seems
plausible that we can make judgments and form preferences about
which lives to lead, where the possibilities include lives with very dif-
ferent beginnings from those of our actual lives. It is not an uncom-
mon thought experiment to wonder what it would be like actually to
be someone else; one might wonder whether one would like to switch
places with another person. Of course, some versions of the thought
experiments are perfectly compatible with Kaufman’s view. For exam-
ple, it is compatible with Kaufman’s view that a thick individual can
imagine himself in some other person’s job or life circumstances now.
But what does appear to be incompatible with Kaufman’s view is the
possibility of imagining that one has led another individual’s life from
the beginning. And yet it does not seem incoherent to form the prefer-
ence for having led another individual’s life from the beginning or to be
relieved that one has not.
Regrettably, we do not have a decisive argument that the judgments
and preferences discussed above are possible. It is in fact hard to see
how to argue for their possibility. We do, however, wish to point to a
possible source of confusion in the considerations Kaufman adduces
on behalf of the contention that such judgments and preferences are
without any basis and thus irrelevant. Recall Kaufman’s claim:
This is why certain possible occurrences that leave my metaphysical

essence intact but which nevertheless extinguish my subjective sense
of myself as myself are things which, like death, I could not survive,
such as brain zaps, philosophical amnesia, permanent coma, some
versions of reincarnation, or ‘merging with the infinite.’ And, like
death, I regard these possibilities as bad insofar as they deprive me
of the goods of life. Knowing that my essence might continue on
without me, as it were, is no comfort.11
Suppose a person is subject to a “brain zap” that significantly changes

one’s “quasi memories,” preferences, values, and beliefs. As a result of
this direct electronic stimulation of the brain, the individual has a very
different personality from the one he had prior to the stimulation. We
agree with Kaufman that this would be tantamount to death for the orig-
inal individual. Once we have begun our lives and formed our person-
alities—our “thick selves,” as it were—we have a strong interest in not
having these personalities (the total configuration of memories, beliefs,
preferences, values, and so forth) radically altered in certain ways. When
such alterations take place, we consider this tantamount to death.
But there is a difference between an alteration of this sort and a
hypothetical case in which one’s life circumstances would have been
different from the beginning (thus allowing for the development of a
different personality or thick self). An alteration of the sort in question
involves a radical discontinuity (to which one does not consent volun-
tarily). That is, one’s metaphysical essence is associated with one thick
self, and then this association is severed and another thick self becomes
attached to the same metaphysical essence; we have strong negative
reactions to these sorts of cases. We do not want our lives to be discon-
tinuous in this sort of way, and thus we have a strong negative reaction
when we contemplate such possibilities.
But Kaufman appears to infer from this sort of negative reaction to
the “alteration” cases a conclusion about hypothetical cases in which
one’s circumstances are different from the beginning (thus allowing for
the gradual and continuous development of a very different kind of
personality or thick self). This kind of inference is problematic, because
one’s aversion to alteration and thus discontinuity (of certain sorts) is
irrelevant to contexts that are different from the actual circumstances
from the beginning. That an individual of course would object to being
subject to an involuntary brain zap that would significantly alter his
personality does not show that he could not reasonably prefer that he
had been raised in very different circumstances from the actual ones.
Apart from the invocation of intuitions about cases of alteration
and thus discontinuity, we do not see that there is any sort of argu-
ment in Kaufman’s article that would establish that the judgments and
preferences about leading entirely different lives discussed above are
necessarily incoherent, mysterious, or without any basis. Although this

kind of argument is not explicit in Kaufman’s work, we suppose some-
one might argue as follows. The thick individual has a point of view or
perspective—a set of experiences, memories, desires, and values—by
reference to which he can coherently form preferences about leading
various kinds of lives. But a thin person by definition does not have
such a point of view and thus cannot coherently be thought to form
preferences about different lives. After all, where would such prefer-
ences come from? What would they be based on?
But this argument rests on a mistake. It supposes that the prefer-
ence for leading one total life (a life from the beginning) as opposed to
another must be generated from the perspective of the thin self or meta-
physical essence. This is admittedly impossible, but it is also unneces-
sary. On the picture we are suggesting, it is possible for us, construed
as thick persons (with the appropriately rich perspective of a thick per-
son) to make judgments and form preferences about various scenarios
in which our metaphysical essences are associated with different thick
selves. So the preferences about such scenarios are generated from the
perspective of our thick selves, even though they are about the possi-
bility of our thin selves being associated with other thick selves. There
is nothing incoherent in this picture, as far as we can see.
We have pointed to a number of contexts in which human beings
naturally make judgments and form preferences about leading very
different lives (from the beginning). We have not sought to dispute
Kaufman’s claim that thick persons could not have been born much
earlier than they were actually born. We have instead raised ques-
tions about Kaufman’s claim that thick personhood is the only notion
of personhood relevant to concerns about the dates of our death and
birth. If it is coherent to prefer that one had been raised in very differ-
ent circumstances from the beginning, then there is no bar to saying
that it would be coherent to prefer that one had been born consider-
ably earlier. In Kaufman’s terms, one would be exhibiting a preference
that one’s metaphysical essence be associated with a different thick
person.
It is of course possible to insist that the sorts of judgments and
preferences we have claimed are coherent are at a deeper level sim-
ply incoherent or without any basis. But it is important to see that one
would thereby be calling into question some fairly common kinds of
practices (of making certain judgments and forming certain prefer-
ences).12 Further, we do not see a strong argument in Kaufman’s article
(or elsewhere) to convince us that we should abandon these practices.
When he seeks to argue for his conclusion, Kaufman appears to go from
considerations pertaining to radical discontinuity and disruption to
those pertaining to alternative circumstances that are different from the
beginning (but involve no such alteration or disruption).
IV. BELSHAW’S “CONSERVATION CLAIM”

In his recent essay, “Death, Pain, and Time,” Christopher Belshaw pres-
ents an approach that is very similar to that of Kaufman. As we under-
stand Belshaw’s argument, he wishes to start with the assumption that
almost everyone does indeed have an asymmetric attitude toward
prenatal and posthumous nonexistence. That is, the vast majority of
people, if not everyone, will consider that the time of their death can
deprive them of goods but will not think of the time of their birth as
similarly depriving them of goods. He then wishes to offer an explana-
tion for this asymmetry. Belshaw says:
We want, most of us, for the past to be as it is, and so are neither indif-
ferent to its shape, nor interested in amending it. Older people can,
on occasion, express a wish to be younger. Historians sometimes, and
unthinkingly, say they’d like to be older than they are. The rest of us
generally do not. We recognise that our being born at a certain time
is, in large part, responsible for who we are today. Someone born at
a different time just wouldn’t, in an everyday sense, be me. And so
we want neither to reduce prenatal nonexistence, making ourselves
older, nor, as a means of regaining youth, to increase it. Nor are we
simply indifferent to past nonexistence, caring neither one way nor
the other when we were born. Rather, our concern is with conserva-
tion, with keeping the facts of prenatal nonexistence just as they are.
For many of us recognise, and many more can easily be brought to
recognise, that a concern for the present to be as it is, and for us to
be who we are, implies a concern that the past be as it was, and thus
that we be born when in fact we were. . . . Indeed, the point can be put
more dramatically: to want to be born at a different time is, in effect,
to want not to exist, and for someone else to exist in your place. It’s
not surprising that this is something only a very few of us want.13
Belshaw’s view is presumably similar to Kaufman’s in that he is sug-

gesting that to want to be born at a different time is to want to be a
different thick person (in Kaufman’s terminology) and that very few
persons would want this. According to Belshaw, we do not want to
have been born earlier because we recognize (or could easily be brought
to recognize) that if we had been born earlier, our personalities would
have been significantly different. The key assumption is that we would
not want our personalities to be significantly different (from the begin-
ning, as it were).
But this assumption has problems that are similar to those that
plagued Kaufman. Belshaw is seeking to explain what he takes to be a
very prevalent and almost universal attitude toward birth. His crucial
supposition is that almost everyone would not prefer to have had a
different personality from the beginning. But this does seem to be a
very conservative principle, and it is not at all clear that it is true. Surely
many individuals would not want to have had a different personal-
ity from the beginning. These individuals—perhaps they are a majority
(even a large majority)—have led basically good lives under generally
favorable circumstances. Even many people who have struggled con-
siderably will no doubt prefer not to have had a different personality
from the beginning (where they do not have control over the nature or
features of this other personality).
But surely there are many deeply unhappy persons—persons who
grew up under conditions of horrible poverty or terrible physical or
emotional abuse. Many of these individuals suffer from the emotional
scars of their troubled early childhoods, and quite a few of them lead
very unhappy lives. Why wouldn’t these people be willing to take the
risks involved in having a different personality? And of course it is not
only individuals from disadvantaged or troubled backgrounds who
suffer from deep, persistent, and unpleasant emotional and mental
problems; chronic depression and schizophrenia are not reserved for
those who have grown up in poverty or suffered abuse. Why is it so
obvious that almost everyone would want to keep the past as it is because
they want desperately to keep the basic features of their personality
as they are? This would seem to be conservative in the extreme and to
be based on an unwarranted assumption that almost everyone is suf-
ficiently satisfied with their personalities that they would be unwilling
to take the risks associated with having a very different personality.
Now we want to emphasize that it does seem reasonable that most
people would not prefer to have been “someone else” (in the sense
of having a very different personality, and—importantly—not being
able to select the nature and features of this personality). But this is
not enough for Belshaw’s stated purpose, because he believes that the
desire to hold fixed the time of one’s birth is almost universal. At the
beginning of “Death, Pain, and Time,” Belshaw states the point starkly:
“We wish to die later. But we don’t wish to have been born earlier. Our
future nonexistence matters to us in a way that past nonexistence does
not.”14 Later he criticizes approaches that seek to explain our asymmet-
rical attitudes toward prenatal and posthumous nonexistence in terms
of our attitudes toward pain and pleasure, saying “our asymmetrical
attitude regarding pleasures and pains is, I shall maintain, complex and
untidy, while that concerning nonexistence is more straightforward,
and relatively easy to understand.”15
But the considerations discussed above indicate that it is not plausi-
ble to assume that there is an almost universal desire to maintain one’s
personality as it is (and not risk assuming another personality, over the
features of which one has no control). The situation here is more “com-
plex and untidy” than Belshaw supposes. If this is so, and if there is
indeed an almost universal lack of a wish to have been born earlier,
then this wish (and the attendant asymmetry in our attitudes toward
prenatal and posthumous nonexistence) cannot be explained in terms
of Belshaw’s “conservation claim.”16
NOTES
1. Thomas Nagel defends the deprivation account in “Death,” reprinted in
Mortal Questions (Cambridge: Cambridge University Press, 1979). For develop-
ments of this approach, see, for example, Brueckner and Fischer, chapter 2; and
F. M. Kamm, “Why Is Death Bad and Worse than Pre-Natal Non-Existence?”
Pacific Philosophical Quarterly 69 (1988), pp. 168–174; and Morality, Mortality:
Volume I (New York and Oxford: Oxford University Press, 1993), esp. pp. 13–71.
2. Nagel, 1979, pp. 8–9.
3. For a discussion of these issues, see Brueckner and Fischer, chapter 2.
4. Frederik Kaufman, “Pre-Vital and Post-Mortem Non-Existence,”
American Philosophical Quarterly 36 (1999), pp. 1–19.
5. Ibid., p. 14.
6. It is not clear that Kaufman needs so strong a conclusion as that it is
impossible for an individual to have been born significantly earlier than he actu-
ally was born. It would seem to be sufficient, for Kaufman’s purposes, that
certain counterfactuals would be true (whether or not the impossibility claim is
true). So it would seem to be sufficient that it is true that anyone born consider-
ably earlier would not be identical to the individual in question. And this could
be true compatibly with the (perhaps remote) possibility that the individual in
question could have been born considerably earlier. The counterfactuals and
the possibility claims correspond to different modalities, and strictly speaking,
what appears to be relevant to Kaufman’s argument are merely the counter-
factuals. For a discussion of the logic of arguments involving these two impor-
tantly different kinds of modalities, see John Martin Fischer, The Metaphysics
of Free Will (Oxford and Cambridge, Mass.: Blackwell Publishers, 1994), esp.
pp. 87–110.
7. Kaufman, 1999, p. 11.
8. Ibid., p. 11.
9. Ibid., p. 12.
10. Ibid., p. 12.
11. Ibid., p. 12.
12. Consider the following parable, which we owe to Glenn Pettigrove.
Odysseus, on his long journey home from Troy, is paid a visit by Zeus one
Mediterranean afternoon. Zeus informs Odysseus that the gods have been con-
templating human attitudes toward life and death. Not surprisingly, a heated
dispute has arisen on Mount Olympus. In order to resolve the dispute, Zeus
has proposed that they perform a little experiment. Because of his reputation
for craft and cunning, Odysseus has been selected as the subject of the experi-
ment. The experiment is simple, requiring nothing more from Odysseus than a
single choice between two options: (1) complete, eternal annihilation, effective
tomorrow, or (2) to be born fifty years earlier than he had been and live to the
ripe age of one hundred (twice as long as his lifetime under option 1), at which
time he would be completely and eternally annihilated.
After securing from Zeus the assurance that the life under option (2) would
contain at least as much honor as his current life, it seems likely that the cun-
ning Odysseus would choose an earlier birth over an immediate death. And
it seems that many, if not, most of us would choose similarly. Of course, in
Kaufman’s terminology the choice would be for our metaphysical essence to
have been associated with a different thick person. Perhaps Kaufman would
contend that upon careful reflection most persons would not make the choice
for earlier birth, but it is not at all clear that this is so. Our attitudes toward the
time of our birth—and thus the possibility of having a different thick self—are
more complex and ambivalent than Kaufman supposes.
13. Christopher Belshaw, “Death, Pain, and Time,” in Philosophical Studies 97
(2000), pp. 317–341, esp. pp. 324–325.
14. Ibid., p. 317.
15. Ibid, p. 317.
16. We are grateful to Dominick Sklenar for helpful conversations about the
topics of this essay, and to Frederik Kaufman and Glenn Pettigrove for gener-
ous comments on a previous version.
5
Earlier Birth and Later Death:

Symmetry Through Thick and Thin
John Martin Fischer
You see, he [Bob Marshall, the first Adirondack 46er and the founder of the Adirondack
Society] had been so dreadfully afraid, so sure that he had missed it all, born too late ever
to taste the freedom of the wilderness as, he imagined, Lewis and Clark had tasted it long
ago in the wilder west. He tried to be rational about the deprivation. He allowed as how,
had he been privileged to join that ‘most thrilling of all American explorations,’
he probably would have been ‘bumped off’ by Indians or typhoid fever before he was 25.
. . . Always, even after his most remote bushwhacks in the wilds of Idaho and Alaska,
he would remember this lakeside country [the Adirondacks]: ‘Real wilderness to me,
as exciting in a different way as the unexplored continent which I had missed by
my tardy birth’.1
I. THE ISSUES
Although we human beings typically think that death can be a bad
thing or misfortune for the individual who dies, it is not easy to
explain why or how death can be a bad thing, when (and if) it in fact
is. (The commonsense view about the possibility of death’s badness
need not be interpreted so that it entails that death is always bad
for the individual who dies, or that immortality would always and
under all circumstances be good.) A natural way to seek to explain
death’s badness, when it is indeed bad, is to say that death deprives
the individual of what would otherwise have been, on balance,
good. Since death deprives the individual of life, it is bad insofar
as it deprives the individual of what would have been (overall)
good. In his classic paper, ‘Death,’ Thomas Nagel (1970) suggested
this sort of explanation of death’s badness, and various philoso-
phers have followed him in defending the deprivation thesis about
death’s badness.2
The deprivation account of death’s badness has various problems.
It must explain how a mere deprivation—an experiential blank in this
63
instance—can be bad for an individual. (Normally, we suppose that

only unpleasant experiences can be genuinely bad for an individual,
or at least that such badness includes or has as an accompaniment
unpleasant experiences.) Also, if the dead individual no longer exists,
how can he or she be harmed? That is, who is the subject of the harm (or
bad or misfortune), and when exactly does this harm take place? These
and other conundra have received considerable attention, and I will not
address them in this essay.
Rather, I shall focus here on what Nagel (1986, p. 228) has called the
most puzzling aspect of the family of puzzles pertaining to death: the
intuitive asymmetry between prenatal and posthumous nonexistence.
Lucretius, a Roman follower of Epicurus, agreed with the Epicurean
idea that we should not fear our deaths, or regard them as something
whose prospective occurrence should be deemed undesirable, pre-
cisely because death cannot be a bad thing for the individual who has
died. One of the arguments presented by Lucretius is the Symmetry
Argument. Because death (considered as an experiential blank—an
indefinitely long period of nonexistence) is the mirror image of birth,
and since we do not consider it bad or regrettable that we were born
when we were born, rather than considerably earlier, we should not
consider it an unfortunate or undesirable thing that we die when we
actually die, rather than considerably later. After all, birth at the actual
time of our births, rather than earlier, is a deprivation in a way similar
to the way in which death at the actual time of our deaths, rather than
later, is a deprivation. Holding fixed the time of my birth, if I were to
die later than I actually die, I would have more life (good or bad, on
balance); similarly, holding fixed the actual time of my death, if I were
to have been born earlier than I actually was born, I would have more
life (good or bad, on balance).
Of course, one could turn the Symmetry Argument on its head and
conclude that we should begin to regret the actual time of our births.
This is precisely what Fred Feldman does:
There are, after all, two ways in which we can rectify the apparently
irrational emotional asymmetry [between late birth and early death].
On the one hand, we can follow Lucretius and cease viewing early
death as a bad thing for Claudette. On the other hand, we can at least
try to start viewing late birth as a bad thing. My suggestion is that in
the present case the latter course would be preferable. I think it must
be granted that our emotional reactions toward pleasures lost by
early death are quite different from our emotional reactions toward
similar pleasures lost by late birth. If my proposal is right, this emo-
tional asymmetry is irrational (1992, p. 222).
It is open to Feldman to point out that, just as it is not rational to be
excessively concerned or preoccupied by our eventual deaths, it is also
Earlier Birth and Later Death: Symmetry Through Thick and Thin 65
not appropriate to focus excessively on the fact of our late births. But
it has seemed to many that our judgment that we are (and should be)
indifferent to our prenatal nonexistence is on firmer ground than our
prospective judgments about our future deaths. This, in conjunction
with the mirror-image claim, yields the view that we ought to be indif-
ferent about our prospective deaths (or, at least, that we should regard
the fact that we will die at a particular time, rather than a later time,
as a matter of indifference). The Symmetry Argument thus provides a
compelling challenge for the proponent of the deprivation account of
death’s badness.
In his article, ‘Death,’ Nagel famously suggested a potential explana-
tion of our asymmetry in attitudes in terms of a certain sort of asym-
metry of possibility: whereas it is possible that an individual die later
than he actually die, it is in some important sense impossible that an
individual—that very same individual—be born considerably earlier
than the actual time of his or her birth. In an important footnote, Nagel
expressed doubts about this explanation:
I confess to being troubled by the above argument, on the grounds
that it is too sophisticated to explain the simple difference between
our attitudes to prenatal and posthumous nonexistence. For this rea-
son I suspect that something essential is omitted from the account
of the badness of death by an analysis that treats it as a deprivation
of possibilities. My suspicion is supported by the following sugges-
tion of Robert Nozick. We could imagine discovering that people
developed from individual spores that had existed indefinitely far
in advance of their birth. In this fantasy, birth never occurs naturally
more than a hundred years before the permanent end of the spore’s
existence. But then we discover a way to trigger the premature
hatching of these spores, and people are born who have thousands
of years of active life before them. Given such a situation, it would
be possible to imagine oneself having come into existence thousands
of years previously. If we put aside the question whether this would
really be the same person, even given the identity of the spore, then
the consequence appears to be that a person’s birth at a given time
could deprive him of many earlier years of possible life. Now while
it would be cause for regret that one had been deprived of all those
possible years of life by being born too late, the feeling would dif-
fer from that which many people have about death. I conclude that
something about the future prospect of nothingness is not captured
by the analysis in terms of denied possibilities. If so, then Lucretius’s
argument still awaits an answer (pp. 370–71).
The Nozickean spore example has its analogues in contemporary bio-
medical science. That is, it is (or may soon be) possible to freeze a fer-
tilized sperm and egg, and thaw them out at some later time. Apart
from considerations about technological feasibility, it certainly seems

to be genuinely conceivable (not conceptually incoherent) and thus
metaphysically possible that a particular sperm and egg have existed
considerably earlier than they actually existed; thus, it is at least meta-
physically possible (even if not feasible) that a particular individual
have been born considerably earlier than he or she actually was born.3
II. THE BELSHAW/KAUFMAN SOLUTION:

ASYMMETRY OF POSSIBILITY
Recently, Frederick Kaufman (1999, 2000, unpublished) and
Christopher Belshaw (1993, 2000a, 2000b) have (in slightly different
ways) defended Nagel’s original suggestion of the asymmetry of pos-
sibility: although it is possible for us to die later, it is not possible for us
to have been born much earlier. More specifically, these proponents of
the ‘asymmetry of possibility’ view distinguish what might be called
a ‘thin’ notion of selfhood or the individual from a ‘thick’ notion. The
thin notion corresponds roughly to metaphysical personal identity
(understood in a certain way); the thin self is perhaps individuated
by reference to ‘bodily identity’ or, (slightly) more carefully, in terms
of identity of the sperm and egg from which the body has developed.
(Alternatively, I suppose the thin self could be understood in terms of
a ‘soul’ or simple, nonphysical substance without particular mental
contents.) The thin self can remain the same (metaphysically identi-
cal) through significant changes in ‘mental’ characteristics, including
memories, preferences, beliefs, intentions, plans, values, and charac-
ter traits. In contrast, the ‘thick’ self (to a rough first approximation
that will be refined below) includes the mental characteristics. It cor-
responds to the idea of personality or basic character (as given by or
generated from interaction with a particular set of experiences).
In this essay, I shall focus primarily on Kaufman’s development of
the asymmetry of possibility view (although I shall also allude briefly
to aspects of Belshaw’s position, as appropriate). Kaufman says:
When we ask whether a currently existing person could exist at an
earlier time and still be the same person who exists now, we need
to be careful about how we construe ‘person,’ a multiply ambigu-
ous concept. I shall argue that a version of Nagel’s original response
to the symmetry argument can be sustained if we are clear about
the account of persons that is relevant to the issue of the badness of
death.
Consider an account of persons which focuses on metaphysical
essences. A person, on this view, is simply a particular essence, and
that person exists in all possible worlds which contain that essence.
The details of one’s actual life are wholly contingent features of an
individual. On this understanding of ‘person,’ since the features of

one’s actual life are not in any way constitutive of the person one is,
it is possible for one to be shorn of all the attributes of one’s actual life
and remain the same person throughout the changes (1999, p. 11).
This is the ‘thin’ notion of personhood or ‘selfhood.’ In contrast stands
the ‘thick’ notion:
. . . [I]nsofar as concern about death is driven by concerns that one’s
conscious personal existence will be extinguished forever, the fact
that one’s metaphysical essence might occur in different times and
places seems beside the point. . . . Hence, in discussions about the evil
of death, a fuller account of ‘person’ is needed, one which incorpo-
rates the defining characteristics of our lives, characteristics which
make up our conception of ourselves as particular persons with par-
ticular memories, beliefs, projects, and commitments. This, after all,
is what death deprives us of.
. . . While large differences in the time or place of our thin selves
will produce persons unrecognizable to the thick persons we cur-
rently are, less dramatic changes will result in persons with whom
we will bear more or less affinity. . . . when I reflect on my biogra-
phy, it seems that being in the right (or sometimes the wrong) place
at the right time makes all the difference. The opportunities seized
(or lost), the accidents avoided (or encountered), the people met,
the moral and physical challenges thrust upon me; these and many
other life-defining events all depend on my existing at a particular
time. Indeed, were my essence to have existed even slightly earlier,
the effect on my present conscious personal life would be enormous.
that I could have existed earlier (1999, p. 14).
Kaufman applies the distinction between thin and thick personhood to
Lucretius’s challenge as follows:
. . . So far all we know is that thick persons cannot exist earlier than
they do. But if one’s biography is, in some broad sense, constitutive of
one’s person, then it also seems that one cannot die later than one will,
for longer life will change the conditions of one’s life, and thus pro-
duce a different person. In other words, if thick persons cannot exist
earlier, neither can they die later—the symmetry argument again.
However, this objection can be met by noting that additions to
one’s biography do not necessarily produce a different person. Thick
persons can, as it were, become ‘thicker,’ that is, acquire additional
features without undoing what is already there. Thus, I, as I currently
am, could go live with the Eskimos; but I, as I currently am, could
not have lived with the Eskimos all along.
. . . Irrespective of whether a future self would be me, the impor-
tant point is that it is possible for some future self to be me, since
additions to one’s biography need not disrupt what has already been
established. If the appropriate connections among my current psy-
chological states and those of a future self hold, then, subjectively, I
can reidentify myself as myself at a later time. What strains credulity,
though, is how a prior existing self could be me, thickly conceived,
since attempting to imagine myself existing at an earlier time does
disrupt my biography thus far. . . . A life might unfold in a variety
of ways, but imagining it starting earlier destroys the current bio-
graphic self, and with it the subject who can be deprived by death
(1999, p. 15).
In a recent paper, Kaufman encapsulates the view as follows:
I am suggesting, with Parfit, that metaphysical personal identity is
not ‘what matters’ when we contemplate life and death; what mat-
ters is psychological continuity. Concern about one’s own death has
very little if anything to do with the extinction or continuation of
one’s metaphysical personal identity. Questions about death engage
us at an intimate level and our concerns are understandably about
the psychological continuum that we conceive ourselves to be. A
god’s eye perspective that identifies an earlier existing person as
metaphysically identical to me seems beside the point.
A life conceived as a psychological continuum of overlapping
memories, intentions, beliefs, commitments and structured by partic-
ular projects, events, loved ones and concerns, shows us how a person
can be deprived of time after death but not similarly deprived of time
before birth. Whereas a given psychological continuum can readily
accommodate additional future experiences without disrupting its
previous narrative structure, no existing narrative could accommo-
date earlier experiences and retain its previous narrative structure,
given contingent facts of human life. What you do tomorrow does
not disrupt what you have already done, but you cannot retain your
current biography if ‘you’ were alive 100 years ago. For you would
have to picture yourself remembering, say, your favorite TV show as
a child and also amusing yourself before television was invented.
. . . Thus while a given person can conceivably die later than he or
she in fact does, that same person (read: structured psychological
continuum) cannot exist earlier and retain the biography of any cur-
rently existing person (unpublished, pp. 14–5).
It will be helpful (in coming to grips with the ideas suggested above) to
consider the dilemmatic argument presented by Kaufman (2000). Either
we are concerned about the thin notion of the self or the thick notion,
when we are worried about our own death. In order for an individual to
be deprived of something (in the relevant sense), he or she must at least
be capable of desiring or preferring it; but a thin self has (and can have) no
desires or preferences. And, whereas the thick self can have preferences,
it is impossible that the thick self have existed earlier. What is impossible,
in this sense, cannot be considered a meaningful deprivation.4
III. REPLY TO THE ASYMMETRY OF POSSIBILITY VIEW

Consideration of Kaufman’s development of the Asymmetry of Possibility
view will help us to gain a more nuanced understanding of the subject
of the harm of death and also what we care about in caring about con-
tinuing to live. I begin by simply noting that we do in fact sometimes
make judgments that appear to involve regret or disappointment at not
having been born or raised in very different circumstances. I (and my
coauthor) have discussed such judgments in chapter 4.5
Additionally, Glenn Pettigrove says:
The contestable nature of Kaufman’s claim comes to the fore when
one considers a variety of religious positions on the afterlife. Most
doctrines of reincarnation hold that one has no conscious recollec-
tion of memories from one life to the next. If there is any remnant
which one might loosely call a character which perdures over sev-
eral incarnations, it is something well below the conscious, psycho-
logical threshold with which Kaufman is concerned. Nevertheless,
untold numbers have found comfort in the face of death through
a doctrine of reincarnation. Several Buddhist teachings provide an
even more startling contrast to Kaufman’s claim, insofar as the meta-
physical substance which was connected with a particular biological
and psychological profile merges indistinguishably with the vast sea
of metaphysical substance which gives life to the cosmos. Here again
hundreds of generations have found comfort in the thought that
their lives would not end with death, even though their psychologi-
cal continuum would be broken and their metaphysical substance
would not even be readily isolable after death (2002, pp. 410–1).
I have some doubts about the coherence of the various doctrines of rein-
carnation alluded to by Pettigrove.6 I find plausible Bernard Williams’
contention that it is not rational now to care about the future states of
an entity with a character not suitably related to one’s character now
(although, of course, it is tricky to specify the nature of the relationship
in question).7 It is even more difficult to understand how it can be com-
forting to envisage a future in which one has merged metaphysically
with The Infinite.8 But it seems to me natural to take seriously the idea
that we could have been born and raised in very different circumstances;
it is not obvious that this is based on a confusion. After all, it seems that
one might regret not having been born to wealthier parents, or not hav-
ing been born and raised in a more civilized century, and so forth.
But certainly, Kaufman might argue that our ordinary, commonsense
ways of expressing these thoughts might not survive critical scrutiny.
What about the first horn of Kaufman’s dilemma, according to which
one cannot be deprived of anything unless one has the capacity to desire
or prefer it, and a thin self cannot have such mental states? Of course,
there are well-known problems with the attempt to connect deprivation
too closely with desires and preferences. For example, a newborn infant
does not have the preference that his inheritance (which is in a trust
fund on his behalf) not be embezzled; yet embezzling it would deprive
him of something to which he is entitled and thus harm him. It is pre-
sumably necessary to refine the account of the connection between the
relevant mental states and deprivation (and harm) to invoke the poten-
tial to develop such mental states; but this sort of revision threatens to
lead to problems of its own (such as the apparent implication that abor-
tion at even an early stage would be wrong).
I would suggest that when we speak of deprivation and harm, the
subject is not the thin self (qua thin self), but some more complex entity,
such as the thin self conjoined with a capacity to step back from any
particular thick self and evaluate such selves. After all, we have points
of view or perspectives on the world, and the ‘I’ that is a potential sub-
ject of deprivation and harm is a conscious entity with a point of view.
This potential subject of deprivation and harm can be considered an
aggregate or combination that includes at least a thin self (‘metaphysi-
cal essence’) and the capacity to step back from any particular associ-
ated thick self, evaluate it in light of various possible thick selves, and
form desires and preferences about which thick selves should be asso-
ciated with one’s thin self (metaphysical essence). Perhaps the subject
of the harm is an aggregate of three components: a thin self, a particular
thick self, and the associated capacity to step back and reflect and form
preferences of the sort sketched above. In any case, the ‘I’ who can be
deprived by early death and also late birth can be interpreted to include
the relevant capacity. Thus, there is no incoherence in supposing that
an individual (the appropriate aggregate individual) can be harmed by
an early death, even if it is granted that there is some plausible sort of
connection between deprivation, harm, and preferences.
Now the question of whether I can be deprived and thus harmed
by late birth becomes: is it possible for my thin self to have been born
considerably earlier with the same sort of thick self as I currently have?
And I do not see why not. Note, first, that it is coherently conceivable,
and thus metaphysically possible, that one’s thick self have come into
being considerably earlier than it actually did (contrary to the second
horn of Kaufman’s dilemma). Here I am supposing that we need not
hold all of the details of our lives fixed, in order to have the same thick
self (in the relevant sense). It is just implausible that I care about all of the
details of my current thick self, and Kaufman agrees, saying, ‘Not every
detail of our lives is absolutely necessary for us to be the psychological
persons we currently are’ (1999, p. 14).9 So, for instance, it is implausible
that I care about holding fixed the exact times of my various experiences,
given that their fundamental contents are not altered. And it appears to
be coherently conceivable, and thus metaphysically possible, that my
thick self have come into existence, with all of its fundamental content,
significantly prior to its actual birth.
Note that this claim is no less defensible than a claim Kaufman is
willing to concede: that the thin self could have been born significantly
prior to its actual time of birth. After all, if the Kripkean view of the
essential features of the thin self is correct (and Kaufman does not reject
it), then in order for my thin self to have been born significantly earlier,
my parents would have had to have existed significantly earlier, and
(among other things) the very same sperm and egg would have had to
have been fertilized significantly earlier, and so forth. Of course, these
scenarios are wildly and fantastically implausible, and yet Kaufman is
willing to suppose that they are at least metaphysically possible.10 I do
not see how it would be relevantly different with respect to thick selves.
Yes, it is wildly and fantastically implausible that the same thick self
should have existed significantly earlier. But that is not the issue: just
as with the thin self, it is at least possible that the thick self could have
existed considerably earlier. So we do not as yet have an asymmetry of
possibility (given the assumption, shared by everyone, that the thick
self can die later than it actually dies).
The claim that it is possible, in the relevant sense, for my thick self
to have come into being (along with my thin self) considerably earlier
than it actually did points us to possible worlds that involve radical
departures from the actual world. Since it is antecedently wildly and
totally implausible and unlikely that such a thing should happen (that
the same thick and thin selves came into being considerably earlier than
they actually did), the possible worlds in which this does indeed happen
are very ‘distant’ in logical or metaphysical space from the actual world.
A proponent of the asymmetry of possibility view might seek to invoke
this fact in some way, but I do not see how this is relevant to the argu-
ment. The argument posits a certain sort of asymmetry of metaphysi-
cal possibility, but there is no such asymmetry; the implausibility of the
possibilities in question (and thus the distance in logical or metaphysical
space of the relevant possible worlds) is quite beside the point.
It is significant to note that it is not uncommon, and arguably not
clearly irrational, to regret the fact that a rather unlikely scenario did
not become actual. Suppose one buys a lottery ticket, and it does not
win. Still, one might regret that it did not win, knowing full well that the
scenario in which this particular ticket wins is extraordinarily unlikely.
Unlikely scenarios, so long as they are metaphysically possible, can cer-

tainly be the object of hope (prospectively) and regret and disappoint-
ment (retrospectively). It is clearly not absurd to regret that one’s ticket
lost, and to regard it as unfortunate for one—even while admitting that
it was an extreme long shot. Hope and disappointment can take as their
objects possible scenarios that are unlikely (and thus correspond to dis-
tant possible worlds). One can find many examples of this sort of phe-
nomenon, ranging from the stock market to the philosophical job market
to one’s loyal support for a particular sports team. I do not see any way
to argue that it is not rational to regret, or consider it unfortunate for me,
that my ticket did not win the lottery (or that I did not get a particular
philosophy position for which I had the no-doubt-shocking temerity to
apply, or even that the San Francisco Giants failed yet again to win the
pennant . . .).11
Now it certainly is true that we ought not waste much mental energy
thinking about remote possibilities. It would follow that I should not
excessively focus on or obsess over my lack of success in the lottery (or,
say, my failure, yet again, to secure that endowed chair in philosophy
at Harvard). It is indisputably true that one should not focus excessively
on these matters. But, of course, it is equally the case that one ought
not focus too much on death—on the fact that I will die when I actually
die and not later. I do not see any difference in this respect between
our attitudes toward regrettable (although unlikely) scenarios and our
attitudes toward matters concerning death.
Both Kaufman and Belshaw seek to brush aside wildly implausible
scenarios as irrelevant. Kaufman says:
that I could have existed earlier (1999, p. 14).
If this is correct, the same should be said about the thin self, as pointed
out above. But Kaufman does not explicitly conclude that it is meta-
physically impossible for the thin self to have been born earlier. In fact,
as noted above, he concedes that this is metaphysically possible. The
whole point of the distinction between thin and thick notions of the self,
I would have thought, is that, whereas thin selves are conceded to have
the possibility of considerably earlier births, thick selves are not; but I
do not see how Kaufman, Belshaw, or anyone could defend this differ-
ence. Further, given that it is not obviously irrational or inappropriate
to regret the nonoccurrence of even extremely unlikely (but metaphysi-
cally possible) scenarios, it has not been shown that our intuitive asym-
metry in attitudes toward posthumous and prenatal nonexistence can
be explained and justified.12
IV. MY ANSWER TO LUCRETIUS AND A DEFENSE

I and my coauthor, Anthony Brueckner, have suggested an alternative
answer to Lucretius’s challenge. We have argued that the intuitive, wide-
spread asymmetry in our attitudes toward posthumous and prenatal
nonexistence (or, if one prefers, experiential blanks) is a special case of
a more general asymmetry in our attitudes toward our past and future
experiences. More specifically, we care especially about future pleasures,
and we are relatively indifferent to past pleasures. Since death deprives
us of future pleasant experiences, whereas late birth merely deprives us
of past pleasures, it is not surprising that we care about the time of our
death, but are relatively indifferent to the time of our birth. The specific
attitudinal asymmetry is an instantiation of a more general, apparently
deep-seated asymmetry in our attitudes toward the future and past.13
Kaufman however offers the following criticism:
Brueckner and Fischer purport to explain our asymmetric attitudes

about the time before we exist and the time after our deaths by sub-
suming those attitudes under a more general Parfitian temporal bias.
If we are interested only in explaining why our attitudes about these
allegedly symmetrical times are so different, Brueckner and Fischer
have indeed offered an explanation. But a general temporal bias is
still a bias. If we ask instead whether we are justified or warranted or
correct in holding such different attitudes about the time before our
births and the time after our deaths, then Brueckner and Fischer’s
explanation for our having different attitudes misses the point. One
does not justify a bias by putting it in the context of a larger bias; a
racial bias is not justified by pointing out that people are xenophobic
(perhaps because of evolution), even though a general xenophobia
might explain the prevalence of racial bias.
Brueckner and Fischer move from noting our preferences about
prenatal and post mortem nonexistence to insisting on the rational-
ity of those preferences. But our temporal preferences are not ratio-
nal just because they are preferred (unpublished, p. 5).
Parfit struggles with the status of the general asymmetry in our atti-
tudes. Although Parfit suspects that the general asymmetry is irratio-
nal, he does not argue for this view, and he only concludes that the
asymmetry is, at best, not irrational (rather than rational). If the specific
asymmetry in our attitudes about death and birth is not irrational, then
this would indeed provide an answer of some degree of plausibility
to the challenge posed by Lucretius. After all, Lucretius contends that
since prenatal and posthumous nonexistence are mirror images of each
other, our asymmetric attitudes are irrational.
I suppose however that one can pose a deeper challenge on behalf
of Lucretius. One can ask how it can be rational to have the asymmetric
attitudes, given the mirror-image fact. Granted, we have a natural ten-

dency toward the asymmetry, and it is not obviously irrational. But why
is it rational? Why continue to have the asymmetry? What is the point?
I do not know how to answer this difficult and probing question in
a completely satisfactory way. I would like to end by trying, in a pre-
liminary way, to sketch an approach to addressing the question. This
approach, I fear, will leave more questions unanswered, but perhaps it
will be fruitful as a way of beginning to think about the general asym-
metry. At the macroscopic level in our universe, causation flows from
past to present to future. That is, causation goes forward, and not back-
ward, in time. Thus, insofar as we care about bringing about effects in
the world, we care especially and distinctively about the present and
future (rather than the fixed past). There is a clear survival benefit to
creatures who care especially about the future, so from a purely evolu-
tionary perspective, there seems to be a ‘point’ to some sort of general
asymmetry in our concern about the past and future. Given this, and
the difficulty of ‘fine-tuning’ such an asymmetry in attitudes, it would
not be surprising (or inappropriate) that we have the general asymme-
try in our attitudes toward our own future pleasures and our own past
pleasures.
Again, presumably creatures with this specific sort of attitudinal
asymmetry will have a greater chance of maximizing pleasure over
time, and, arguably, being happier. Again, there is a clear survival
advantage to having such an asymmetry. As above, given the difficulty
of ‘fine-tuning’ such an asymmetry—of turning it off and on as the
context requires—it would not be surprising or inappropriate that we
would have asymmetrical attitudes toward our own future pleasures
and our own past pleasures, even in particular instances in which it is
clear that such an asymmetry will not affect one’s long-term pleasure
or happiness (or chances for survival). I hope it will emerge, at least in
a preliminary way, that there can be a rationale or nonarbitrary point
to the Parfitian general asymmetry of which our asymmetric attitudes
toward early death and late birth are a specific instance. Insofar as it is
rational to care about pleasure, happiness, and even survival, the gen-
eral asymmetry is, arguably, rational. (Of course, one could call into
question the rationality of caring about such things, but it is at least
plausible that these are extremely widely regarded as rational to care
about.14)
I do not think it is plausible that being xenophobic can be justified in
a similar way (by a good evolutionary argument). Certainly it is plau-
sible that favoring one’s own interests and those of one’s family and
friends (and perhaps nation) can have an evolutionary justification. But
certain kinds of (excessive and inappropriate) fear of different people,
and associated attitudes (say, of racism), are not similarly justifiable
—indeed, one supposes that they could plausibly be shown to be disad-
vantageous across a wide variety of circumstances.15
REFERENCES
Belshaw, Christopher (1993), ‘Asymmetry and Nonexistence,’ Philosophical
Studies, Vol. 70, pp. 103–16.
—— (2000a), ‘Death, Pain, and Time,’ Philosophical Studies, Vol. 97, pp. 317–41.
—— (2000b), ‘Later Death/Earlier Birth,’ in Peter A. French and Howard
K. Wettstein, eds., Midwest Studies in Philosophy, Vol. 24 (2000), pp. 69–83.
Feldman, Fred (1992), Confrontations with the Reaper: A Philosophical Study of the
Nature and Value of Death, Oxford University Press, New York.
Fischer, John Martin, ed. (1993), The Metaphysics of Death, Stanford University
Press, Stanford.
Fischer, John Martin (1994), ‘Why Immortality is Not So Bad,’ International
Journal of Philosophical Studies, Vol. 2, pp. 257–70 (chapter 6).
Fischer, John Martin and Brueckner, Anthony (1986), ‘Why Is Death Bad?,’
Philosophical Studies, Vol. 50, pp. 213–21 (chapter 2).
Fischer, John Martin and Brueckner, Anthony (1993a), ‘The Asymmetry of Early
Birth and Late Death,’ Philosophical Studies, Vol. 71, pp. 327–31.
Fischer, John Martin and Brueckner, Anthony (1993b), ‘Death’s Badness,’ Pacific
Philosophical Quarterly, Vol. 74, pp. 37–45.
Fischer, John Martin and Curl, Ruth (1996), ‘Philosophical Models of Immortality
in Science Fiction,’ in Immortal Engines: Life Extension and Immortality in
Science Fiction and Fantasy, University of Georgia Press, Athens, GA, pp. 3–12
(appendix to chaper 6).
Fischer, John Martin and Speak, Daniel (2000), ‘Death and the Psychological
Conception of Personal Identity,’ in Peter A. French and Howard K. Wettstein,
eds., Midwest Studies in Philosophy, Vol. 24, pp. 84–93 (chapter 4).
Kaufman, Frederick (1999), ‘Pre-Vital and Post-Mortem Non-Existence,’
American Philosophical Quarterly, Vol. 36, pp. 1–19.
Kaufman, Frederick (2000), ‘Thick and Thin Selves: Reply to Fischer and
Speak,’ in Peter A. French and Howard K. Wettstein, eds., Midwest Studies in
Philosophy, Vol. 24 (2000), pp. 94–7.
Kaufman, Frederick (unpublished), ‘Earlier Birth and Later Death: The Answer
to Lucretius.’
Nagel, Thomas (1970), ‘Death,’ Noüs, Vol. 4, pp. 73–80. Reprinted in Fischer
1993, pp. 59–69 and pp. 370–1 (references to page numbers below are to the
reprinted article).
Nagel, Thomas (1986), The View From Nowhere, Oxford University Press, New York.
Parfit, Derek (1984), Reasons and Persons, Clarendon Press, Oxford, pp. 165–85.
Perry, John (1978), A Dialogue on Personal Identity and Immortality, Hackett
Publishing Co., Indianapolis.
Pettigrove, Glenn (2002), ‘Death, Asymmetry and the Psychological Self,’ Pacific
Philosophical Quarterly, Vol. 83, pp. 407–23.
Williams, Bernard (1973), ‘The Makropulos Case: Reflections on the Tedium of
Immortality,’ in Problems of the Self, Cambridge University Press, Cambridge,
pp. 82–100. Reprinted in Fischer, 1993, pp. 73–92 and pp. 371–2.
NOTES
1. From The Adirondack Explorer, January 2001, p. 33 (this article first
appeared in Wilderness magazine). I am grateful to Steve Schwartz for this
quotation. Steve also informs me that an Adirondack 46er is someone who has
climbed all 46 peaks in the Adirondacks that are over 4000 feet. As he puts it, ‘It
is kind of like getting your Ph.D. in masochistic outdoors activity.’
2. Other defenses of the deprivation account include: Feldman, 1992;
Fischer and Brueckner, 1986 (chapter 2), 1993a, and 1993b; Fischer and Speak,
2000 (chapter 4).
3. See Brueckner and Fischer, 1986.
4. Kaufman himself puts the dilemma as follows: ‘I conclude that thick me
cannot be deprived by there not being a different thick self in its stead, because
thick me cannot be a different self, irrespective of how much I might wish it
to be so; and since thin me has no preferences, it too cannot be deprived by
not having had a different thick self attached to it.’ But surely the issue is not
whether thick me can be a different self, but whether thick me (holding fixed
the fundamental content, but not necessarily all the details) could have existed
at a much earlier time.
5. For similar examples, see Pettigrove, 2002.
6. Kaufman expresses his own doubts: ‘This is why certain possible occur-
rences that leave my metaphysical essence intact but which nevertheless extin-
guish my subjective sense of myself as myself are things which, like death,
I could not survive, such as brain zaps, philosophical amnesia, permanent coma;
some versions of reincarnation, or “merging with the infinite” ’ (1999, p. 11).
7. See Williams, 1973, and Fischer, 1994.
8. There are interesting treatments of ‘fusion’ of selves in various works
of science fiction. For a discussion, see Fischer and Curl, 1996 (appendix to
chapter 6).
9. For a critical discussion of the extreme view that we care about holding
fixed all of our thick selves, what Belshaw calls the ‘Conservation Thesis,’ see
Fischer and Speak, 2000, pp. 91–2. It is very implausible that I want to hold fixed
even the fact that I broke a blood vessel in my leg when I was young and had
to stay in bed for weeks. After all, I can be confident that I would still have the
same basic personality and character, and the same important configuration of
memories, even sans this experience and associated memories . . . .
10. Kaufman says, ‘I am willing to concede, in other words, that on a thin
account of “person,” we could have existed earlier. But I deny that that concep-
tion of “person” is what is at issue when we ask about the badness of death’
(1999, p. 12).
11. Gretchen Weirob, the philosophy professor (on her deathbed) in John
Perry’s (1978) delightful, A Dialogue on Personal Identity and Immortality, says:
‘Even the possibility of something quite improbable can be comforting, in cer-
tain situations. When we used to play tennis, I beat you no more than one time
in twenty. But this was enough to establish the possibility of beating you on
any given occasion, and by focusing merely on the possibility I remained eager
to play. Entombed in a secure prison, thinking our situation quite hopeless, we
may find unutterable joy in the information that there is, after all, the slimmest
possibility of escape. Hope provides comfort, and hope does not always require
probability’ (p. 2).
12. David Hershenov (in personal correspondence) makes an interesting
point. He says: ‘Leaving aside the possibility that the thick self exists earlier,
aren’t all accounts of a thick or biographical or narrative self committed to a first
moment or (vague period) of existence in which there is not much of a thick self
or personality or narrative? (Doesn’t there logically have to be a beginning of a
narrative?) Since we existed at that time with very little in the way of a settled
psychology and looked forward to a future, couldn’t we at that state have been
open to many possible experiences and thick selves developing? We could have
been placed in countless different cultures and would have cared about our-
selves and our futures. If that is so, then at least at a certain time in our existence
we could have expressed prudential concern, but without a thick narrative or
biographical self. We could have been happy at that moment raised a Moslem
in Iran in another time period or Christian in modern America. So at least at one
time in our existence, we should have been neutral between earlier (pre-birth)
deprivation and future deprivation (so far as the narrative-self approach is con-
cerned). And if we could care equally then about any number of thick selves we
might acquire, why couldn’t we recapture that concern?’
13. See chapter 2.
14. Note that I am not here contending that we can analyze or somehow
‘reduce’ value notions, such as ‘good’ and ‘bad’ (and so forth) to such notions
as happiness, flourishing, or survival. For a discussion of related issues, see
Parfit, 1984.
15. I have read an earlier version of this essay to the Department of
Philosophy, University of Rochester; I am very grateful to various members
of the audience for their helpful comments. Also, I have been helped by a
discussion with David Hershenov’s University of Buffalo graduate seminar,
Fall 2004.
6
Why Immortality Is Not So Bad
John Martin Fischer
There’s an old joke. Uh, two elderly women are at a Catskills mountain resort, and one
of ‘em says: ‘Boy the food at this place is really terrible.’ ‘The other one says, ‘Yeah,
I know, and such . . . small portions.’ Well, that’s essentially how I feel about life. Full of
loneliness and misery and suffering and unhappiness, and it’s all over much too quickly.
Woody Allen, Annie Hall
I
I shall begin by laying out some of the key elements of Bernard Williams’s
fascinating and influential discussion of immortality, ‘The Makropulos
Case: Reflections on the Tedium of Immortality’.1 Williams discusses a
character in a play by Karel Čapek (which was made into an opera by
Janaček.) This character had various names with the initials EM. When
she was 42 years of age, her father gave her an elixir of life which rendered
her capable of living forever (at the biological age of 42). At the time of
action of the play, EM is aged 342. As Williams puts it, ‘her unending life
has come to a state of boredom, indifference and coldness. Everything
is joyless . . . In the end, she refuses the elixir and dies, and the formula is
destroyed by a young woman (despite the protests of some older men!).’
For my purposes here, it will be useful to begin by distilling from
Williams’s rich and intriguing discussion his general framework for
analyzing models of immortality. This framework involves positing
two criteria which must be met if a given model of immortality is to
be appealing to an individual. First, the future person (posited by the
model) must be genuinely identical to the individual. (This means not
just being qualitatively similar or having several identical properties; it
means being genuinely identical—the same particular person.) Second,
the life of the future person must be attractive (in a certain way) to the
individual—the life of the future person must be ‘suitably related’ to
the goals and projects of the individual.
This framework is really very simple and natural. It says that, in
order for a model of immortality to be attractive to an individual, the
model must posit a future scenario in which the individual can recog-
nize himself—someone genuinely identical to the individual. Further,
79
the life of oneself in the future must be appealing; presumably, it cannot

involve constant torture, onerous labor, tedium and so forth. The two
conditions presented by Williams can be dubbed the ‘identity condi-
tion’ and the ‘attractiveness condition’.
Now the problems with EM-type immortality are supposed by Williams
to pertain primarily to the second condition, although he also adduces
considerations pertinent to the first.2 With regard to the second condition
Williams constructs a dilemma. Either EM’s character (her basic goals,
projects, dispositions, and interests) remain the same over time, or they
change. If they remain the same, then indefinitely many experiences will
lead to detachment or boredom: ‘a boredom connected with the fact that
everything that could happen and make sense to one particular human
being of 42 had already happened to her’.3 But if the character changes, it
is unclear whether the second condition is satisfied, because it is unclear
how to assess the new projects and goals in light of the old ones.
Williams’s point is that it is not merely a contingent fact that eter-
nal life would be unattractive; this unattractiveness is alleged to be an
essential feature of eternal life.4 Williams says:
. . . perhaps, one day, it will be possible for some of us not to age. If
that were so, would it not follow then that, more life being per se
better than less life, we should have reason so far as that went . . . to
live for ever? EM indeed bears strong, if fictional, witness against
the desirability of that, but perhaps she still laboured under some
contingent limitations, social or psychological. . . . Against this, I am
going to suggest that the supposed contingencies are not really con-
tingencies; that an endless life would be a meaningless one; and that
we could have no reason for living eternally a human life. There is no
desirable or significant property which life would have more of, or
have more unqualifiedly, if we lasted for ever. In some part, we can
apply to life Aristotle’s marvellous remark about Plato’s Form of the
Good: ‘nor will it be any the more good for being eternal: that which
lasts long is no whiter than that which perishes in a day’ [Ethica
Nicomachea 1096b4].5
II
I wish to examine Williams’s thesis that immortality is essentially unap-
pealing for creatures like us. First, I shall briefly consider Williams’s sug-
gestions about the identity condition. Then I shall turn to the attractiveness
condition. Consider the following passage from Williams’s essay:
Some philosophers have pictured an eternal existence as occupied in
something like intense intellectual enquiry. . . . The activity is engross-
ing, self-justifying, affords, as it may appear, endless new perspec-
tives, and by being engrossing enables one to lose oneself. . . . But if
Why Immortality Is Not So Bad 81
one is totally and perpetually absorbed in such an activity, and loses

oneself in it, then as those words suggest, we come back to the prob-
lem of satisfying the condition that it should be me who lives for
ever.6
Similarly, Williams argues against the appeal of the Spinozistic idea that
intellectual activity is the most active and free state that a person could
be in. Specifically, Williams argues against Stuart Hampshire’s formula-
tion of a doctrine he alleges is shared by both Spinoza and Freud, that
one’s only means of achieving this distinctness as an individual, this
freedom in relation to the common order of nature, is the power of
the mind freely to follow in its thought an intellectual order.’ The
contrast to this free intellectual activity is ‘the common condition of
men that their conduct and their judgments of value, their desires
and aversions, are in each individual determined by unconscious
memories.7
But since Williams believes that such unconscious motivations are
indeed part of the self, he accuses the Spinozistic conception of freedom
of aspiring to be free from the self, which entails a loss of individuality
itself. Thus, again, Williams claims that to lose oneself in intellectual
activity is literally to lose oneself. If such activity were the dominant com-
ponent of immortality, it could not be of interest to an individual in the
sense in which the individual is especially interested in his or her own
future; thus, Williams is here primarily concerned with his first criterion
for the desirability of immortality—the identity criterion. Williams goes
on to say:
As those who totally wish to lose themselves in the movement can
consistently only hope that the movement will go on, so the consis-
tent Spinozist—at least on this account of Spinozism—can only hope
that the intellectual activity goes on, something which could be as
well realised in the existence of Aristotle’s prime mover, perhaps, as
in anything to do with Spinoza or any other particular man.8
But it seems to me that an activity in which it is tempting to say that one
‘loses oneself’ is one in which the content of one’s experiences is focused
outward: one is thinking about something besides oneself. An engross-
ing and absorbing activity causes one to ‘lose oneself’ in the sense that
one is not self-absorbed. But it is quite another matter to claim that the
experiences involved in such activities are themselves not one’s own.
Even though one has ‘lost oneself’ in something in the sense that one
is not narcissistically focused even in part on oneself, it does not follow
that one cannot look at a future with such experiences as genuinely
one’s own future.
I would suggest, then, that Williams’s remarks about ‘losing oneself
in the movement’ do not call into question the possibility of an immortal
life in which a certain particular individual continues to exist (and

can envisage him or herself in the future). Even if one’s life is heavily
invested in activities in which one ‘loses oneself’, one can still under-
stand these activities to be part of one’s own future; the crucial distinc-
tion here is between the content of the relevant experiences and their
ownership.
III
I now turn to Williams’s second condition—the attractiveness condi-
tion. As pointed out above, Williams here constructs a dilemma: either
one’s character remains fixed, or it is allowed to change over time. I
shall begin with the first horn of Williams’s dilemma; that is, I shall be
assuming that the individual in question has roughly speaking a fixed
character over time.
The specific problem with the first sort of immortality (in which
character is held fixed) is its putatively inevitable tendency to become
boring and alienating. Williams puts the point as follows:
In general we can ask, what it is about the imaged activities of an
eternal life which would stave off the principle hazard to which EM
succumbed, boredom. The Don Juan in Hell joke, that heaven’s pros-
pects are tedious and the devil has the best tunes, though a tired
fancy in itself, at least serves to show up a real and (I suspect) a pro-
found difficulty, of providing any model of an unending, supposedly
satisfying, state or activity which would not rightly prove boring to
anyone who remained conscious of himself and who had acquired
a character, interests, tastes and impatiences in the course of living,
already, a finite life.9
There are various philosophical defenses of the thesis that immortal-
ity (of the sort under consideration here) would be necessarily boring
and thus would run afoul of the attractiveness condition. I certainly
cannot here fully defend the idea that there are some pictures of such
immortality which are not necessarily unattractive in this (or any
other) way, but I wish to make a gesture in this direction by pointing
to what appear to me to be some salient errors in Williams’s defense
of the thesis that such immortality is necessarily boring.
The first error can be seen to come from (or at least be encour-
aged by) a particular formulation employed by Williams. He says
that the defenders of the desirability of immortality must provide
a ‘model of an unending, supposedly satisfying, state or activity
which would not rightly prove boring to anyone who remained
conscious of himself and who had acquired a character, interests,
tastes and impatiences in the course of living, already, a finite life’.10
The use of the phrase ‘an unending, supposedly satisfying, state or
activity’, is infelicitous insofar as it suggests (but of course does not

strictly speaking entail) that the endless life in question must consist
in a single state or activity. Later, Williams says that the defender of
the desirability of immortality must point to ‘something that makes
boredom unthinkable . . . something that could be guaranteed to be
at every moment utterly absorbing. But if a man has and retains a
character, there is no reason to suppose that there is anything that
could be that’.11 Again, this passage (especially the use of the sin-
gular pronouns ‘something’ and ‘anything’) at least suggests that
the endless life must consist in some single utterly absorbing thing.
Finally, Williams considers an eternal existence occupied in activi-
ties of intense intellectual inquiry. He says that ‘it seems quite unrea-
sonable to suppose that [these activities] would have the fulfilling or
liberating character that they do have for [an individual who actu-
ally engages in such activities], if they were in fact all he could do or
conceive of doing’.12
But why suppose that any one single supposedly absorbing activity
must be pursued at the expense of all others? Why can’t such activities
be part of a package in an immortal life, just as we suppose that they
should be in a mortal life? Certainly, an immortal life could consist in
a certain mix of activities, possibly including friendship, love, family,
intellectual, artistic and athletic activity, sensual delights, and so forth.
We could imagine that any one of these would be boring and alienat-
ing, pursued relentlessly and without some combination of the others.
In general, single-minded and unbalanced pursuit of any single kind
of activity will be unattractive. But of course from the fact that one’s
life will be unending it does not follow that it must be unitary or unbal-
anced. That one’s life is endless clearly does not have the implication
that one must endlessly and single-mindedly pursue some particular
sort of activity.
It might be useful again to consider Williams’s demand for ‘some-
thing that makes boredom unthinkable . . . something that could be
guaranteed to be at every moment utterly absorbing’. His claim is that
‘nothing less will do for eternity’.13 But the justification for this demand
is unclear. Why, in particular, should there be an asymmetry (of the
sort implied by the demand) in the standards for the attractiveness
of a finite life and an infinite life? Surely, we think of certain mortal
lives which involve considerable stretches of boredom and even pain
nevertheless worth living and even very appealing. Given this, why
think that an immortal life with such features would not be on bal-
ance appealing? Why think that because a life is unending, it must be
uniformly pleasing in order to be on balance attractive? The inference
here is not more compelling than the inference noted above from the
unending nature of immortal life to some single unitary activity which
it putatively must contain.
Suppose one says that one finds some activity ‘endlessly fascinat-
ing’. This could mean various different things. First, it could mean that
whenever one turns to the activity (in the normal course of one’s life),
one finds it on balance fascinating. Second, it could mean that when-
ever one turns to the activity (in the normal course of one’s life), one
finds it filled with fascinating moments—perhaps even densely packed
with fascinating moments. Finally, I suppose it could (just possibly)
mean that one pursues the activity forever and finds it at every moment
fascinating. Thus, with regard to the schema, ‘endlessly—’, one must
distinguish at least three different notions: reliability, density, and infinite
extensibility.
Now imagine that an unending life contains some activity which one
finds ‘endlessly fascinating’. It surely does not follow from the fact that
an unending life contains an endlessly fascinating activity that the activ-
ity must be endlessly fascinating in the sense of infinite extensibility.
An unending life can contain an endlessly fascinating activity in the
sense of reliability or density. Further, I see no reason simply to assume
(as Williams seems to) that in order for an endless life to be attractive,
it must contain an activity (or even set of activities) that is endlessly
fascinating (or endlessly appealing in any way) in the sense of infinite
extensibility. I should think that it is even an open question whether in
order for an endless life to be attractive, it must contain an activity that
is endlessly fascinating (or endlessly appealing in any way) in any of
the senses.
I wish now to develop a distinction which I believe is important to
assessing the appeal of immortality. Having laid out the distinction, I
will suggest that the tendency to think that immortality must be bor-
ing and alienating may come in part from attending solely to one of
the categories involved in the distinction; this is another mistake of the
proponents of the thesis that immortality is necessarily boring.
Some pleasurable experiences, it seems, are in some sense ‘self-
exhausting’. In the case of these pleasures, once (or perhaps a few
times) is enough. That is to say, when one experiences such pleasures
one tends not to want to repeat them—even at some point relatively far
in the future. Some such pleasures are frankly disappointing; in the case
of these, we find that some highly touted or much anticipated pleasure
is just not what it was made out to be, and we simply conclude that
it is not worth pursuing these in the future. But there are other such
pleasures which are not necessarily disappointing; rather, they may be
entirely fulfilling but in some way ‘complete in themselves’. More spe-
cifically, they seem to be complete in the sense that, having experienced
such a pleasure, one has no desire to experience it again at any point in
the future.14
I take it that everyone has had his share of disappointments, so it is
not necessary to dwell on these. But it will be useful to consider some
examples of the ‘non-disappointing’ self-exhausting pleasures. Suppose,

for instance, that you have the goal of doing something just (or at least
primarily) to prove to yourself that you can do it. Imagine, for example,
that you are somewhat afraid of heights, and you have been working
hard to overcome this phobia. You form the goal of climbing Mt Whitney
just to show yourself that you have overcome the fear—just to show
yourself that you can control your life and overcome obstacles. Upon
climbing the mountain, you may in fact be very pleased and proud.
Indeed, you may be deeply satisfied. But also you may have absolutely
no desire to climb Mt Whitney (or any other mountain) again. You have
accomplished your goal, but there is no impetus toward repeating the
relevant activity or the pleasure that issues from it.
I speculate that there are quite a few activities and resulting plea-
sures that are relevantly similar to those in the above case. Some of
these are activities in which one sets out to prove something to oneself
or other people. Others may be activities in which one sets a goal which
is essentially ‘comparative’ in some way—one wants to win a race or
some prize, one wants to be the brightest, most productive, most popu-
lar, fastest, and so forth (in some given context). Frequently (although
certainly not invariably), upon reaching such essentially comparative
goals, one finds them either disappointing or ‘complete in themselves’;
in any case, there is relatively little energy or impetus to repeat the
accomplishments. (Of course, the energizing aspect of such accom-
plishments will vary with the nature of the accomplishment and the
individual’s personality; for some individuals, such achievements only
whet the appetite for more, whereas this is not the case for others.)
I suspect, then, that the class of self-exhausting pleasures (both dis-
appointing and not) is rather large. But these are not the only sort of
pleasures. There are also ‘repeatable pleasures’. Here an individual may
well find the pleasure highly fulfilling and completely satisfying at the
moment and yet wish to have more (i.e., to repeat the pleasure) at some
point in the future (not necessarily immediately). Certain salient sen-
sual pleasures leap immediately to mind: the pleasures of sex, of eating
fine meals and drinking fine wines, of listening to beautiful music, of
seeing great art, and so forth. These, or many of them, seem to be—at
least for many people—repeatable pleasures. (Note that the distinction
between self-exhausting and repeatable pleasures must be relativized to
particular individuals; this having been said, there will presumably be
some similarities across different individuals.)
It is not evident that the distinction between self-exhausting and
repeatable pleasures can be understood or explained in terms of other
notions. That is, it is not clear that the repeatable pleasures are ‘higher’,
‘more noble’, ‘more intrinsically compelling’, ‘more complex’, ‘more
intense’, and so forth. It just seems to be a fact about us that we find
that some pleasures are self-exhausting and some are repeatable, and it
is not clear how even to begin to give an illuminating reductive account

of this distinction.15
Of course, even repeatable pleasures may become boring or unap-
pealing if distributed too closely (or in an otherwise inappropriate pat-
tern). I suppose that even the most delectable lobster thermidor would
quickly become revolting if consumed at every meal. But, as noted
above, it is a mistake to suppose that the pleasures must be experi-
enced in this way. Given the appropriate distribution of such pleasures,
it seems that an endless life that included some (but perhaps not only)
repeatable pleasures would not necessarily be boring or unattractive.
Perhaps some of the proponents of the ‘necessary boredom’ thesis tend
to attend solely or primarily to the self-exhausting pleasures (and asso-
ciated activities). But once it is seen that there are also repeatable plea-
sures, the prospects of a certain sort of immortality are not nearly so
grim.
I wish to say a bit more about the distinction between self-exhaust-
ing and repeatable pleasures. As the discussion proceeds, I hope it will
become evident just how implausible it is to deny that there are repeat-
able pleasures (or that there can continue to be repeatable pleasures
that form part of a mix of pleasurable experiences that extends indefi-
nitely into the future). As a help in further discussing the nature and
role of repeatable pleasures, I shall now relate the story of André and
his beloved goose liver:16
We had just been served the usual airline fare. The man sitting next
to me, call him André, tasted his food deliberately, paused thought-
fully for a moment as if he were extracting what little pleasure could
be found in the morsel, and then pronounced judgment: ‘Surprising,
yes this is really rather nice.’ He had a cultured European accent
and the appearance of a man dissipated not by wanton and reckless
living, but by the civilized excess of too much of the good life. I said
something to the effect that I thought all airplane food was awful
and this seemed to be no exception. André looked at me with a type
of patient parental disappointment. My comment had revealed how
little I knew about life. ‘Well, of course, this “food” is terrible— not
really food at all. But this is an airplane, isn’t it? And the point is that
this turkey is much superior to what one normally finds in such envi-
rons. That is the pleasure in it.’ It became clear that André’s senses
were far more refined than mine. He had trained himself to glean
what little enjoyment could be found even in something so bland as
a turkey sandwich on United.
He began to relate the various meals he had eaten at different times.
And this was how we at last came to the topic of the beloved goose
liver. A goose liver, you see, properly nurtured and prepared, simply
is better than the best of any other food. André became quiet for a
time—lost in reveries like one remembering old and dear friends.
He began slowly, reverently to recall for me the rare times when he

had found his beloved goose liver. There were the times growing up
in Hungary—a country which, as everyone knows, really is the best
country at producing goose liver. Later there were great moments
when he would return to Hungary to visit his relatives; they would
scrimp and save in order to have the week’s wages necessary to pro-
cure the goose liver. Certainly this was extravagant, but so great was
his joy eating the meal that everyone at the table felt it was a small
price to pay.
There were other rare occasions in places like Vienna and New
York where André would find and become reacquainted with his
beloved goose liver in new surroundings. But such moments car-
ried with them tremendous opportunities for disappointment. Not
infrequently, the prized liver would be ruined by a clumsy chef who
completely lacked the proper respect for the bounty he was prepar-
ing. Once, however, André was traveling through a little town in the
Swiss Alps. He happened upon an average-looking restaurant around
dinnertime. There on the menu was the daily special—goose liver. He
inquired after the details of the dish— was it fresh, how was it pre-
pared, and so forth. The answers encouraged him to order the meal.
Upon its arrival at his table, André was surprised beyond his wildest
dreams. He exclaimed to the waitress that he must meet the chef, for
there were only two or three men in the world (he knew them all)
who could prepare the beloved goose liver so expertly. How was it
possible that the masterpiece could be produced so casually here?
Much to André’s surprise, when the chef was brought to the table,
he turned out to be one of the famous chefs who had prepared André
a meal years earlier. (The chef had some family business in the area
and was cooking in the restaurant as a favor to the owner who was
his friend.) The chef was, of course, delighted to find someone who
truly appreciated the treasure which had been laid before him, and
the two talked late into the night. André extended his stay in the
town three days. He ordered goose liver every night.
Evidently, André’s enthusiasm is food. Surely, the pleasures of the goose
liver are repeatable pleasures for André. And it seems that André does
not need such exotic culinary adventures to achieve significant repeat-
able pleasures; indeed, he gets such pleasures from a wide variety of
gastronomic experiences, both elaborate and pedestrian. Further, I see
no reason to think that André’s pleasures would cease to be repeatable,
if part of an immortal life (in which the pleasures are appropriately dis-
tributed. Goose liver for breakfast, lunch and dinner would no doubt
rather rapidly turn even André’s stomach).
To extend the point. Really, it seems that there are many repeatable
pleasures; when one thinks about it—and specific accounts such as that
of André help to bring home the point—Williams’s necessary boredom
thesis becomes very implausible. Think, for instance, of the pleasures

of listening to great music. I get extraordinary pleasure from listen-
ing to Bach’s Second Partita for the Unaccompanied Violin. (Whereas
I am certainly not immune to gastronomical delights, Bach’s Second
Partita is my beloved goose liver.) And I see no reason why it would
cease to be a repeatable pleasure, if part of an immortal life (in which
there were an appropriate mix of activities and pleasures). Certainly,
there are other such pleasures, such as the pleasures of visiting a great
art museum, or a great and beautiful city, such as Paris, Venice or San
Francisco. (I cannot imagine ever getting tired of the view of the city of
San Francisco from the Golden Gate Bridge, or the feeling of the fog
engulfing me in Golden Gate Park, or the beautiful plaintive sound
of the foghorns in the distance. I have no tendency to think that these
pleasures would become less compelling, unless pursued in a single-
minded or compulsive fashion.)
In this section I have in a very sketchy way suggested a distinction
between self-exhausting and repeatable pleasures. Although I have not
analyzed or developed the distinction in detail, I have suggested that it is
a mistake to suppose that all pleasures are relevantly similar to the self-
exhausting sort. I wish briefly here to allude to a treatment of these issues
which (like Williams’s) is insufficiently attentive to the distinction in
question. In Kierkegaard’s pseudonymous essay ‘The Rotation Method’,
the aestheticist ‘A’ properly rejects the idea that there must be one activity
which is the sole source of pleasure and which is pursued relentlessly
over the course of a lifetime. Rather, ‘A’ endorses a system of rotating
pleasures just as an efficient farmer might rotate his crops to achieve a
better result. But even with the rotation method ‘A’ finds life boring:
Starting from a principle is affirmed by people of experience to be
a very reasonable procedure; I am willing to humor them, and so
begin with the principle that all men are bores. Surely no one will
prove himself so great a bore as to contradict me in this.
. . . All men are bores. The word itself suggests the possibility of
a subdivision. It may just as well indicate a man who bores others
as one who bores himself. Those who bore others are the mob, the
crowd, the infinite multitude of men in general. Those who bore
themselves are the elect, the aristocracy; and it is a curious fact that
those who do not bore themselves usually bore others, while those
who bore themselves entertain others.17
But whereas Kierkegaard’s hedonist ‘A’ avoids some of the errors dis-
cussed above by adopting the rotation method, he evidently does not
avoid the error of ignoring or underestimating the repeatable pleasures.
Given the existence of such pleasures, a life with a suitable arrangement
of them need not be boring. And I do not see why an immortal life with
such a mix of repeatable pleasures would necessarily be boring.
Kierkegaard wished to convince us to turn away from hedonism and

toward spiritual and religious experiences. I have suggested that he
ignored the possibility of a range of pleasures which clearly are accessi-
ble even to persons who do not have spiritual or religious experiences.
But for those who do indeed have such experiences, there would seem
to be even more reason to embrace immortal life; surely, the deep and
resonant rewards of spiritual and religious experience would not some-
how become wooden or etiolated, if part of an endless life. What reason
is there to suppose that such experiences would change their character
in such circumstances?
Williams usefully distinguishes between ‘conditional’ and ‘categori-
cal’ desires.18 The conditional desires are desires for certain things,
given that one will continue to live. Someone surely will want adequate
clothing, food, shelter, and so forth, on the condition that he or she will
continue to be alive. But such a person may not prefer to continue to
live. Preferences which imply an answer to the question of whether
one wishes to be alive are categorical desires. Presumably— although
Williams does not explicitly say this—there can be both ‘positive’ and
‘negative’ categorical desires. A positive categorical desire implies the
desire to continue to live, whereas a negative categorical desire implies
the desire not to continue to live.
Perhaps the distinction between self-exhausting and repeatable plea-
sures can go some distance toward illuminating Williams’s claim that
one would lose one’s positive categorical desires in an immortal life.
Granted, this might be true if one focused exclusively on self-exhaust-
ing pleasures. After a while—perhaps a long while—these desires
would lose their capacity to ground categorical desires and to propel
one into the future. But I see no reason to think that the repeatable plea-
sures would lose their energizing and ‘propulsive’ character. Further,
spiritual and religious experiences would seem to be relevantly similar
to the repeatable pleasures in this respect; they seem capable of provid-
ing the basis for positive categorical desires, even in an immortal life.19
So far I have been concerned to discuss the first horn of Williams’s
dilemma pertinent to the attractiveness condition (presented above).
That is, I have discussed the necessary boredom thesis in the context of
a relatively fixed character. Let me now say just a few very brief words
about the second horn, according to which the relevant individual’s
character changes over time. Williams suggests that it is now unclear
that the individual will find such immortality attractive, given that it is
unclear that there is the appropriate relationship between the individu-
al’s current character and future goals, values, and interests.
This sort of case notoriously raises fascinating but complex issues.20
But the basic point is that it seems that an individual could value such
an existence if he or she felt that the change in character would result
from certain sorts of sequences. That is, if I felt that my future character
will be different from my present one as a result of appropriate reflec-

tion at future times upon my experiences given my ‘then-current’ char-
acter, then I might well value such an existence. One’s attitudes toward
future changes of character depend on how and why the changes take
place.
Surely in our ordinary, finite lives we envisage certain changes in
our values and preferences over time. For example, one may currently
value excitement and challenge; thus, one might wish to live in an
urban area with many career and avocational opportunities (but with
lousy weather and a high crime rate). Still, one might envisage a time in
the future when one will be older and will prefer warm weather, seren-
ity, and security. One can certainly envisage a time when one will prefer
to live in a condominium in a warm, safe place, even if one currently
thrives on life in Manhattan. And one need not look at the future stages
of one’s life (in which significant changes in values and preferences
have taken place) as unattractive; certainly, they are not so unattractive
as to render death preferable!
Thus, there are quite ordinary cases in our finite lives in which we
envisage changes in our characters—our values and preferences—and
which are not so unattractive as to render death preferable. Why, then,
could not the same be true of immortal existence? As above, why set
such radically different standards for immortal life and mortal life?
Granted, if one’s character is changed by brainwashing, coercion,
deception or various other methods, one might find the resultant
existence thoroughly unattractive. But why assimilate all changes of
character to these? And a devoted conservative republican may find
it unthinkable that she become a liberal democrat, even by rather less
exotic means of transformation. But it is not evident to me that such a
person would actually prefer death. An even so, there is no reason to
assimilate all changes of character to such a change; all that is required,
in order to defend the thesis that immortality is not necessarily unat-
tractive (on this horn of the dilemma), is that there be certain changes of
character plausibly envisaged as part of an immortal life which would
not be so unattractive as to render death preferable.
IV
In this essay I have explored some of the philosophical puzzles pertain-
ing to immortality. More specifically, I have used Bernard Williams’s
important and influential discussion as a springboard for analyzing
what I take to be certain problems with the claim that immortality
is necessarily unattractive. I have argued that it is unfair to suppose
that, in order for immortality to be attractive, it must consist of some
single activity pursued at the expense of others. Further, it is unfair to
demand that, in order for immortality to be attractive, it must consist
of entirely pleasurable or agreeable experiences; why suppose the

standards for immortal life are in this respect different from the stan-
dards for mortal life? Also, one may be entirely ‘lost’ in an engross-
ing activity in the sense of not focusing (primarily) upon oneself; it
is quite another matter to say that the relevant experiences are not
one’s own. Finally, it is important to distinguish two different kinds of
pleasures: self-exhausting pleasures and repeatable pleasures. A life
without repeatable pleasures might well eventually become boring.
But it is a mistake to suppose that an immortal life must contain only
self-exhausting pleasures at the expense of repeatable pleasures. The
repeatable pleasures—perhaps together with spiritual and religious
experiences—could provide a reasonable basis for positive categorical
desires even in an immortal life. It has been a recurrent theme of my
discussion that it is quite unfair to set radically different standards for
finite life and immortal life.
NOTES
I am very grateful to careful and insightful comments by Mark Ravizza.

Also, I have benefited from the comments of anonymous readers for the
International Journal of Philosophical Studies. Finally, some of the material in
this essay is based on ideas that also appear (in telescoped form) in the intro-
ductory essay in The Metaphysics of Death and ‘Models of Immortality’ (see
notes 1 and 2).
1. Bernard Williams, ‘The Makropulos Case: Reflections on the Tedium of
Immortality’, in Bernard Williams, Problems of the Self (Cambridge University
Press, 1973), pp. 82–100; reprinted in John Martin Fischer (ed.) The Metaphysics
of Death (Stanford, CA: Stanford University Press, 1993).
2. For a general taxonomy of models of immortality and a discussion of the
bearing of Williams’s two criteria on several of these models, see the appendix
to this chapter.
3. Williams, op. cit. note 1, p. 90.
4. Presumably, the essential boredom thesis is meant to apply to creatures
of a certain sort—creatures relevantly similar to us. Otherwise, it would follow
from the thesis that God’s existence is boring and unattractive (insofar as God
is essentially everlasting).
5. Williams, op. cit. note 1, p. 89.
6. Ibid., p. 96.
7. Ibid. p. 97.
8. Ibid., p. 98.
9. Ibid., pp. 94–5.
10. Ibid.
11. Ibid.
12. Ibid.
13. Ibid.
14. This notion of ‘completeness in itself’ is different from Aristotle’s notion
according to which certain activities—energeiai—are complete in themselves.
Aristotle distinguishes energeia from kinesis, which are not complete in them-
selves. Roughly, Aristotle’s distinction corresponds to activities which are
movements toward a certain product and which are not complete until the pro-
duction of the product, and activities which are not so understood.
At Metaphysics Theta Six, Aristotle introduces the ‘tense test’ to distinguish
energeia and kinesis. According to the tense test, if the verb ‘X-ing’ is an energeia
verb, then ‘I am X-ing’ entails ‘I have X-ed’. For example, ‘I am enjoying myself’
entails ‘I have enjoyed myself’. If the verb is a kinesis verb, ‘I am X-ing’ entails
‘I have not X-ed’. For example, ‘I am learning [something]’ entails ‘I have not
learned [the thing]’. There is an analogue of the tense test which is a non-
linguistic phenomenon. The proper parts of energeia X are also X’s: the proper
parts of enjoyings are enjoyings. The proper parts of kinesis Y are not also Y’s:
the proper parts of a walking from A to B are are not walkings from A to B.
For some discussions of the tense test, see: J.L. Ackrill, ‘Aristotle’s Distinction
Between energeia and kinesis’, in R. Bambrough (ed.) New Essays in Plato and
Aristotle (New York: Humanities Press, 1965); and Terry Penner, ‘Verbs and the
Identity of Actions—A Philosophical Exercise in the Interpretation of Aristotle’,
in O.P. Wood and G. Pitcher (eds) Ryle: A Collection of Critical Essays (Garden
City, NY: Doubleday, 1970).
15. It is an interesting philosophical question: Why are some pleasures self-
exhausting and others repeatable?
16. For the story of André I am indebted to Mark Ravizza. Since the original
publication of this essay, I have become aware of the cruel practices involved
in producing goose liver; I would not have used Mark Ravizza’s otherwise nice
(and true) story, if I had known of these practices.
17. Soren Kierkegaard, ‘The Rotation Method’, in Either/Or, in A Kierkegaard
Anthology, ed. Robert Bretall (New York: The Modern Library, 1946), pp. 21,
23–4.
18. Williams, op. cit. note 1, pp. 85–6.
19. It has been brought to my attention that there may indeed be some
experiences in life that we savor and value (to the extent we actually do)
precisely because we know that we will not enjoy them forever. It is difficult
for me to know whether this is really the case, and to what extent (if so). But
let me grant that it is true. This admission would not in itself undermine
my strategy of argumentation, for even if certain pleasures are expunged or
diminished, the repeatable ones may still make immortal life worthwhile.
And it is also worth noting that there certainly are painful and unpleasant
experiences associated precisely with the fact that we cannot have certain
relationships and experiences forever: loss and death notoriously impose
great pain and suffering upon us. I see no reason to suppose that the dim-
inution in pleasures issuing from immortality would be greater than the
diminution in pain and suffering.
20. See, for example, Derek Parfit, Reasons and Persons (Oxford: Oxford
University Press, 1984).
Appendix to Chapter 6: Philosophical
Models of Immortality in Science Fiction
John Martin Fischer and Ruth Curl
Science fiction (SF) is often described as a literary genre well suited to

philosophical speculation. SF and philosophy share a common interest
in the question of immortality, and comparisons and contrasts can be
made regarding their respective treatments of the theme. We propose
here a sketchy taxonomy of different models or pictures of immortality
offered by philosophers and SF writers. After noting important differ-
ences in these models, we shall suggest that some problems and con-
cerns expressed by philosophers and SF writers alike are the result of
conflating different models. It is our hope that these comparisons will
provide a preliminary sense of the way SF can be said to function as
philosophical discourse.
Our discussion will use as its base the analytical framework pre-
sented in Bernard Williams’s influential discussion of immortality, The
Makropulos Case: Reflections on the Tedium of Immortality.1 This simple
and natural framework involves two criteria to make immortality
truly appealing: first, there must be a future in which an individual
can recognize himself or herself—someone genuinely identical to the
individual, not just qualitatively similar or with several identical prop-
erties. Second, the future life of the individual must be appealing (in
some way) to that individual; it cannot involve constant torture, hard
labor, tedium, or the like. These conditions can be dubbed the identity
condition and the attractiveness condition. With these, we can construct a
taxonomy of different models of immortality (see Table 6.1).
Although our focus will be the immortality of sentient creatures
or constructs, another treatment of immortality in SF is also possible:
universe immortality, in which there is an attempt to overcome laws
of entropy to create an immortal world, forever self-perpetuating. The
center of attention here is not the immortality of sentient creatures but
rather the immortality of the physical universe.2 Only SF seems to deal
with universe immortality; and while it is not our focus, this vision of
immortality merits attention because it has no corollary in other fields
of literature or philosophy.
93
Table 6.1. A taxonomy of immortality

Immortality of the universe Immortality of sentient creatures
Nonatomistic Atomistic
Stream-of-consciousness Stream-of-consciousness
retaining loss
Serial Nonserial
Disjoint lives Connected lives
(Methods) (Forms)
Biological Nonbiological Solipsistic Nonsolipsistic
Vampirism “Downloading”
mind into
Chemicals computer
Cryonics or Cyborgs
Suspended animation
Relativistic
Cloning time dilation
Body transfers
We turn now to depictions of immortality pertaining to sentient enti-

ties, beginning with a distinction between nonatomistic and atomistic
concepts of immortality. The former involves a kind of fusion of differ-
ent individuals into a type of immortal entity; the latter involves the
immortality of individuals. The nonatomistic model usually involves
the merging of various individuals into some sort of superorganism.
The individual’s stream of consciousness may either be retained, as in
Greg Bear’s Blood Music, or lost, as in Arthur C. Clarke’s Childhood’s End
and one episode of Robert A. Heinlein’s Methuselah’s Children.3
In Blood Music, a brilliant researcher, after losing his job, injects him-
self with lymphocytes he has genetically manipulated so that he can
smuggle them out of the lab and continue his research. The altered
Philosophical Models of Immortality in Science Fiction 95
lymphocytes then invade the biosphere and trigger the mutation of

humanity into a new organism composed of individually intelligent
cells. Eventually the cells unite to form a superintelligent being. Each
cell can either function separately or compartmentalize with other cells,
which can then isolate themselves to work on various problems. Bear’s
vision of the mutation and transformation of humanity is best expressed
in the novel’s last lines: “Nothing is lost. Nothing is forgotten. It was in
the blood, the flesh. And now it is forever” (BM 247).
In Childhood’s End, children and adolescents transform beyond the
comprehension of the rest of humanity. Clarke’s vision clearly shows a
complete but unintentional and uncontrollable break with human char-
acteristics, memories, and emotions. In Methuselah’s Children, members
of the Howard Families (immortals in a mortal world) fleeing persecu-
tion on Earth and searching for a hospitable planet encounter the Little
People, who “in an utterly basic sense . . . differed from humans in kind.
They were not individuals. No single body of a native housed a discrete
individual. Their individuals were multi-bodied, they had group ‘souls.’
The basic unit of their society was a telepathic rapport group of many
parts. The number of bodies and brains housing one individual ran as
high as ninety or more and was never less than thirty-odd” (MC 134–5).
Clearly, there can be different versions of the nonatomistic approach,
including differences in the nature of the transition from individuals to
composites, which can be a genetic mutation (as in Blood Music, ignor-
ing for the moment the manipulation by Vergil) or a nonmutational
evolutionary transformation (as in Childhood’s End). And there can be dif-
ferences in the nature of the composites: for example, there may be one
or many composites, and the existence of the composites might be rela-
tively desirable or undesirable.
But any sort of nonatomistic immortality—even one in which the
nature of the composite’s existence is relatively attractive—appears to
run afoul of Williams’s first criterion: the identity criterion. Arguably,
the types of fusion envisaged in nonatomistic models (even ones that
somehow preserve individual streams of consciousness) do not allow
individuals to look forward to their own future existence. As such, these
nonatomistic models are not very appealing models of immortality.
Is it then appropriate to eliminate nonatomistic models? When we
look to the future we seem to care about the welfare of our communi-
ties and friends, the planet and its nations. We might care about the
continued development of the arts, the preservation of natural beauty,
and the attainment of human rights and distributive justice; but we care
especially about how we ourselves will fare—we especially look forward
to future pleasurable states of ourselves and particularly reject prospec-
tive unpleasant future states of ourselves. So, for example, if we are told
that some future individuals will be tortured horribly for days, we can
genuinely regret this; however, if we are then told that those people
will be ourselves, we are horrified—we especially regret this. Thus, indi-

viduals might care to some extent about a future in which individuals
have become group entities of certain sorts; indeed, it might even be
desirable in some sense. But we do not and cannot look at such a pros-
pect in the special and especially vivid way we look at future scenarios
in which we exist as individuals. This special sense in which we care
particularly about what happens to us is not engaged by nonatomistic
models of immortality.
Since nonatomistic models seem to run afoul of the identity condi-
tion, let us instead turn to atomistic models of immortality. In this class
there are serial and nonserial models. In serial models of immortality, the
individual in question in some ways lives a series of lives; in nonserial
models, the individual simply leads an indefinitely long single life.
The atomistic serial model of immortality comes in at least two ver-
sions: the disjoint-lives serial model and the connected-lives serial model. In
the disjoint-lives model, one individual lives an indefinitely long series of
lives without internal psychological connections: there are no significant
continuities or connections of memory or other psychological states, such
as values, beliefs, desires, and intentions, from one life to the next. In this
view, the self is some sort of soul or bare particular without any essential
mental contents. When the soul enters a new body, the person itself per-
sists, even if there are no remaining memories, beliefs, preferences, val-
ues, or intentions. This model recalls the Hindu model of reincarnation. A
possible metaphor is the tulip bulb—the different lives correspond to the
different plants and flowers that spring from the bulb from one year to
the next, whereas the persisting self corresponds to the essential bulb.
But, like the nonatomistic model, the disjoint-lives serial model runs
afoul of the identity condition. It is unclear how an individual could
recognize a future individual as genuinely identical to himself or herself
if there is no psychological connection between the two (including con-
nections of memory). We do not know if it is metaphysically coherent to
suppose that persistence of personal identity means the persistence of a
bare, psychologically empty soul; there are deep perplexities here into
which we cannot go. Even if the model is metaphysically coherent, the
identity condition does not seem to be satisfied in the relevant way, a
way that makes it possible for us to recognize ourselves in the future sce-
nario. That is, even if there is no insuperable ontological problem with
the disjoint-lives serial picture, there is an epistemic problem: presented
with a description of a future scenario, there is no way individuals can
recognize or identify themselves. And, if the relevant future person has
no psychological connection to the current individual, why should the
individual care especially (in the way one cares especially about oneself)
about this future person? Given this problem, the disjoint-lives model
is unappealing—it cannot capture the sense in which we might value
especially our own immortality.
Unfortunately, the connected-lives serial model fares no bet-

ter. Imagine, if you can, what it would be like to lead one life—to go
through childhood, adolescence, and all the stages of life—accumulate
memories and associated values, and then begin again: go through a
second childhood (but with memories of the previous life), a second
adolescence (but with memories of the previous life plus the new child-
hood), and so forth. What would it be like to be a small child carry-
ing memories of adolescence, marriage, raising a family, seeing one’s
children grow up, and so forth? The model of full or robust psycho-
logical connections within serial lives seems either entirely incoherent
or entirely unattractive; in any case, it surely does not meet the two
criteria. Nor does it seem possible to weaken the psychological connec-
tions in any natural or appealing way; in particular, it does not seem
plausible that one could have only certain memories (just enough to
be able to recognize oneself as a persisting entity) at certain stages of
life. This would involve “blackouts” of parts of memory at some stages
and not others—an almost stroboscopic and bizarre picture of memory.
SF often puts a skeptical valence on this sort of connected-lives serial
model. Consider Clarke’s The City and the Stars (1956). In this novel
people are reborn into new bodies without memories of previous lives;
then, as they near adulthood, they gradually remember their previous
lives. From the perspective of the “new” individual, it would surely be
disconcerting to be suddenly flooded by a vast set of old memories of
earlier lives; and—more relevant to the issue of immortality—from the
perspective of the “old” individual, it would not be pleasant to stop
being conscious one day, then reawaken with a new set of memories
involving a new childhood and adolescence.
Thus, even though SF novels may claim that a character leads many
lives, our ruminations above lead us to call this possibility into question.
On closer scrutiny, these novels do not depict characters who them-
selves lead different lives in the senses required by the serial model.
A particular character (say, Lazarus Long) does not lead many lives;
rather, he leads one extended life in which many other people play roles.
Many lives become part of his life when they intersect it—but Long
himself does not genuinely lead a series of lives. (A life with a series of
people need not be a series of lives.)
Our taxonomic trail leads finally to atomistic nonserial conceptions
of immortality, of which there are various versions. Primarily, these
involve different ways of generating or maintaining the indefinitely long
life, and different ways of viewing the nature of the life (pattern and
distribution of experiences, relationship to other lives, and so forth).
Let us first consider the different ways of generating or maintaining
a nonserial atomistic form of immortality. There are a number of horror
stories in which a vampire draws on the life force of another to con-
tinue existence. Literature and films have produced many variations
on this theme; the vampire does not always follow the Count Dracula
formula. Numerous films, for example, depict beautiful young women
who seduce young men to feed off their energy; and Oscar Wilde’s The
Picture of Dorian Gray (1897) features a protagonist who remains young
while his portrait ages. In E. E. “Doc” Smith’s Lensman series the
Overlords live off the life force of the Velantians. In one Balzac story, an
old man lives off young girls. A recent Stephen King film, Sleepwalkers
(1992), depicts a young man who gains vitality from the innocence or
purity of young virgins, whom he kills in order to devour their life
force, or souls. Obviously, the topic continues to generate discussion.
In Anne McCaffrey’s series comprising Crystal Singer (1982),
Killashandra (1985), and Crystal Line (1992), humans have developed
a symbiosis with a spore, which makes them extremely long-lived.
Unfortunately, they must periodically return to the planet of the spores
to avoid a terrible death. (This somehow resembles the need to visit
one’s parents regularly—at least in our families!) In McCaffrey and
Jody Lynn Nye’s The Death of Sleep (1990), the protagonist, who engages
in cryogenic sleep, ages only four or five years in seventy-two. Another
type of life prolongation is envisioned in Hugo Gernsback’s Ralph 124C
41+: A Romance of the Year 2660 (1925), in which a man reacts to a sci-
entist’s revival of a dead dog by exclaiming, “I only regret for myself
that you had not lived and conducted this experiment when I was a
young man, that I might have, from time to time, lived in suspended
animation from century to century, and from generation to generation
as it will now be possible for human beings to do.”4 This would not
be continued conscious existence with stroboscopic memory, but rather
stroboscopic consciousness of a certain sort. In some novels, cloning
gives characters a form of immortality. For example, in Heinlein’s Time
Enough for Love (1973), Lazarus Long is more or less cloned as his own
daughters.
Another biological method of achieving immortality consists in so-
called body transfers, which presupposes the falsity of the “bodily
identity” criterion of personal identity. In The World of Null-A (1948) and
The Players of Null-A (1956), by A. E. van Vogt, Gosseyn’s consciousness
transfers from one body (when it is destroyed) to another. As long as
there are bodies, he can exist forever. Of course, conceptually one can
distinguish between various sorts of body transfers. In some cases the
brain is transferred to a different body. In others the brain itself is not
transferred, but the mental state is, as in the film Invasion of the Body
Snatchers (1956); in some of the latter sorts of cases, there can be telepor-
tation as well as mental transfer.5
There are also rather less exotic (though by no means mundane)
biological methods of generating and maintaining immortality, as
portrayed in Mary Shelley’s “The Mortal Immortal” (1910) and Larry
Niven’s Ringworld (1970), Ringworld Engineers (1979), and Protector
(1973), all of which feature immortal beings. In Shelley’s story, a young

apprentice drinks the creation of his master and becomes immortal. In
Niven’s books, individuals live for centuries by using a drug especially
tailored for their chemistry; without the drug, they die. Any human
who eats the Spice of Life becomes a Pak Protector. Human Protectors
undergo a physical change that makes them almost unrecognizable as
human and a mental change that makes them protect whatever society
they are in at the moment. These beings seem to be biological analogues
to Isaac Asimov’s robots, and they follow laws (instincts, in this case)
similar to his Laws of Robotics.
There are also nonbiological methods of generating and maintaining
immortality. In Neuromancer (1984) William Gibson creates a kind of
human immortality by allowing the transfer of human mental states to
computers. In Gregory Benford’s Great Sky River (1987) the transfer is
accomplished through the insertion of computer chips into the human,
resulting in a combination of biological and mechanical capabilities:
though the body may die, the “mind” continues. These procedures
involve mental transfer (“downloading” of the “mind”) not accompa-
nied by actual brain transfer.6
Other SF authors increase human longevity or create immortality
by augmenting or supplanting normal human biological capacities
through mechanical means. In McCaffrey’s The Ship Who Sang (1969), a
future society trains deformed but mentally functional babies to work
in cyborg-type bodies if the parents so choose. This falls under the
rubric of cyborg-type models of generating and maintaining atomis-
tic nonserial immortality. In other works, robots are created and then
allegedly made sentient. Their mechanical nature makes them more or
less immortal. Thus, in Asimov’s Robot and Foundation series, Daneel
first acquires a feel for human phenomena and then leans more and
more toward the human, becoming telepathic to get a better insight
into human nature and reasoning. Finally, he makes plans to transfer
his knowledge and memories (which in a robot are also his essence)
to the brain of a Solarian child, thereby becoming mortal. But it is a
prolonged mortality because Solarians, like all Spacers, live three or
four hundred years. Further, if he can perform the operation once, he
can do it again, particularly because this child is a hermaphrodite who
will produce at least one offspring that is for all intents and purposes
“itself.” So, unlike Andrew Martin in “The Bicentennial Man,” Daneel
leaves his “option” of immortality open.
One other nonbiological way to produce immortality (one might call
it “relativistic” immortality) involves time travel, as in Joe Haldeman’s
The Forever War (1974), in which time travel paradoxes are manipulated
to achieve a sort of immortality.7
Having briefly surveyed the methods of generating and maintaining
immortality (in particular, atomistic nonserial immortality), we now
turn to the nature of immortal lives—their relationship to other lives

and the pattern and distribution of their experiences.
First, consider a kind of solipsistic model. Heinlein’s “ ‘—All You
Zombies—’ ” (1958) features an endless temporal loop in which the pro-
tagonist is a man who travels in time. But the pattern of his time travels
indicates that he is in fact his own father, mother, and baby. There are
other, nonsolipsistic, conceptions of the nature of atomistic nonserial
immortality. One posits the “lone immortal” who lives among other
individuals, all mortals. There are at least two versions of the lone
immortal model—one in which the lone immortal is known by (cer-
tain) others to be immortal, and another in which the immortality is a
secret. Such models appear, respectively, in Asimov’s The End of Eternity
(1955) and Shelley’s “The Mortal Immortal.” In another conception the
immortal is not alone—perhaps others are immortal, as in Methuselah’s
Children, or perhaps everyone is immortal.
We have found problems with all models of immortality except for
atomistic nonserial approaches. Nonatomistic models do not seem to
meet the identity criterion, and some atomistic models—in particular
the connected-lives serial models—run afoul of the attractiveness cri-
terion (if not also the identity condition). But what about the atomistic
nonserial models of immortality? Surely some methods of generat-
ing and maintaining immortality (such as feeding off the blood and
vitality of others) make the resulting immortality less attractive. And
some pictures of the nature of such immortality (such as the solipsistic
model) make immortality unappealing. But not all methods of generat-
ing immortality are similarly problematic, and not all concepts of the
nature of such immortality are straightforwardly problematic. Is there
something about the nature of atomistic non-serial immortality that
renders it, on reflection, necessarily undesirable?
Though some philosophers argue for this undesirability,8 some SF
models are not so pessimistic. A common trait in SF is its faith in the
ability of technology to accelerate the moment in the process of his-
tory when desirable immortality can be experienced. And today, there
is already the hope that the human life span can be extended (through
cryonics, for example) long enough to allow us to outlive the immediate
causes of death and in a sense live to see the dawn of immortality. Yet
SF has negative models, too, and can be every bit as critical of positive
aspirations as are many philosophers. One brief example: though some
SF novels depict efforts to achieve immortality through transformation
into robots or mechanical beings, perhaps an equal number offer the
opposite maneuver: a reverse immortality, or “Pinocchio Syndrome,” in
which an immortal strives to become mortal (not to die, but to become
“subject to mortality”). Somehow, even facing the prospect of immortal
existence, human (mortal) qualities still retain such value that they are
worth the reversal.
Despite a certain symbiosis between models in SF and philosophy,

SF may be more open to the possibility of transformation of the human
body and life span. But in the end, is SF any more willing to abandon
human limits? That vast and intriguing question is, unfortunately,
beyond the scope of this essay.9
NOTES
1. Bernard Williams, The Makropulos Case: Reflections of the Tedium of Immor-
tality (Cambridge: Cambridge University Press, 1973), 82–100.
2. We see such concern expressed in Robert A. Heinlein’s “Waldo” (1942)
and Isaac Asimov’s The Gods Themselves (1972), in which an alternate universe
is discovered and energy is drawn from it, thus invalidating the law of the
conservation of energy and avoiding entropy. In Poul Anderson’s Tau Zero
(1970) the universe contracts until there is too much energy contained in too
small a volume and the contracting universe explodes to begin the process of
expansion again. In Gregory Benford’s Timescape (1980), tachyons—particles
that travel faster than light—can make universal wave functions split into two
or more universes if a causal paradox is created by the tachyonic interaction.
These works all depict science fiction’s underlying concern with the mortality
of the universe.
3. Greg Bear, Blood Music (New York: Ace Books, 1986) [BM]; Arthur
C. Clarke, Childhood’s End (New York: Ballantine Books, 1980); Robert A. Heinlein,
Methuselah’s Children (New York: Signet Books, 1958) [MC]. Later page references
are to these editions.
4. Hugo Gernsback, Ralph 124C 41+: A Romance of the Year 2660 (New York:
Frederick Fell, 1950), 65.
5. For various examples of this, along with an incisive and comprehensive
philosophical discussion of the nature of personal identity, see Derek Parfit,
Reasons and Persons (Oxford: Clarendon Press, 1984).
6. See Rudy Rucker, Software (New York: Ace Books, 1982).
7. For a philosophical discussion of time travel paradoxes, see Paul Horwich,
Asymmetries in Time (Cambridge: MIT Press, 1987).
8. See Williams, The Makropulos Case.
9. For more complete citations and a bibliography of Science Fiction discus-
sions of death and immortality, see John Martin Fischer, ed., The Metaphysics of
Death (Stanford: Stanford University Press, 1993), pp. 411–413.
7
Epicureanism About Death

and Immortality
John Martin Fischer
I want to live forever: but just what is it that I thereby want? Prior to 1874
(or thereabouts) my want would have seemed quite clear: I would have wanted to live for
an unending sequence of years, one year for each natural number—an omega-sequence
of years. But our horizon has since been expanded by the teachings of Georg Cantor.
The natural numbers all together amount only to the smallest order of infinity, aleph-null.
There are countess greater infinities that dwarf aleph-null as surely as aleph-null dwarfs
our customarily allotted three score and ten. Why settle for a piddling aleph-null years
if there are limit cardinals out there to vault over, inaccessible cardinals waiting
to be surpassed?
. . . trans-omega longevity is (conceptually) possible: there are possible worlds that endure
beyond a single omega-sequence of years, and a person can survive in these worlds from
one omega-sequence to another.
. . . I want trans-omega longevity, but not at any cost. Wanting to live beyond a single
omega-sequence of years is, for me, a conditional want, as is wanting to live to be 100.
Both wants are conditional, at the very least, upon my still having my wits about me,
and upon there still being a fair balance of pleasure over pain. In claiming that
trans-omega longevity is desirable, I claim only that there is some possible world,
even if quite remote from our own, in which I have trans-omega existence and the above
conditions are satisfied. Some, it is true, have argued that such conditions could never
be satisfied even for ordinary immortality because a life too long inevitably leads to
perpetual boredom. I suspect that those who argue in this way either lack imagination
or become too quickly jaded with the good things in life . . .
Phillip Bricker, “On Living Forever” (presented at The American Philosophical
Association, March, 1985)
I. INTRODUCTION
Epicureans take seriously Boethius’ thought that philosophy has its
consolations. In her important work on Hellenistic philosophy, Martha
Nussbaum has offered an interpretation of Hellenistic philosophy
according to the “medical model.”1 On this approach, philosophy
is not a neutral, detached methodology, but a way of helping us to
grapple with problems that otherwise would confuse and distress us.
Philosophy, then, is a kind of therapy. Nussbaum both attributes this
view to Epicurus and his followers (such as the Roman philosopher,
Lucretius) and also endorses it. The Hellenistic philosophers sought to
103
apply their philosophical therapy to such issues as the fear of death,

the nature and mysteries of love, sexuality, and potentially unruly emo-
tions, such as anger.
Here I shall focus on Nussbaum’s reconstruction, interpretation,
and defense of Lucretius’s “main argument” that it is irrational to fear
death.2 I shall also offer some reflections on what she calls the “banquet
argument” of Lucretius. According to this argument, we should real-
ize that life is like a banquet: “it has a structure in time that reaches a
natural and appropriate termination.”3 Here I wish briefly to add to my
previous defense of the thesis that immortality would not necessarily
be unattractive.4
In my view, philosophy is a perfectly neutral device. It can be
employed by those who seek reassurance and freedom from anxiety.
It can also increase confusion and perplexity. Even at its best, it may
reveal puzzles and problems of which we were previously unaware.
Of course, it is always up to us how exactly we use the deliverances of
theoretical reasoning. It is not a good idea to ruminate excessively on
insoluble dilemmas—a sensible view that would be endorsed, presum-
ably, by practical philosophy—or to allow them to dampen our spirits.
But it may be that philosophy shows us, what we feared inchoately,
that, as Thomas Nagel puts it, “ . . . a bad end is in store for us all.”5
II. THE MAIN ARGUMENT AND PREVIOUS DISCUSSION

Nussbaum presents Lucretius’s main argument as follows:
1. An event can be good or bad for someone only if, at the time
when the event is present, that person exists as a subject of at
least possible experience, so that it is at least possible that the
person experiences the event.
2. The time after a person dies is a time at which that person does
not exist as a subject of possible experience.
3. Hence, the condition of being dead is not bad for that person.
4. It is irrational to fear a future event unless that event, when it
comes, will be bad for one.
5. It is irrational to fear death.6
Nussbaum points out that Nagel has rejected the first premise of the
main argument because of its insistence on a connection between bad-
ness and experience. Nagel offers two examples. The first involves an
individual who is betrayed behind his back; even though the individual
never comes to know about this betrayal (or, let us say, experience any
unpleasant consequences of it), Nagel contends that the betrayal can be
a bad thing for the individual. In the second example, a person loses all
higher mental functioning in an accident (or as a result of a stroke); this
is alleged by Nagel to be a loss for the person, even if the individual is
now (after the accident) contented. On Nagel’s view, death is bad for
Epicureanism About Death and Immortality 105
the individual who dies not in virtue of involving unpleasant experi-

ences, but insofar as it is a deprivation of the good things in life (the
“deprivation thesis about death’s badness”).
Nussbaum disagrees:
. . . Nagel does not make it clear exactly how an event located com-
pletely outside a life’s temporal span diminishes the life itself. The
cases he actually analyzes are not by themselves sufficient to show
this, since in each of them a subject persists, during the time of the
bad event, who has at least a strong claim to be identical with the
subject to whom the bad event is a misfortune. In the betrayal case,
this subject is clearly the very same, and is a subject of possible, if
not actual, experience in relation to that event. In the second case, it
is hard not to feel that the continued existence of the damaged per-
son, who is continuous with and very plausibly identical with the
former adult, gives the argument that the adult has suffered a loss at
least part of its force. Where death is concerned, however, there is no
subject at all on the scene, and no continuant. So it remains unclear
exactly how the life that has ended is diminished by the event.7
Why exactly is it thought to be so important to produce an example

in which “the subject does not persist?” I shall return to this question
below, but I would first suggest that a quite natural response would be
that, in such a circumstance, it is impossible for the individual to have
any unpleasant experience as a result of the event which purportedly
is bad for him or her. That is, it is plausible to suppose that the reason
why the subject’s going out of existence is problematic is that (on the
assumption that death is an experiential blank), the (nonexistent) agent
cannot have any unpleasant experience. This thought makes it natural
to seek to develop examples in which it is indisputably impossible for
the (still existing) individual to have any unpleasant experience as a
result of the purportedly harmful event, and yet the person does appear
to be harmed.8
Consider the following two examples. The first is a modification of the
case presented by Nagel; it employs the signature structure of preemp-
tive overdetermination found in the “Frankfurt-type” counterexamples
to the Principle of Alternative Possibilities.9 Here is my presentation of
the case:
Imagine first that the example is as described by Nagel. You are

betrayed behind your back by people who you thought were good
friends, and you never actually find out about this or have any
bad experiences as a result of the betrayal. But now suppose that
these friends were (very) worried that you might find out about the
betrayal. In order to guard against this possibility, they arrange for
White to watch over you. His task is to prevent you ever from find-
ing out about the betrayal. So, for example, if one of the individu-
als who betrayed you should decide to tell you about it, White can
prevent him from succeeding: White can do whatever is required to
prevent the information from getting to you. Or if you should begin
to seek out one of the friends, White could prevent you from suc-
ceeding in making contact. I simply stipulate that White is in a posi-
tion to thwart any attempt by you or your friends to inform you of
what happened.10
Since everything that actually happens among your friends and to
you and your family is exactly the same in my version and Nagel’s
version, I claim that it is plausible that the betrayal harms you. That
is, it is plausible that the betrayal harms you in Nagel’s version, and
if harm supervenes on what “actually happens to you” (in some
physico/causal sense) and your loved ones, then you are harmed in
my version of the case. But in my version, it is not just true that you do
not experience anything unpleasant as a result of the betrayal—you
cannot.
The second case owes much to an example by Jeff McMahon.11 Here
is the example:
. . . your daughter is trekking in the Himalayas while you are at home
in the United States. Tragically, she dies in an accident. I believe that
you are harmed by your daughter’s death—a bad thing has happened
to you—even before you find out about it. Suppose, further, that you
die without ever finding out about the accident in the Himalayas;
imagine, for example, that you die of a heart attack just five minutes
after your daughter dies. You never find out about her death, and,
given plausible assumptions about the situations of you and your
daughter, you cannot find out about it. Nevertheless, it seems to me
that you have been harmed (at least, for the five minutes of your con-
tinued life) by the death of your daughter. And here it is not merely
the case that you do not have any unpleasant experiences as a result
of your daughter’s death; in addition, it is, at least on a very natural
understanding of “possibility,” impossible for you to have any such
experiences as a result of her death.12
Nussbaum has responded to the latter case as follows (and, presum-
ably, her comments would also apply to the former):
I do not find Fischer’s counterexamples altogether convincing:
like the Nagel examples I criticize, they all involve a subject who
continues to exist, however briefly, during the time when the bad
event takes place. Even if the mother dies shortly after her daugh-
ter’s death, and without receiving news of it, the idea that a bad
thing has happened to her surely rests, at least to some extent, on
the thought that the mother is there in the world when the daughter
dies. There is a her for the bad thing to happen to. This, of course,
is not true of one’s own death; the bad event just is the cessation of
the subject (Lucretius profoundly suggests that we believe death to
be bad for us through a mental sleight of hand, in which we imag-
ine ourselves persisting and watching our own loss of the goods of
life). The right parallel, then, would be the case in which the mother
and the daughter die at precisely the same instant. In this case I
think we would not confidently assert that the mother has suffered
a bereavement.13
III. SUITS AND HETHERINGTON
III.1. Suits
David Suits does not find my modification of Nagel’s example entirely
convincing.14 Indeed, he says:
This [the modified version of the Nagel example of betrayal behind
one’s back] seems to be a quite fanciful—no, a desperate—attempt
to bolster the example. First of all, we are never in a position to know
that any precaution against harm (for that is all White is) is guaran-
teed to be successful in a case such as betrayal, where the effects can
be far-ranging and difficult to trace . . .
Second, it seems to me that if White is really so clever as all that,
then he could make his job immeasurably easier simply by prevent-
ing the secret betrayal in the first place. So now the question is this:
What is the difference between, on the one hand, a secret betrayal
which, on account of magic, can have no bad effects whatsoever
on you, and, on the other hand, there never having been a secret
betrayal after all? . . . Let’s invent a counter-story: All your life is char-
acterized, as far as you can tell, by the unwavering loyalty of your
friends. Nothing whatsoever in your experience leads you to believe
that any of your friends are not after all your friends; in fact, all your
experience is to the contrary. All attempts to discover betrayal have
come to naught. What shall you do with the hypothesis that there
might nevertheless be some secret betrayal? What will your thera-
pist say about your speculations that there is a very cunning Mr.
White who is preventing all relevant effects of this secret betrayal
from reaching you?
In what sense then could it be said that something happened that
was bad for you? Well, the only answer is that if there was a secret
betrayal, then it was after all a betrayal. Now of course to call some-
thing a betrayal is to lead us to expect harmful consequences. That
is the way we have come to know the world. . . . The best that can be
said is that if, somehow, I were absolutely convinced that the “vic-
tim” would not and could not be harmed in any way, then I would
have to say that what takes place is not a betrayal at all.15
In reply to Suits, I would begin by pointing out that the example is
indeed fanciful, and admittedly so. It is a thought-experiment, with all
of the attendant methodological risks (and, I believe, benefits). Granted:
in ordinary life, we are never in a position to know that a given precau-
tion against harm is guaranteed to be successful. I am not proposing
this as empirically plausible or feasible, but as conceivable and thus
metaphysically possible. Imagine, if you will, that White has Godlike
foreknowledge of the future. I do not believe that the philosophical point
is affected by Suits’ contention that we (as we actually are) could never
have the required sort of certainty.
Further, it is quite beside the point that White “could make his job
immeasurably easier simply by preventing the secret betrayal in the
first place.” This may be true, but in the example as I presented it,
White does not prevent the secret betrayal. One could certainly tell a
different story, but, in the story I told, White is a merely “counterfac-
tual intervener”; the example thus has the characteristic structure of a
“Frankfurt-type case” (as pointed out above). If the story I told is coher-
ent, then a theorist is intellectually required to take it into account—to
show how its point fits with his own view, even if his own view fits
nicely with another story.
In the “counterstory” told by Suits, there is no act of betrayal—at
least as far as anyone can tell. This is fundamentally different from my
story, in which an act of betrayal does in fact take place. Simply put,
there is a basic, clear difference between a case in which a betrayal actu-
ally takes place and one in which a betrayal does not take place, but
would have under certain hypothetical circumstances (would have, let
us say, but for the intervention of White). In ordinary life, given no evi-
dence of betrayal, it would be unhealthy obsessively to seek evidence
of a betrayal (and one’s therapist would legitimately be concerned!).
But, again, that is quite beside the point. The example is one in which it
is simply stipulated that there was a secret betrayal, and we are invited
to consider whether this in fact harms an individual who never has any
unpleasant experience as a result. Of course, it is not a suggestion of the
story that in ordinary life one should obsessively seek evidence of the
infidelity of friends and loved ones!
Suits continues to press his case:
The best that can be said [about the example as presented by Fischer]
is that if, somehow, I were absolutely convinced that the “victim”
would not and could not be harmed in any way, then I would have
to say that what takes place is not a betrayal at all. I might not know
what to call it . . .
A White-managed secret betrayal is no different from a merely

hypothetical betrayal. Real betrayals, as we have come to know
them, are like the incautious firings of guns: if they do not on some
particular occasions have bad consequences, then they are at least
very risky. And so we invoke precautions which, on the basis of past
experience, we expect will minimize such risks. Suppose everyone
agrees that all reasonable precautions are in place. We fire the gun,
and no one is harmed. Is this bad for someone? We are having fun
shooting at paper targets; we are in an enclosed firing range with
thick concrete walls; no one else is around. Who dares to complain?
Who comes forward and says that something bad has happened
as a result of our target practice, even though no one has been harmed,
and even though no one can make out a case for possible harm, given our
precautions? Incautious firing of guns is risky, but once the pre-
cautions are in place, then firing the gun is not at all incautious.
Similarly, betrayals are risky, but once White is in place, then there
is no betrayal after all.16
For my purposes, it does not matter what we call the behavior in ques-
tion—a “betrayal” or (say) “characterizing you negatively behind
your back,” or whatever. The question simply is whether such behav-
ior harms you. It is stipulated that you experience nothing unpleas-
ant as a result of the behavior in question; now the issue is whether,
nevertheless, you have been harmed. It may be that a White-managed
secret betrayal is no different from a merely hypothetical betrayal
insofar as you do not experience anything unpleasant as a result of
the behavior in either case; but it does not follow that there is no dif-
ference between the cases with respect to the issue of whether you are
harmed. As I said above, an actual betrayal is different from a merely
hypothetical betrayal: in the case of an actual betrayal, something has
happened which (arguably) has harmed you. Suits does not discuss
Nagel’s other case (the case of the individual who is reduced to the
state of a “contented infant” by (say) a stroke), but I would make the
parallel claim about this case: an actual stroke is crucially different
from a merely hypothetical stroke (even though, by hypothesis, the
individual in question does not experience anything unpleasant in
either case).
Yes, typically betrayals are risky; but the story does not purport to
portray a typical case. A theory needs to be right even in atypical cases.
A “real betrayal” (or a betrayal under normal circumstances) is like
the incautious firing of a gun: there is a considerable risk of causing
unpleasant experiences in both cases. When one has taken the sort of
precautions described by Suits, then firing the gun is not incautious; as
with the White-managed “betrayal” example, there is no risk of caus-
ing unpleasant experiences. But the cases are significantly different.
Whereas I am inclined to say that the negative characterizations of you

by your so-called friends harm you, by the very nature of the behavior,
I have no similar inclination to say that a mere firing of a gun, where
there is no chance of hitting anyone, can harm anyone (except, perhaps,
the gun-firer, but that raises different issues . . . ). The contention is not
that all riskless behavior (or that all riskless behavior which would be
risky, but for the precautions) harms others; the contention is only that
some does.
III.2. Hetherington
Stephen Hetherington offers a fascinating critique of Nagel’s example
of the (alleged) betrayal, even as modified a la the Frankfurt-type exam-
ples.17 Hetherington says:
By being betrayed, some of your beliefs are rendered false. More
vitally, some of your personally important beliefs are rendered false.
For a start, you believe that your friends are loyal to you in standard
ways; moreover, you care that this belief of yours be true. The betrayal
makes the belief false, though. And this harms you, even though you
are wholly unaware of its doing so, indeed even if (as in Fischer’s
case) you could never experience any consequence of the betrayal.
The harm occurs because the falsification of your belief diminishes
you as someone who wishes to believe only what is true about what-
ever is important to you. You wish to have those true beliefs; your
wish is not being fulfilled. So, although this harm is one of which
you are unaware, it is a harm nonetheless. If you were to realize that
your belief was false, you would be upset. And even if—perhaps
because your circumstances are as described in Fischer’s case—you
could never come to realize that your belief is false, its being false
still makes you that much less cognitively successful as a person than
you would wish to be. You are—by now being mistaken about some-
thing that matters to you—that much “out of step” with the world,
notably with some parts of the world that matter to you. Insofar as it
matters to you to be right about what matters to you, therefore, your
being mistaken about what matters to you harms you. That harm is
of at least metaphysical significance, as your status as a true believer
on what you care about is harmed.18
Hetherington goes on to elaborate the relevant sort of harm:
Your realizing that you are being harmed in that way inflicts a further
harm of its own; a fortiori, so does its being impossible for you ever to
find out that you were harmed. What is that further harm? It is the
harm of human absurdity. If the belief is important enough to you,
and if the betrayal is sufficiently fundamental, then your life might
well have become somewhat absurd as a result.
. . . I am talking about an objective sort of absurdity. It is objective,

in that it is not an awareness, either actual or even possible, of a dis-
crepancy; instead, it is the existence of a basic discrepancy, one that
can exist between a person and the world as a whole, and one that
can exist without the person being aware of it, perhaps even with her
being unable to be aware of it. This sort of absurdity is also cognitive,
in that it is a matter of a person’s failing to be aware of how poorly at
least some of her cognitive efforts are informing her about the world
in which she has to live. It is essentially her being mistaken about that
world.
. . . To the extent that your belief in your friends’ loyalty is also
important to you, that absurdity is even tragic, no matter whether or
not you are aware of this tragic dimension to your life. You trust the
friends; you show them your feelings; presumably, you interact with
them in what you assume is a context of respect and honesty. All
the while, however, they know that you do not know how untrust-
worthy they are in relation to you. They are aware of how misplaced
is your trust in them. . . . Even if your friends are not laughing at you
behind your back, they could be; and in that sense, the world is doing
so, at any rate.19
On Hetherington’s view, the badness of the betrayal is understood in
terms of creating a certain sort of dissonance or discordance; one has a cer-
tain set of beliefs about important features of our lives, one structures
one’s life around these beliefs, and yet they are false. As Hetherington
puts it, “A person can be harmed by something insofar as it renders false
some belief of hers whose truth matters to her.”20 In contrast, of course,
death cannot be a bad thing by creating a dissonance or discordance
between one’s cognitions and the external reality, since death destroys
the cognizer. If what disturbs us about the Nagel-type betrayal case
(in its various versions) is that it creates a kind of absurd discordance,
then, since death cannot be bad in this way, the example cannot be
legitimately employed in order to defend the deprivation thesis about
death’s badness.
Although I find Hetherington’s analysis helpful, I am not sure that
it captures the entire truth about the badness of the betrayal (or death).
In any case, it is interesting to note that Hetherington (like Suits) only
focuses on Nagel’s betrayal example, and not on the example of an
individual who is reduced to the mental status of a contented infant (as
a result of a stroke or accident). It seems to me that the stroke (or acci-
dent) has indeed been a bad thing for the individual, even though (by
hypothesis) he or she has not experienced anything unpleasant (and
cannot do so) as a result. Further, it seems to me that the badness here
cannot be analyzed in the way suggested by Hetherington in regard to
the betrayal case.
As opposed to the betrayal case, in the case of the stroke victim

there need not be a discordance of the sort pointed to by Hetherington.
The stroke victim presumably has no memories of his previous activi-
ties, and need not be radically mistaken about his current situation
and abilities. He does not believe that he has capabilities that are in
fact wildly “out of sync” with reality; he need not be fundamentally
mistaken about the world or his relationship to it. I do not believe that
the stroke victim is absurd; his situation is sad and maybe even tragic
(in a certain way), but not absurd. He has not misplaced loyalty or trust,
and there is no tendency to suppose that the world is, as it were, laugh-
ing at him.
The two examples, then, are different. If one agrees that the stroke or
accident has indeed harmed or been a bad thing for the individual, then
one cannot explain the badness in terms of the sort of discrepancy and
absurdity discussed by Hetherington. It seems to me that the badness is
not so much a matter of the discordance or discrepancy between inter-
nal cognitive states and the external world, but more “direct”—the tak-
ing away of capacities to function and experience that were possessed
prior to the unfortunate event. It is the lack of these more complex
capacities and experiences in itself, rather than some sort of discrep-
ancy between the individual’s awareness and reality, that is bad. Or so
it seems to me.
Nussbaum states, in a passage quoted above, “In the second case
[the stroke case], it is hard not to feel that the continued existence of the
damaged person, who is continuous with and very plausibly identical
with the former adult, gives the argument that the adult has suffered a
loss at least part of its force.” One might then argue that there is even in
this case a kind of discordance or discrepancy—between the former and
current capacities of the individual. But the same can be said of death;
it creates a disparity between prior and subsequent capacities—the lat-
ter having been reduced to zero. What cannot be said, in the case of
death, is that there remains an existing individual who has been dimin-
ished (and thus that the remaining capacities are greater than zero, as it
were)—but it is essentially controversial, within this dialectical context,
whether the removal of the subject is consistent with badness: it cannot
simply be assumed here that the removal of the subject is not consistent
with badness, and thus that there must be a nonzero capacity remain-
der. I turn now to a more careful discussion of precisely this sort of
dialectical subtlety.
IV. DEATH AND DIALECTICAL STALEMATES

Elsewhere I have sought to describe an argumentative structure I have
called a “Dialectical Stalemate.” I have suggested that this sort of struc-
ture is found in many of the most intractable philosophical puzzles:
Frequently in philosophy we are engaged in considering a cer-

tain argument (or family of arguments) for some claim C. The
argument employs a principle P. Allegedly, P supports C. Now
the proponent of the argument may be called upon to support the
principle, and he may do so by invoking a set of examples (or
other considerations). Based on these examples (or other consid-
erations), he argues that the principle and thus also the philo-
sophical claim are to be accepted.
But the opponent of the argument may respond as follows. The
examples are not sufficient to establish the principle P. One could
embrace all the examples and not adduce P to explain them: rather,
it is alleged that a weaker principle, P*, is all that is decisively
established by the examples (or other considerations). Further, P*,
in contrast to P, does not support C. Finally, it is very hard to see
how one could decisively establish P. One reason it is so difficult is
that it at least appears that one cannot invoke a particular example
which would decisively establish P without begging the question in a
straightforward fashion against either the opponent of P or the oppo-
nent of C. Further, it also seems that one cannot invoke a particular
example which would decisively refute P without begging the ques-
tion against the proponent of P or the proponent of C. These condi-
tions mark out a distinctive—and particularly precarious—spot in
dialectical space.
I shall call contexts with roughly the above form, “Dialectical
Stalemates.”21
Take, for example, the Basic Argument for the Incompatibility of Causal
Determinism and the sort of free will that involves alternative possibili-
ties.22 The argument, dubbed the “Consequence Argument” by Peter
Van Inwagen, proceeds from the point that causal determinism implies
that all our behavior is the consequence of the past and the laws of nature
to the conclusion that we lack the sort of free will that involves genuine
access to alternative possibilities (if causal determinism obtains).23 The
argument can be given in different forms, but typically it employs prin-
ciples that putatively encode our commonsense views about the fixity
of the past and natural laws. Sometimes the argument employs a modal
principle that allegedly captures intuitive ideas about the transfer of
powerlessness: if one is powerless over one thing, and powerless over
that thing’s leading to another, one is powerless over the other thing.
A number of examples can be adduced that seem to support a
Principle of the Fixity of the Natural Laws. The problem is that there are
different ways of seeking to capture the intuitive, commonsense notion
of the fixity of the laws of nature. None of the examples appear to sup-
port the incompatibilist way of capturing the kernel of truth in those
examples over the compatibilist way. That is, consider the following two
Fixity of the Laws Principles:
(IFL) For any action Y, and agent S, if it is true that if S were to do

Y, some natural law which actually obtains would not obtain, then
S cannot do Y.
(CFL) For any action Y, event b, agent S and times t1, t2, and t3, (t1
prior to t2 prior to or simultaneous with t3), if (1) Y’s occurring at
t2 is inconsistent with the laws of nature, or (2) Y’s occurring at t2
would cause some event b’s occurring at t3 and b’s occurring at t3 is
inconsistent with the laws of nature, then S cannot at t1 do Y at t2.24
The problem is that none of the examples adduced by incompatibil-
ists such as Carl Ginet and Van Inwagen, or by anyone else (as far as
I know), can show decisively that IFL is to be preferred to CFL. That is,
the relevant data do not support one principle over the other.
Consider, for example, Van Inwagen’s examples of someone produc-
ing a machine that would cause things to go faster than the speed of
light, or someone engaging in selective breeding to produce a human
being who could get along without vitamin C. He points out that no
one has it in his power to do such things, insofar as the laws of nature
entail that nothing travels faster than the speed of light, and that human
beings cannot get along without vitamin C. It is obvious, however that
such examples do not support IFL over CFL.25 If one states that causal
determinism might turn out to be true, and, after all, it is “just obvious”
that we are free in the sense that involves alternative possibilities, this
would clearly be question-begging in the relevant dialectical context.
Exactly similar considerations apply to the notion of the fixity of the
past, as well as to the modal principle (or related principles) sometimes
employed in the argument for incompatibilism.26
As I said above, a modal principle is sometimes employed as part of
the Consequence Argument (for the incompatibility of causal determin-
ism and freedom). The modal principle, the “Principle of the Transfer
of Powerlessness,” is structurally parallel to the “Principle of Closure
of Knowledge Under Known Implication.” Indeed, the principles are
the same, except for the different interpretations of the relevant modal-
ity.27 On this principle, if someone knows that p, and knows that p
implies q, then he knows that q. Just as the Principle of the Transfer of
Powerlessness is employed to generate a sort of free will skepticism
(i.e., incompatibilism about causal determinism and the sort of free-
dom that involves genuine access to alternative possibilities), so the
Principle of Closure of Knowledge Under Known Implication is some-
times employed to generate epistemological skepticism.
The argument would go as follows.28 Evidently, I know that there
is a laptop in front of me. But there being a laptop computer in front
of me entails that I am not a brain-in-a-vat being stimulated to falsely
believe there is a laptop in front of me, and I know this. Thus, given the
Principle of the Transfer of Knowledge Under Known Implication, it
follows that I know that I am not a brain-in-a-vat being stimulated to

falsely believe (for instance) that there is a laptop in front of me. I do
not, however, know this; in the lingo, I cannot “rule out this skeptical
counterpossibility.” Thus, modus tollens gives us the conclusion that I
do not know that there is a laptop computer in front of me (or, for that
matter, any contingent proposition about the external world).
Now one could simply state that it is “just obvious” that I know that
there is a laptop in front of me, even though I cannot rule out the skep-
tical counterpossiblity and thus I do not know that I am not a brain-
in-a-vat. Thus, someone could insist that the Principle of the Transfer of
Knowledge Under Known Implication is invalid. The problem here is
painfully clear: the data invoked are essentially contested, and thus it is
question-begging, in the relevant dialectical niche (in which epistemo-
logical skepticism is being considered seriously), simply to assert it.
Consider, also, the “Principle of the Transfer of Nonresponsibility.”
This is the same modal principle as above, except for the interpreta-
tion of the modality. It states that if you are not responsible for one
thing, and you are not responsible for that thing’s leading to another,
you are not responsible for the other.29 An incompatibilist about
causal determinism and moral responsibility might use this modal
principle in an argument for this sort of incompatibilism. Assuming
causal determinism, some fact about the distant past, together with
the laws of nature, entail the present and future facts. I am not morally
responsible for the past fact. Further, I am not morally responsible for
the laws of nature, and thus, I am not morally responsible for the past
fact’s leading to the present and future situations of the universe.30
Now, given the Principle of the Transfer of Nonresponsibility, it fol-
lows that I am not morally responsible for the present (and thus for
my behavior).
There are various ways of responding to this sort of argument.
I believe there are promising ways of seeking to show the modal
principle—the Principle of the Transfer of Nonresponsibility—to
be invalid.31 But I do not think the following strategy is promising.
A compatibilist might simply point out that it is “just obvious”that I
am morally responsible for my behavior, even if causal determinism
is true. (Perhaps this is because it is “just obvious” that I am free to do
otherwise, even if causal determinism obtains.) Further, such a com-
patibilist might concede that I am not morally responsible for the dis-
tant past, nor am I morally responsible for the connection between the
past and the present (given that this connection instantiates a natural
law). So this sort of compatibilist simply insists that it is “just obvi-
ous” that the Principle of the Transfer of Nonresponsibility is invalid.
It should however be (painfully) clear that this move is dialectically
“traif” (unkosher). It simply begs the question against the incompatibil-
ist. A compatibilist about causal determinism and moral responsibility
might be entirely justified in rejecting the modal transfer principle, but

not on this basis.
In my previous work on these subjects, I suggested that it is one of
the salient characteristics of a perennial or classic philosophical prob-
lem that it involves the signature structure of a Dialectical Stalemate.32
I added that the Free Will Problem (which is really a family of related
problems and puzzles) is a “true philosophical classic” in part because
it is an environment rich with Dialectical Stalemates.
It should be evident that the debate discussed above about death is a
Dialectical Stalemate. Nussbaum points out that the examples invoked
by Nagel, even as modified by me, cannot decisively show that death
is a bad thing for the individual who dies. Even if it is plausible to say
that the examples of betrayal and stroke involve badness for the rel-
evant individuals, the examples are different from death in that they
involve the persistence of the subject. Thus they cannot in themselves
decisively show that death can be a bad thing for the individual who
dies. In framing and considering general principles relating to harm, it
is evident that there will be different principles, and that the examples
in question will not in themselves decisively support a principle strong
enough to entail that death can be bad for the individual who dies over
a slightly weaker principle—a principle that does not have this conclusion.
As Nussbaum points out, all of the examples “involve a subject who
continues to exist, however briefly, during the time the event takes
place.”33 But it is hard to see how one can get rid of this feature and
have an example that would not beg the question at issue about death’s
badness. It would seem that, given the definition of death, any case in
which the event in question removes the subject would be essentially
contested. Thus we have a classic Dialectical Stalemate.
It is important, however, to keep in mind that a Dialectical Stalemate
need not result in our inability to make any philosophical progress or to
come to any useful philosophical conclusions. In a Dialectical Stalemate,
no example can in itself decisively establish the relevant conclusion (with-
out begging the question). Nussbaum is completely correct to point out
that none of the examples above can in themselves decisively establish
that death can be a bad thing for the individual who dies. But I want
to highlight what I have written about how to proceed in this sort of
argumentative neighborhood:
I do not however think that Dialectical Stalemates should issue in philo-
sophical despair. An opponent of the principle under consideration may
demand that its proponent provide examples which absolutely require
one to accept the principle. But I would claim that this is unreasonable.
It may even be true that it is necessarily the case that if a philosopher
argues for a certain general principle by giving examples, a weaker
principle can be found that is the strongest principle the examples
support (strictly speaking). The crucial issue becomes whether it
is plausible to accept the stronger principle, if one accepts the weaker

principle. Considerable philosophical ingenuity can be displayed in
generating examples, which invite one to accept the stronger principle
as well as the weaker principle, or in explaining in a non-ad hoc fashion
exactly why one should only accept the weaker principle. Alternatively,
philosophical creativity can issue in a restructuring of the problem;
that is, one might find some other principle P* that can be employed
to establish C, or perhaps one can show in some way that C wasn’t that
interesting after all.34
In my work on Free Will, I have suggested a certain kind of restructuring
of the traditional debates. I have contended that there are importantly
different kinds of freedom (or control), and that seeing this can help us
to sidestep some of the Dialectical Stalemates embedded in traditional
debates about Free Will. In our discussion of death here, I would opt
for the other course of action alluded to above. That is, I would insist
that it is unreasonable to demand an example that would in itself deci-
sively show (in a nonquestion-begging way) that death can be a bad
thing for an individual. (I actually believe that Nussbaum may well be
in agreement with me about this; her explicit contention is merely that
the examples in themselves do not show that death can be a bad thing
for the individual who dies, and this leaves it open that other consider-
ations may be invoked.)
I stated above that the crucial question in a Dialectical Stalemate
may be whether it is plausible to accept the stronger principle, if one
accepts the weaker principle. So the key issue is whether the differ-
ence between the stronger and weaker principles makes a difference to
the issue at hand. I would argue that, whereas the Nagel-type cases
discussed above all involve the persistence of a subject, it does not
seem plausible that this feature is crucial—that it is this feature (perhaps
together with others that are present in the examples) that inclines us
to say that the relevant individual is harmed, and in the absence of
which we would not be so inclined. So, my view is that if it is plausible
to hold that the individual is harmed in the betrayal, stroke, and trek-
king cases, it would also be plausible to maintain that he is (or can be)
harmed by death.
Recall that Nussbaum stated (about the trekking case), “The right
parallel, then, would be the case in which the mother and the daughter
die at precisely the same instant. In this case I think we would not confi-
dently assert that the mother has suffered a bereavement.”35 It is correct
that it would not be dialectically kosher to invoke the original version
of the example as decisive evidence on the basis of which to conclude
that death can be a bad thing for an individual. And I would agree with
Nussbaum that one cannot conclude with confidence that the mother
has suffered a bereavement, simply in virtue of consideration of her
version of the example. But it would not be inappropriate to employ the
original version of the example, together with suitable versions of the

other examples, to generate the preliminary conclusion that a persist-
ing subject can be harmed even in contexts in which he is unaware of
the harm and does not suffer as a result. Then, if it is plausible that the
difference between death and these contexts does not make a difference
to badness, one could conclude that, even in Nussbaum’s version of the
trekking example, the mother has been harmed.
Nussbaum holds that the Nagel-type examples do not in themselves
decisively establish that death can be a bad thing for the individual who
dies. She does however have reservations about the main argument.36
Her worries stem not so much from the examples, but from consider-
ations about our ongoing projects in life. Although I am in agreement
with Nussbaum that the examples do not in themselves decisively
establish that death can be a bad thing for the individual who dies, they
are helpful insofar as they point us to that conclusion; they help to pro-
vide a strong plausibility argument for it. They challenge the Epicurean
to state why one should think that the examples are importantly dif-
ferent from death. More specifically, why exactly should the pertinent
feature—that death deprives us of the subject—make the difference in
question? It is perhaps natural to suppose that the problem with there
being no subject is that this issues in an impossibility of experience; but we
have seen that the mere impossibility of experience is not what makes it
the case that death cannot be bad, if it were indeed the case that death
cannot be bad for the individual who dies. So why exactly does it mat-
ter that the subject is removed?
V. LUCRETIUS’S “PROFOUND INSIGHT”

Nussbaum states, “Lucretius profoundly suggests that we believe death
to be bad for us through a mental sleight of hand, in which we imagine
ourselves persisting and watching our own loss of the goods of life.”37
Stephen Rosenbaum also highlights this view of Lucretius, according
to which at least part of our view that death is bad can be explained
in terms of a natural, pervasive mistake—a tendency mentally to proj-
ect ourselves into the future as somehow “there” and “watching” and
perhaps even “suffering,” even after our death. The mistake consists in
covertly assuming that one will continue to exist and have a point of
view, even after death.
In his more recent work on death, Nagel also emphasizes this subjec-
tive tendency to project oneself into the future:
It hardly needs saying that we are accustomed to our own exis-
tence. Each of us has been around for as long as he can remember;
it seems the only natural condition of things, and to look forward
to its end feels like the denial of something which is more than a
mere possibility. It is true that various of my possibilities—things

I might do or experience—will remain unrealized as a result of
my death. But more fundamental is the fact that they will then
cease even to be possibilities—when I as a subject of possibilities
as well as of actualities cease to exist. That is why the expectation
of complete unconsciousness is so different from the expectation
of death. Unconsciousness includes the continued possibility of
experience, and therefore doesn’t obliterate the here and now as
death does.
The internal awareness of my own existence carries with it a par-
ticularly strong sense of its own future, and of its possible continu-
ation beyond any future that may actually be reached. It is stronger
than the sense of future possibility attaching to the existence of any
particular thing in the world objectively conceived—perhaps of a
strength surpassed only by the sense of possible continuation we
have about the world itself.38
I believe that we do indeed have the sort of tendency noted by
Lucretius (and Nagel), but that it is unclear whether its removal
would, on balance, be helpful. This is because I think that sometimes,
at least, our subtle projection of our perspective into the future, even
after death, is comforting; we picture ourselves at our own funeral
(listening to the no-doubt glowing eulogies of family, loved-ones, or
friends), or, perhaps less nobly, we picture ourselves enjoying other
emotions upon observing reactions by others to our own deaths. Now
perhaps, strictly speaking, we are simply picturing the eulogies and
the reactions, and not our own enjoyment of them. But it is Lucretius’s
suggestion that we tend to assume, perhaps subtly, that we are (some-
how) still able to experience or be aware of the world, even after our
deaths.
Consider these passages from an amusing piece by the comedian,
Richard Lewis:
[Eulogizing myself at my own funeral] would be sort of a dream
come true . . . My feeling is that since everyone in my life (except
God) puts some kind of spin on me as to what and why and who
and where I am, I at least deserve to get my licks in when it’s me in
that . . . casket.
I really don’t want to scare people who show up to pay their last
respects for me, but I feel that I owe it to my soul (before it gets too set
in its ways) to put in my own two cents. At the risk of sounding cocky,
I expect and want a lot of people at the last good-bye and the more cry-
ing the better, because I suffered a lot of emotional shit and it would
do my corpse good to hear a little sobbing, albeit too little too late.
I don’t want a small, private funeral. Not just because I will prob-
ably be forever narcissistic but also because I so rarely left my house
when I was living. I think it would be nice to see people for a change,
even if I am dead. So come on down!!!39
David Sedaris’s short story, “The Last You’ll Hear from Me,” also con-
tains some deliciously malevolent projections into the future:
Dear Friends and Family,
By the time you receive this letter I will be dead. Those of you attend-
ing this service are sitting quietly, holding a beautiful paperweight, a
gift from the collection, which, in life, had been my pride and joy. You
turn the paperweight over in your hands, look deep inside, at the object
imbedded in the glass, be it a rose of a scorpion, whatever, and through
your tears you ask, “What is death like?” By this time I certainly know
the answer to that question but am unable to give details . . .
If my instructions were followed the way I wanted them to be
(see attached instruction envelope #1), this letter is being read to you
from the pulpit of The Simple Shepherd Church of Christ by my best
friend, Eileen Mickey (Hi, Eileen), who is wearing the long-sleeved
Lisa Montino designer dress I left behind that always looked so good
on me. (Eileen, I hope you either lost some weight or took it out some
on the sides or you’re not going to be able to breathe. Also, remem-
ber it needs to be dry-cleaned. I know how you and your family love
to skimp, but please, don’t listen to what anyone says about Woolite.
Dry-clean!)
Most of you are probably wondering why I did it. You’re ask-
ing yourselves over and over again, “What could have driven Trish
Moody to do such a thing?”
You’re whispering, “Why, Lord? Why take Trish Moody? Trish
was a ray of bright sunshine, always doing things for other people,
always so up and perky and full of love. Pretty too. Just as smart and
sweet and pretty as they come.”
You’re probably shaking your heads and thinking there’s plenty
of people a lot worse than Trish Moody. There’s her former excuse
for a boyfriend, Randy Sykes, for example. The boyfriend who, after
Trish accidentally backed her car over his dog, practically beat her
senseless. He beat her with words but still, it might as well have been
with his fists. . . . The Dog’s death was a tragic accident but perhaps
also a blessing in disguise as Randy tended to spend entirely too
much time with it. . . .
What did Trish’s mother say when her daughter, heartbroken
over her breakup with Randy, came to her in search of love and
understanding?
“If you’re looking for sympathy you can find it between shit and
syphilis in the dictionary.”
Perhaps my mother can live with slogans such as this. I know
I can’t.
Neither can I live surrounded by “friends” such as Annette Kelper,

who desperately tries to pretend that nobody notices the fact that
she’s balding on top of her head. That’s right. Look closely—balding
just like a man. Perhaps Randy feels sorry for chrome-dome Annette.
Maybe that’s why he was seen twice in her company in a single five-
day period . . .
Is everyone on earth as two-faced as Annette Kelper? Is every-
one as cruel as Randy Sykes? I think not. Most of you, the loved
ones I left behind, are simple, devoted people. I urge you now to
take a look around the room. Are Randy Sykes and Annette sitting in
the audience? Are they shifting uncomfortably in the pew, shielding
their faces with the 8-by-11 photograph of me I had reproduced to
serve as a memento of this occasion?
Fancy little shitheads! Look at them, take a good hard look at
them. It’s their fault I’m dead. They are to blame. I urge you now to
take those paperweights and stone them. Release your anger! The
Bible says that it’s all right to cast the first stone if someone dead
is telling you to do it and I’m telling you now, pretend the paper-
weights are stones and cast them upon the guilty. I’ve put aside my
savings to pay for damages to the walls and windows. It’s money
I was saving for my wedding and there’s plenty of it so throw! Hurt
them the way they hurt me! Kill them! No one will hold you respon-
sible. Kill them!40
Much of the “fun” described above comes from illicitly assuming that
one would still be around to witness the events in question, even after
death. Much humor surrounding death, as well as many funerary prac-
tices, involve treating the dead as like the living in certain ways (ways
obviously seen to be inappropriate, in the case of the humor).41 I simply
wish to point out here that our (perhaps clandestine) assumption of
our continued presence and capacity to be aware of the world, even
after death, can be a source of comfort and consolation. If death is like
life in this respect, even if it is unpleasant, it is not entirely mysterious
and frightening—it has the ring of familiarity (note that chronic pain
sufferers sometimes think of their pain as “an old friend”). In any case,
removing this tendency to project ourselves into the future (as having a
persisting point of view) may lead to enhanced anxiety about death; the
tendency tames death, and without it, the stark nothingness and total
annihilation can seem more frightening. When death is completely dif-
ferent from life and lacking even in awareness, it can arouse less trac-
table, more unruly fears.
VI. THE BANQUET ARGUMENT

Let us turn to Lucretius’s “Banquet Argument.” As in the case of a ban-
quet, there is a definite pattern or temporal structure to a human life (in
the typical case). The conclusion of the Banquet Argument is something

like this: that our mortality is a necessary condition of our various activ-
ities having meaning and value of the sort we can comprehend and find
attractive. Nussbaum writes, “ . . . the removal of all finitude in general,
mortality in particular, would not so much enable these values [the val-
ues we find in friendship, love, justice, and the various forms of mor-
ally virtuous action, for example] to survive eternally as bring about the
death of value as we know it.”42 She contends that we would not have
the virtues without death. For example, Nussbaum says that courage
involves “a certain way of acting and reacting in the face of death,” and
moderation “is a management of appetite in a being for whom excesses
of certain sorts can bring illness and eventually death. . . . ”43
I have pointed out, in contrast, that death is not the only condition
that could provide a point or content to the virtues. Even in an immor-
tal life, there could be long stretches of physical and/or emotional dis-
ability, depression, anxiety, boredom, loneliness, and so forth. These
sorts of potential conditions could certainly give a shape to our lives
and content to the virtues. We do not need death in order to have dan-
ger, and to provide considerable impetus to ourselves to strive to avoid
terrible disability, dysfunctionality, and suffering. Nussbaum has sub-
sequently expressed her agreement with this point, saying, “So I agree
with Fischer: we need to take apart the different limitations of a human
life much more precisely, asking exactly how each of them works in
connection with the shaping of value.”44
In this section I wish to make a few tentative gestures toward under-
standing these issues better. First, note that the proponent of the view
that immortality is necessarily bad (such as Bernard Williams45) insists
that there are no conceivable circumstances in which immortal life
would be recognizably human and attractive. If in our immortal lives
we became decrepit or permanently disabled, this would certainly give
the relevant sort of “shape” to our lives, but there would emerge the
concomitant danger that our lives would be unattractive. Thus, the sort
of circumstances I am envisaging would involve the potential for long-
term and significant disabilities and suffering, but with subsequent
regeneration and recovery. This certainly seems both conceivable and
potentially attractive.
It is often stated that an immortal life would have no shape. How
could we care about something essentially amorphous? Consider, for
example, an ordinary physical object, such as a carpet. You might think
that you could simply expand the size of the carpet in your imagination
indefinitely and that you would thereby imagine a very large carpet—
indeed, an infinitely large carpet. But the problem is that a carpet is what
it is—a particular carpet—in virtue of its borders. As one expands the
carpet in one’s mind, it inevitably explodes into shapelessness as the
distinction between the carpet and the noncarpet surroundings becomes
blurred. Similarly, a statue is the particular statue it is, in part, because of

the contours of its borders; expand them indefinitely and one is in dan-
ger of having no sculpture at all, but a huge, formless blob of marble.
But I believe the analogies are misleading here. First, it is not entirely
evident that the thought-experiments issue in their putative conclu-
sions. Perhaps one can think of an indefinitely large carpet! But, more
important, why not think of infinitely long life as similar to an indefi-
nitely long electrocardiogram? (What could be more appropriate in this
context than the representation of one’s heartbeat—the physical engine
of life?) I do not see why one could not have an indefinitely long elec-
trocardiogram, with a given pattern displayed at any given temporal
period. After all, one need not expand to infinity all of the spatial dimen-
sions of the relevant object; one can have an indefinitely long electro-
cardiogram with specific structure and content along the way, since the
vertical dimension need not be extended along with the horizontal one.
It is surely a mistake to think that “shape” need be conflated with fea-
tures of one dimension—the boundless horizontal dimension.
Yes, a banquet has a definite and bounded temporal structure: appe-
tizers, soup, salad, main course, dessert and so forth. Also, our lives
typically have a certain narrative structure with a beginning, middle,
and end (carved up very roughly). It is of course normally thought to
be a virtue—a sign of great wisdom—to accept the finitude of our lives
and not engage in what the Greeks called “plenoexia” or a certain sort
of inappropriate “overreaching.” But the question at issue here is not
about our ordinary, normal lives, but about life’s possibilities, consid-
ered from a distinctively philosophical point of view. Why can a ban-
quet not be a kind of “temporal all-you-can-eat buffet?” Better, why
can we not imagine an indefinitely extended banquet, with suitable
intervals for recovery (and enjoyment of other activities)? This is, after
all, the way “foodies” tend to look at life already!
And our lives have a narrative structure, but why suppose that
essential to this sort of structure is finitude? After all, many people
watch soap operas, which are stories that are seemingly endless.
Perhaps, more carefully put, it seems to me that our lives could have
certain of the distinctive features of narrative structure without finitude.
Specifically, our lives can be thought to have value based on narrative
structure, even apart from whether the lives are bounded temporally.
So, for example, we value succeeding as a result of striving or learning
from past mistakes, rather than merely as a result of a windfall (such
as winning the lottery). The values encoded in human narratives could
still exist, even if the stories were infinitely long; these are a function of
relationships, not finitude. We mean different things when we advert to
the notion that our lives are (or correspond to) “stories,” and that they
have narrative structure. I believe that the chief element is narrative
value, rather than finitude.46
Consider, finally, what I would dub the “Super-Powers Problem.”

In an immortal life, presumably one would know that one has immor-
tality. By the way, this raises interesting questions about the concept of
immortality, as it plays a role in philosophical discussions. I am assum-
ing that the relevant individuals know that they are immortal, and not
simply contingently so; they know that they are invulnerable to death,
not just that in fact they will not die. But if one knows one is invulner-
able to death, one knows one could do lots of things without having
to worry about death—skydiving (without a parachute), rock climb-
ing in the most exotic and precarious places, and so forth. Further, one
knows that, no matter what happens to one, one will not die: so, some-
one could riddle you with bullets and you would nevertheless continue
to live. Given these realizations, one might conclude that one could do
just about anything, and, although such a life might seem at first attrac-
tive, it would be so fundamentally different from our own, finite, lim-
ited human lives as to be incomprehensible to us.
The reply is again that, although the envisaged circumstances
would be very different from our current status, they would not be
sufficiently different to justify the purported conclusion. As with the
virtues, we must remember that there are dangers other than death.
So, even if I realize that I am invulnerable to death and thus that I
would continue to live, even if I were to fall from a high mountain cliff
or were riddled with bullets, I would still realize that I would no doubt
be significantly damaged by such things. Consequent pain, suffering,
and disability would be a constraint against trying such antics, and
would also temper any inclination to suppose that one had “infinite”
or super-powers. Life would be different, but, arguably, analogous to
our finite human lives.47
NOTES
1. Martha Nussbaum, The Therapy of Desire (Princeton: Princeton University
Press, 1994); Martha Nussbaum, “Reply to Papers in Symposium on Nussbaum,
The Therapy of Desire,” Philosophy and Phenomenological Research 59 (1999),
pp. 811–819.
2. The main argument is laid out in Nussbaum, The Therapy of Desire, pp. 201–
202. I make a preliminary stab at discussing these matters in John Martin Fischer,
“Contribution to Symposium on Nussbaum’s The Therapy of Desire,” Philosophy
and Phenomenological Research 59 (1999), pp. 787–792.
3. Nussbaum, The Therapy of Desire, p. 203.
4. Fischer, chapter 6; and Fischer, “Contribution to Symposium on
Nussbaum’s The Therapy of Desire.”
5. Thomas Nagel, Mortal Questions (Cambridge, Cambridge University
Press, 1979), pp. 1–10; reprinted in John Martin Fischer (ed.), The Metaphysics of
Death (Stanford: Stanford University Press, 1993), pp. 61–69. The quotation is on
p. 69 of the reprinted version.
6. Nussbaum, The Therapy of Desire, pp. 201–202.
7. Ibid., pp. 205–206.

8. Various philosophers have faulted Nagel precisely for providing
examples in which (allegedly) the individual merely does not, rather than
cannot, experience anything unpleasant as a result of the event in ques-
tion: Harry S. Silverstein, “The Evil of Death,” The Journal of Philosophy 77
(1980), pp. 401–424; reprinted in John Fischer (ed.), The Metaphysics of Death,
pp. 95–110; Stuart Rosenbaum, “How to Be Dead and Not Care: A Defense of
Epicurus,” American Philosopohical Quarterly 23 (1986), pp. 217–225; reprinted
in Fischer (ed.), The Metaphysics of Death, pp. 119–134; and Walter Glannon,
“Temporal Asymmetry, Life, and Death,” American Philosophical Quarterly 31
(1994), pp. 235–244. I attempt to reply in chapter 3.
9. Fischer, chapter 3. The Principle of Alternative Possibilities states that
an individual is morally responsible for an action only if he could have done
otherwise. Harry Frankfurt first presented a purported counterexample to this
principle (or a template for such a counterexample) in Harry G. Frankfurt,
“Alternate Possibilities and Moral Responsibility,” The Journal of Philosophy 66
(1969), pp. 828–839. I have discussed such examples in (among other places)
John Martin Fisher, The Metaphysics of Free Will (London: Blackwell Publishers,
1994); John Martin Fischer “Recent Work on Moral Responsibility,” Ethics 110
(1999), pp. 93–139; and John Martin Fischer, “Frankfurt-Type Compatibilism,”
in S. Buss and L. Overton (eds.), Contours of Agency: Essays on Themes from Harry
Frankfurt (Cambridge: The MIT Press, 2002), pp. 1–26.
10. Fischer, chapter 3, p. 39.
11. Jeff McMahan, “Death and the Value of Life,” Ethics 99 (1988), pp. 32–61;
reprinted in Fischer (ed.), The Metaphysics of Death, pp. 233–266.
12. Fischer, “Contribution to Symposium on Nussbaum’s The Therapy
of Desire,” p. 789.
13. Nussbaum, “Reply to Papers in Symposium on Nussbaum, The Therapy
of Desire,” pp. 811–812.
14. David Suits, “Why Death Is Not Bad for the One Who Died,” American
Philosophical Quarterly 38 (2001), pp. 69–84.
15. Suits, “Why Death Is Not Bad for the One Who Died,” pp. 76–77.
16. Suits, “Why Death Is Not Bad for the One Who Died,” p. 77.
17. Stephen Hetherington, “Deathly Harm,” American Philosophical Quarterly
38 (2001), pp. 349–362.
18. Ibid., pp. 351–352.
19. Ibid., pp. 355–356.
20. Ibid., p. 352.
21. Fischer, The Metaphysics of Free Will, p. 83.
22. For a more complete discussion, see Fischer, The Metaphysics of Free Will.
I discuss the application of the structure of Dialectical Stalemates on p. 84.
23. Peter Van Inwagen, An Essay on Free Will (Oxford: Clarendon Press,
1983). Also see Carl Ginet, On Action (Cambridge: Cambridge University Press,
1990); Fischer, The Metaphysics of Free Will.
24. Fischer, The Metaphysics of Free Will, pp. 67 and 70.
25. Ibid., pp. 67–76.
26. For a discussion of the issues related to the Fixity of the Past, see Fischer,
The Metaphysics of Free Will, pp. 78–86. I discuss the Dialectical Stalemates rel-
evant to the modal principle, where the relevant modality is “nonresponsibility,”
in John Martin Fischer, “The Transfer of Nonresponsibility,” in J. Campbell,

M. O’Rourke and D. Shier (eds.), Freedom and Determinism (Cambridge: The
MIT Press, 2004), pp. 189–209.
27. Fischer, The Metaphysics of Free Will.
28. For a helpful discussion of such arguments, see Anthony Brueckner,
“Skepticism and Epistemic Closure,” Philosophical Topics 13 (1985), pp. 89–118.
Also, see Fischer, The Metaphysics of Free Will, pp. 23–45.
29. This principle was first formulated and discussed by Peter Van Inwagen
(see Peter Van Inwagen, An Essay on Free Will, pp. 182–188).
30. For an interesting discussion, see David Widerker, “Farewell to the
Direct Argument,” The Journal of Philosophy 99 (2002), pp. 316–324.
31. Mark Ravizza, “Semicompatibilism and the Transfer of Nonresponsibility,”
Philosophical Studies 75 (1994), pp. 61–93; and John Martin Fischer and Mark
Ravizza, Responsibility and Control (Cambridge: Cambridge University Press, 1994),
pp. 151–169.
of Desire,” pp. 811–812.
36. Ibid., p. 812.
37. Ibid., p. 811.
38. Thomas Nagel, The View from Nowhere (Oxford: Oxford University Press,
1986), p. 226.
39. Richard Lewis, The Other Great Depression (New York: Plume Books
[Penguin], 2002), pp. 244–245. Lewis goes on to write:
I have this gut feeling that if I made a “live” appearance it would be a friendly
and cool visitation and everyone would have a sense of calm about it. Everyone
except former “dates from hell” who are still actresses and see my demise as a
great opportunity to showcase. As I’ve mentioned before, if I don’t settle down
in a good relationship but instead drop dead while still adolescently dating
much younger women, I’m certain that my memorial service will mean noth-
ing more to these vixens than a golden opportunity to display some histrionics
(after catching a glimpse of some industry heavies in the synagogue) with the
hopes of turning some heads (not mine anymore) and getting considered for a
future role. After first feigning screams to plan a seed for future auditions for a
potential horror-flick part, they would then, apparently out of the blue, go into
powerful, well-rehearsed monologues from some Mamet play while paying lip
service to my death by frivolously changing a few words here and there to refer
to things that clearly come from my life–like too much masturbation and self-
pity, my limitless quantity of neediness, or my sickening, debilitating habit of
believing that I never did enough for the jerky people who knew how to make
me feel guilty and worthless if I didn’t go to bat for them (p. 245).
40. David Sedaris, Barrel Fever (Boston: Little Brown and Company, 1994),
pp. 18–21.
41. Fischer (ed.), The Metaphysics of Death, pp. 29–30.
42. Nussbaum, The Therapy of Desire, p. 226.
43. Ibid., pp. 227 and 228.

45. Bernard Williams, Problems of the Self (Cambridge, Cambridge University
Press, 1973), pp. 82–100.
46. I explore these issues in greater depth in chapter 9.
47. I benefited from reading a truncated version of this essay at the
University of Buffalo “Conference on Metaphysics and Medicine,” November
2004, organized by Barry Smith. I also discussed the essay with members of
David Hershenov’s University of Buffalo philosophy department graduate
seminar; I am grateful for their thoughtful comments. I have also read ver-
sions of this essay at the philosophy departments at Duke University, The John
Hopkins University, and Washington University in St. Louis. I am very grateful
to thoughtful comments on these occasions.
8
Stories
John Martin Fischer
Stories are those parts of our memories

we can afford to reveal.
Garrison Keillor, “A Prairie Home Companion”
I. VIRTUE AND COMMUNITY IN LE NIÈVE

Recently, at a conference in California, Professor Jules Levain (University
of Paris) gave an interesting talk on the relationship between virtue
and community.1 More specifically, he wished to distinguish various
claims about the relationship between virtue and community and to
evaluate some of them. Two particularly salient claims discussed by
Levain were that community sustains virtue and that community gives
content to virtue. That is to say, Levain wished (among other things)
to evaluate the claims that certain sorts of communities perform cru-
cial and indispensable sustaining and content-providing functions for
various virtues.
Whereas it is relatively clear what it means to suggest that a com-
munity has the function of sustaining certain virtues, it might be use-
ful to say a few words about the alleged content-providing function of
communities. The point here is that forms of communal life “fill in the
detailed prescriptions that make abstract principles into a lived moral-
ity.” As Levain put it,
communities tell us how to gear our general moral principles into a
complex world; without them we would not know what our prin-
ciples bid of us in the particular contexts of social life. . . . For exam-
ple, we might hold to the general principle of respect for others; but
it is only a particular form of communal life which tells us what it
129
is to respect others, how one does this in different contexts, what

the different forms of respect are, and the like. It is only by living
within a complex form of communal life that we can learn these
particularities.
One of the most interesting features of Levain’s presentation was his
extensive use of facts about a town in France during the Second World
War to substantiate the two claims about the relationship between vir-
tue and community. He pointed out that the village of Le Chambon, a
French Huguenot enclave which during the Nazi occupation of France
sheltered about five thousand refugees (mostly Jewish), is often cited
and discussed.2 But Levain claimed that there was another village in
the vicinity of Le Chambon—Le Niève—in which the citizens were also
actively involved in sheltering refugees from the Nazis. Recent work by
a French historian has brought to light some of this activity, and Levain
employed this evidence to good effect.
Levain pointed out that the rescue enterprise in Le Niève had a collec-
tive nature; the community as a whole had an impact on the decisions of
its members to engage in rescue. This influence of the community was
felt despite the fact that the efforts were not explicitly organized, and
many of the decisions to rescue were made by individuals and families
acting outside any organized structure. As Professor Levain put it,
A refugee would show up at someone’s door, or someone would
be asked if she would put one up, and she would have to decide
there and then whether to help out. Sometimes the refugee would
be brought by another villager, who either had no room or no more
room for refugees, or whom the Nazi or Vichy officials had come too
close to discovering. The person asked, or presented with a refugee,
would not always have to put the refugee up at her own house. She
could search for other shelter for the refugee. Other tasks were also
essential—getting false papers, sometimes smuggling persons out of
the village into Switzerland. Simply helping to keep up the facade
that nothing worthy of the Nazis’ attention was going on was a task
shared by all. Yet engaging in any of these activities was very risky.
It was a punishable offense to shelter Jews or to help to do so and in
fact three of the town’s leaders were jailed for a time for engaging in
these activities.
Levain argued that the virtuous way in which individual citizens of
Le Niève made their choices was deeply affected by the community in
which they lived. He adduced the recent research of the historian, who
did extensive interviews of surviving citizens of Le Niève. These citi-
zens did not see their actions and practices as having been supereroga-
tory. When pressed to say why they tried to help the refugees they said,
“It was simply what one had to do,” “She was standing at my door;
Stories 131
how could I fail to help?” and so forth. Levain analyzed the evidence
as follows:
the role of community in sustaining the virtuous activities becomes
clearer. For the fact that others were engaging in, helping, and sup-
porting the rescue activities—even though this was seldom directly
discussed—is part of what helped everyone to define them as “every-
day” and unremarkable. In a context in which one is doing what one
regards as the right thing, yet no others are joining one in this effort,
it is much more difficult to sustain a sense that one is doing precisely
what can simply be expected from anyone and is nothing remark-
able or noteworthy.
Thus, Levain pointed out that the evidence about the people of Le
Niève supported both the claims about the sustaining and also the con-
tent-providing role of community.
In the discussion that followed the talk, many of the participants
praised Levain’s use of a “real-life” example. Indeed, just about every-
one appeared to think it extremely important that the evidence came
from an actual historical example rather than a mere hypothetical case.
Too often, it was noted, philosophers simply make up stories—evi-
dently, “out of thin air”—and precipitously generate conclusions from
them. But why have any confidence that these made-up scenarios can
reveal anything useful about the real world? Further, there was general
consensus that the evidence adduced by Levain did indeed provide
considerable support for the pertinent claims about the relationship
between virtue and community.
At the end of the discussion period, Levain said, “I thank all of you
for this illuminating conversation. I have learned much that will be of
benefit to me in my thinking about these issues in the future. I should
however mention one final point: as far as I know, there is no village of
Le Niève, and no French historian who has unearthed evidence about
it. I simply made up the story.”
At the reception following the talk, there was much distress and ani-
mated conversation. Even the brie and chardonnay could not diminish
the anger and mortification of many of those who had attended the
talk; they felt duped and cheated. And no doubt they were justified in
these feelings. But some at the reception began to wonder: Did it mat-
ter at all to the evaluation of the claims about virtue and community
that Levain had simply made up the example rather than employing
an actual example? (Would it have made any difference to the evalu-
ation of the relevant claims about the relationship between virtue and
community if Levain had used true evidence from Le Chambon rather
than Le Niève? Would it have made any difference if Levain had used
evidence which he falsely—although with justification—believed to be
true of Le Chambon?) Doesn’t the moral force of an example depend
on its “intrinsic” features—its content—rather than on features of its

“genesis”? (Would it have made any difference if Levain had begun by
stating that the town of Le Niève and its inhabitants are the focus of a
recently unearthed novel by a great French novelist?)
Indeed, one of the members of the audience found the professor’s
duplicity highly enlightening (although no doubt regrettable in certain
respects). So enlightening, in fact, that he said to himself that it would
be useful to present the story of the professor’s talk as if it had actually
occurred, even if it had not.3
II. HYPOTHETICAL AND STREAMLINED EXAMPLES
II.1. The Criticism

The story of Professor Levain might prompt one to reflect on certain
aspects of the role of hypothetical examples in philosophy. In this essay,
I shall be particularly interested in the role of hypothetical and sche-
matized or abstract examples in ethics. There are of course many ways
of thinking about ethical problems and many different approaches
to doing ethics. One traditionally powerful (but by no means uncon-
troversial) approach to ethics rather liberally employs what might be
called “streamlined” hypothetical examples. These are schematized
hypothetical scenarios in which only a few details are filled in, and all
the other details are left out.4
Although this methodology has many proponents, it also has its
vigorous critics. In various contexts (although not perhaps as explicitly
and frequently in print) one encounters criticisms of the use of hypo-
thetical examples in ethics.5 Sometimes the criticism takes the form of
mere disdain, but on other occasions it is articulated with a bit more
precision. I wish to attempt to understand this kind of criticism and to
do some preliminary work toward defending the use of streamlined
hypothetical examples in ethics.
It is striking that although many philosophers object—viscerally if not
intellectually—to the use of streamlined hypothetical examples in ethics,
they rarely offer explicit reasons for their view. The reasons they offer (when
they offer reasons at all) are multifarious. Notoriously, Elizabeth Anscombe
objects to the use of such examples in her classic essay, “Modern Moral
Philosophy.”6 She criticizes philosophers who employ these examples on
the grounds that they help to “corrupt our youth” and make them less
likely to accept the absolute moral strictures of religious dogma. According
to Anscombe, these hypothetical examples—in which graphic moral con-
flicts and dilemmas are represented—help to foster the false impression
that God would allow such situations to occur frequently and one might
have to depart from the clear rules given by the church. In her essay, “Does
Oxford Moral Philosophy Corrupt the Youth?” Anscombe says:
Stories 133
A point of method I would recommend to the corrupter of the youth

would be this: concentrate on examples which are either banal: you
have promised to return a book, but . . . and so on, or fantastic: what
you ought to do if you had to move forward, and stepping with your
right foot meant killing twenty-five young men, while stepping with
your left foot would kill fifty drooling old ones. (Obviously the right
thing to do would be to jump and polish off the lot.)7
Similarly, Peter Geach says:
Spiritual writers warn us that the devil often suggests to us the ques-
tion of what we ought to do in circumstances that have not arisen
and may never arise, and tries to hurry us into a wrong hypothetical
decision: he thus may have all the satisfaction of leading us to a mor-
ally sinful intention, without any of the trouble of contriving circum-
stances in which an actual decision would be unavoidable; and we
ought to foil his malice and cunning by giving no other answer to the
question he presses upon us than “I will do as the Lord wills,” and
by trusting in God that in an actual case we shall be given the grace
to decide rightly.8
Now it would be convenient for my purposes to suppose that all the

critiques of streamlined hypothetical examples were as implausible
as these, or equally dependent on very special and not widely shared
assumptions. But this is clearly not the case, and it is important to
attempt to crystallize some more plausible versions of the critical view.
It should be evident from Levain’s story about Le Niève that the criti-
cism of such examples cannot be merely that they are hypothetical. The
story of Professor Levain helps to show that the force and relevance of
hypothetical examples depends on their “intrinsic content” and not on
their veracity or their etiology (for instance, whether they are concocted
by a philosopher or borrowed, say, from a novel).9 While holding fixed
the intrinsic content of these stories but varying their veracity or etiol-
ogy one gets no changes in the relevance of the stories to the ethical
problems under consideration.
But the primary thrust of the skepticism about the use of the typi-
cal hypothetical examples in ethics is not based on the fact that these
stories are not true. These examples are both hypothetical and stream-
lined. It seems to me that the most important basis of criticism is that
the examples are streamlined. And I think the brunt of the criticism
is based on the idea that schematized and abstract examples—that is,
streamlined examples—are not realistic.
But while it is clear to me that this is the kernel of the critique of
streamlined hypothetical examples, it is not yet clear exactly what the
critique consists in and amounts to. Of course, insofar as a streamlined
example lacks robust details, it is to that extent unrealistic. But why
exactly does this matter? Why precisely is it important for an example

to be in this way realistic? In a number of insightful essays, Martha
Nussbaum has developed interesting criticisms of the use of hypotheti-
cal examples (of a certain kind) in ethics.10 Further, she has argued in
favor of turning to novels (by certain authors) for ethical insight. In
seeking to understand the criticism of streamlined examples, it will be
helpful here to lay out some of Nussbaum’s main points.11
Nussbaum argues in favor of reading not just novels but other works
of literature, saying that
many serious dramas will be pertinent as well, and some biog-
raphies and histories—so long as these are written in a style that
gives sufficient attention to particularity and emotion, and so long
as they involve their readers in relevant activities of searching and
feeling. . . . 12
She goes on to say:
But the philosopher is likely to be less troubled by these questions of
literary genre than by a prior question: namely, why a literary work
at all? Why can’t we investigate everything we want to investigate
by using complex examples of the sort that moral philosophers are
very good at inventing? In reply, we must insist that the philosopher
who asks this question cannot have been convinced by the argument
so far about the intimate connection between literary form and ethi-
cal content. Schematic philosophers’ examples almost always lack
the particularity, the emotive appeal, the absorbing plottedness, the
variety and indeterminacy, of good fiction; they lack, too, good fic-
tion’s way of making the reader a participant and a friend. . . .
We can add that examples, setting things up schematically, signal
to the readers what they should notice and find relevant. They hand
them the ethically salient descriptions. This means that much of the
ethical work is already done, the result is “cooked.” The novels are
more open-ended, showing the reader what it is to search for the
appropriate description and why that search matters. . . . [B]y show-
ing the mystery and indeterminacy of “our actual adventure,” they
characterize life more richly and truly—indeed, more precisely—
than an example lacking those features ever could; and they engen-
der in the reader a type of ethical work more appropriate for life.13
II.2. Preliminary Points

Before going on to analyze Nussbaum’s main point, I wish to pause to
note a few things. First, being realistic implies a certain sort of richness
of detail, but of course mere richness of detail does not imply that a sce-
nario is realistic. Thus, there are various different ways of failing to be
realistic. An example may not be realistic simply in virtue of not being
Stories 135
very detailed; such an example might not be “far-fetched” or “fantastic”

(to use Anscombe’s term), and thus may be “realistic as far as it goes.”
Nevertheless, this sort of example would still be unrealistic in the sense
of lacking the richness of detail present in reality. Of course, another
way of failing to be realistic is to be far-fetched or “fantastic”; and this
is quite compatible with richness of detail. Note that Nussbaum does
not recommend careful perusal of the works of Robert Heinlein, for
example; she prefers Henry James and Marcel Proust!
Second, one might wonder why we should turn to literature for illu-
mination at all; that is, why not simply use real life? Montesquieu and
Voltaire held that history is philosophy teaching with examples. Why
not simply employ life itself (or full depictions of it) rather than literary
works? To this Nussbaum responds:
One obvious answer was suggested already by Aristotle: we have
never lived enough. Our experience is, without fiction, too confined
and too parochial. Literature extends it. . . . We can clarify and extend
this point by emphasizing that novels do not function, inside this
account, as pieces of “raw” life: they are a close and careful inter-
pretative description. . . . The point is that in the activity of literary
imagining we are led to imagine and describe with greater preci-
sion, focusing our attention on each word, feeling each event more
keenly—whereas much of actual life goes by without that height-
ened awareness, and is thus, in a certain sense, not fully or thor-
oughly lived.14
A rather extreme version of this view about the importance of litera-
ture—a view not endorsed by Nussbaum—is held by the Hindu sage in
the following very brief story. A great Hindu sage lived in a remote cave
and offered advice to people who flocked to him for his wise counsel.
When asked how he could give such good advice even though he never
leaves his cave, he responded, “Every evening I read the Ramayana,
and that is all I need to know.” But one need not hold such an extreme
position to maintain the basic plausibility of Nussbaum’s response.
II.3. Reply to the Criticism

These preliminary points having been made, let us return to the criti-
cism of streamlined examples based on their purportedly unrealistic
character. Clearly, ethical reflection and practical reasoning—thinking
about morality in general, thinking about particular moral problems,
teaching morality, and so forth—is extremely complex and has many
different aspects. I cannot here attempt to construct a general account
of such reflection and deliberation, but consideration of Nussbaum’s
remarks can help one to distinguish some of the aspects in a fashion
that illuminates the criticism of streamlined examples. I now wish to try
to distinguish at least three different facets of ethical reflection.
Undeniably, part of ethical reflection involves enhancing our capaci-

ties to see what is morally salient and to respond sensitively and appro-
priately to it. These aspects of ethical reflection involve seeking to refine
and deepen our capacity for moral perception and moral response.
The moral perception aspect of ethical reflection is itself complex
and multifarious.15 But I can say a few words to indicate part of what
moral perception involves. We often face highly nuanced, complicated,
and ambiguous situations in which we must first identify the ethically
relevant features. Before we can properly decide how to respond or
what to do, we must—either explicitly or implicitly—see what is mor-
ally relevant and important. As we confront the situation, we “frame
the moral issues”: we give a structure to a moral situation, identifying
and making salient (admittedly sometimes in an inchoate fashion) the
features that are relevant to a moral assessment and to an appropriate
response.16
Having identified what is morally relevant, we are in a position to
begin to make a suitable response, perhaps (but not exclusively) by
applying or relating general principles to the particular elements we
have identified as relevant. The second aspect of ethical reflection—the
moral response aspect—involves seeking to improve our abilities to make
acceptable moral judgments (given our background ethical theories or
moral orientations), our propensities to have emotional responses which
fit with these judgments, and our capacities to act in accordance with
these emotional responses and judgments. Clearly, in this aspect of moral
reflection, we seek to educate ourselves and develop our characters and
emotions in light of our general ethical theories or orientations.
Another aspect of ethical reflection (not emphasized by Nussbaum,
but of course compatible with her discussion) is moral analysis. The
moral analysis aspect involves seeking to generate, test, and refine our
general principles and theories—or, if one wishes to eschew the “the-
ory approach,” one’s general moral orientation. The picture here is that
there are times when we step back from our moral lives and engage
in critical reflection on it; this reflection attempts to systematize and
order the situations as given by our capacity for moral perception. Here
the features identified and described by moral perception are evaluated
and may be combined into more general principles and theories that
give expression to our general moral outlook.
Let me emphasize that these are brief and very incomplete charac-
terizations of the aspects of our moral lives. Further, I again state that
these are only some of the many components of our ethical reflection
and practical reasoning. Further, in practice these aspects do not exist in
airtight compartments; there are considerable overlaps and interlacings
of the various aspects. But, nevertheless, the preliminary distinctions
I have made can help us to see what is correct in the criticism of stream-
lined examples, and what is incorrect.
Stories 137
I suggest that reading novels and other works of fiction fits nicely
and naturally with the purposes of the first and second aspects of
ethical reflection: enhancing the capacities of moral perception and
response.17 Nussbaum is correct to point to the importance of these
aspects of morality and to the relevance of fiction (of certain sorts) to
them. Further, it seems to me that philosophers’ streamlined examples
fit most naturally with the purposes of the third aspect: analysis. Here,
one primary task is to test the generality and plausibility of principles
which “latch onto” or embody features deemed morally relevant. By
distinguishing the various aspects of ethical reflection, one can give an
appealing account of the role of different sources of illumination; one
can see how both fiction (i.e., rich stories) and philosophers’ examples
(streamlined stories) can play important roles (although at different
points in ethical reflection).
Further, perhaps one can see why certain critiques of the use of sche-
matized examples in ethics are unfair (or at least misguided). Of course,
if one wishes to enhance and improve one’s capacities of moral per-
ception and response, the abstract and schematized examples will by
and large be inappropriate.18 If one is seeking to improve the capacity
to pick out and describe the morally relevant features of highly com-
plex situations, a schematized example in which one is “handed the
ethically salient description” probably will not be useful. Further, such
an example may not be particularly helpful in enhancing one’s emo-
tional responsiveness and sensitivity. But to infer from these facts that
such examples can play no useful role in practical reflection is unfair;
it illicitly presupposes that the first two aspects exhaust the domain.19
My suggestion, then, is that the complete dismissal of schematized, hypo-
thetical examples in ethics is too abrupt; it may issue from an exclusive
focus on the first two aspects of practical reasoning at the expense of
the third.
II.4. Further Challenges

I have suggested that the view that streamlined examples are com-
pletely inappropriate in ethics is incorrect because it wrongly leaves
out an aspect of our moral lives: the aspect of analysis. But I now wish
briefly to consider some further challenges to the legitimacy of stream-
lined examples. Someone might press the point that precisely because
streamlined examples are unrealistic, they are even inappropriate to the
aspect of analysis.
First Version of the Worry. The following version of this worry is
indeed suggested by Nussbaum. Unless our emotions are engaged in
a certain way by an example, our moral judgments about it are not
to be trusted, and streamlined examples simply do not adequately
engage our emotions. If this is true, then even if we set aside the aspects
of moral perception and response, there will still be a problem with
employing schematized examples for the more narrow task of evaluat-

ing a proposed moral principle or analysis.20
This view raises deep and difficult questions which are intractable to
quick resolution. It is not absolutely clear to me that Nussbaum’s the-
sis about the necessity of emotional engagement for trustworthy judg-
ment is actually true. While it has some plausibility, we sometimes trust
our moral judgments only when they are made from a cooler, more
detached, less emotionally engaged, perspective. So Nussbaum’s thesis
here may not be true in its general form. Further, even if it is true, it can
be accommodated compatibly with holding that streamlined examples
have a legitimate place in moral reflection.
To explain: Nussbaum’s critique of the use of hypothetical and
streamlined examples in ethics rests not only on the thesis of the neces-
sity of emotional engagement for trustworthy moral judgment, but also
on what might be called an “atomistic” picture of emotional engage-
ment. And I believe it is highly implausible to think that, one’s emotions
having been engaged, they suddenly disengage when one’s attention is
focused on a streamlined example. Imagine that one is in the process of
ethical reflection on a given topic. Suppose one fully immerses oneself
in relevant fiction (Henry James, Proust, and so forth) which appro-
priately engages one’s emotions. Having identified salient ethical fea-
tures and generated appropriate principles, one turns to the assessment
and testing of those principles—and here one uses certain streamlined
examples (as well as other ingredients). Why suppose that in this pro-
cess somehow the emotions suddenly “turn off” or disengage when
considering the streamlined examples; is emotional engagement this
atomistic and episodic?
It seems to me that a more holistic and less atomistic conception of
emotional engagement is closer to the truth about us. Surely, our emo-
tional investment in and engagement with other persons does not work
in a way analogous to the episodic view apparently presupposed by
Nussbaum’s criticism; it is a sometimes painful feature of our lives that
we do not and cannot simply disengage from other persons when they
are absent. And it is also perhaps a comforting feature of us that our
emotional attachments are not so adventitious. If a less episodic con-
ception of emotional engagement is accurate, one can hold the thesis of
the necessity of emotional engagement compatibly with giving a place
to streamlined examples in the analysis component of moral reflection.
Second Version of the Worry. There is another version of the worry that
precisely because streamlined examples are unrealistic, they are inap-
propriate to the aspect of analysis. The idea is this. A streamlined exam-
ple presents us with a very special, artificial, and rarified context. If a
conclusion is drawn about this sort of context, it does not necessarily
transfer straightforwardly to the real world. As a crude example, sup-
pose you strike a match in a room with no gas in it, and you note that
Stories 139
striking a match there does not cause an explosion. You then proceed to
strike a match in a different room—one with lots of gas in it. Needless
to say, you would have made a rather unfortunate mistake—the mis-
take of failing to see that striking a match has different consequences
depending on the other factors that are present in the given context. A
given event may have one meaning in one context (and against a given
set of background conditions), and quite another meaning in another
context (and against another set of background conditions).21 And so a
given moral event or factor may have one moral meaning in a stream-
lined example, and quite another meaning in reality.
I wish to grant the basic insight behind this criticism, but still maintain
that streamlined examples have a place in analysis. The objection shows
that one must be very careful to consider the various possibilities of
“meaning shift” in doing one’s analysis, but it does not show that stream-
lined examples are worthless or inappropriate. Streamlined examples,
even when they cannot be employed to achieve closure, can be highly
suggestive; they can establish strong presumptions and can suggest further
examples in which the factor in question is tested against various back-
grounds, or in which clusters of potentially synergistic factors are tested.
I believe that the meaning shift problem is the most significant reason
why one might be tempted to say that streamlined examples ought to be
avoided precisely because they are unrealistic. But I have suggested—in
an admittedly brief and tentative way—that even this possibility leaves
room for legitimate use of streamlined examples. And it is very impor-
tant to keep in mind the potential benefits of employing streamlined
examples; these need to be weighed against the risks. A primary goal of
abstraction and schematization in moral reflection is to create the ana-
logue of “controlled experiments” in science: one wants to hold all other
factors fixed, and test one particular factor for ethical relevance.22
The fruits of controlled experimentation in ethics, as in the sciences,
can be quite significant. And it must be kept in mind that this procedure
does not in itself imply that individual morally relevant factors combine
“additively” and “non-synergistically” to determine an overall moral
assessment of a situation, just as the procedure in the sciences does
not imply that striking a match will not have different consequences,
depending on the background conditions. The procedure of abstraction
and schematization—of “streamlining,” if you will—can assist in deter-
mining whether a factor is morally relevant; but how this factor combines
with others is left open, and it is useful to keep in mind the possibility of
significant interactions, synergisms, and “meaning shifts.”
III. CONCLUSION
My purpose has been to seek to find a place for philosophers’ stream-
lined examples—hypothetical and schematized—in moral reasoning.
I have attempted to accommodate the insights of such philosophers as

Nussbaum about the role of reading fiction and using less schematized
(and in this respect more “realistic”) examples. I have most definitely
not argued that one should exclusively employ such examples in ethics;
rather, I have emphasized that practical reasoning and ethical reflec-
tion has many aspects in which various sorts of examples are appropri-
ate. I have attempted to explain why I believe that there are legitimate
and appropriate uses of streamlined examples in ethical reflection.23
Whereas this may seem to be a fairly weak thesis, it will provide some
comfort to those who have been rebuffed by the proponents of the
thesis that such examples are never appropriate and are to be shunned
completely.24
NOTES
1. Professor Levain is one of the most distinguished ethicists in France.
Although he is well known in France, he is not widely known in the United
States. It is ironic that Jacques Derrida has received so much attention here
whereas Levain remains obscure, given Levain’s stature in French academic
circles.
2. P. Hallie, Lest Innocent Blood Be Shed. Also, P. Sauvage’s film, “Weapons
of the Spirit.”
3. As far as I know, there is no Professor Jules Levain—I simply made
him up. The idea for this section was suggested to me by a talk delivered by
Lawrence Blum to the Riverside Philosophy Conference, May 1991. The sort of
duplicity portrayed in the text did not actually occur. The idea for this section
was in part suggested by interesting remarks by David Solomon in response
to Blum, and I have borrowed from Blum’s nice essay. Blum’s essay, “Virtue
and Community,” has subsequently been published in his collection of essays,
Moral Perception and Particularity (New York, 1994), 144–69.
4. The “Trolley Problem” cases, and related cases, are frequently discussed
examples of this sort, but of course there are many others. For a selection of
essays that present and discuss many such hypothetical and streamlined cases,
see Ethics: Problems and Principles, edited by John Martin Fischer and Mark
Ravizza (Fort Worth, Tex., 1992).
5. Recently Jonathan Dancy has argued for “particularism” about moral rea-
sons: Moral Reasons (Oxford, 1992). According to particularism, considerations
count as moral reasons only given their entire context (pp. 55–7, 89). For a sus-
tained critique of the use of thought experiments primarily in metaphysics, see
Kathleen V. Wilkes, Real People: Personal Identity Without Thought Experiments
(Oxford, 1988).
6. Philosophy 33 (1958): 1–19.
7. G. E. M. Anscombe, “Does Oxford Moral Philosophy Corrupt the Youth?”
The Listener, February 14, 1957, p. 267. (Subsequent issues of The Listener contain
what can only be described as vituperative correspondence about the issues
raised by this piece.) Taking a similar tack to that of Anscombe, Tom Weller
asks, “If A and B were drowning, and you could only save one of them, would
you . . . have lunch or go to a movie?” (The Book of Stupid Questions [New York,
Stories 141
1988]; as quoted in F. M. Kamm, Morality, Mortality Volume I: Death and Whom to

Save from It [New York, 1993], 75).
8. Peter Geach, in Nuclear Weapons and Christian Conscience, edited by
Walter Stern (London, 1981), 91. I am indebted to Gerald Dworkin for bringing
this passage to my attention.
9. Note that Levain’s story of Le Niève pertains to moral matters, but more
properly to descriptive features of practical reflection (rather than normative fea-
tures, strictly speaking). That is, the story is about how communities function to
sustain and give content to virtues, and these are primarily descriptive features of
moral matters. I would contend that the point developed in the text about the irrel-
evance of etiology or veracity also applies to normative features of moral matters.
Given that Levain’s story of Le Niève pertains to descriptive aspects of moral-
ity, shouldn’t the falsity of the story be relevant? I do not believe it is. Stories are
told, and examples adduced, for various purposes; if the story is told to bring
some general phenomenon to the attention of the relevant audience—some phe-
nomenon which otherwise might escape notice and for which there is considerable
independent evidence—it can be perfectly useful and appropriate even if false.
10. Martha C. Nussbaum, Love’s Knowledge: Essays on Philosophy and Literature
(New York and Oxford, 1990).
11. It is helpful to turn to Nussbaum’s work to understand the criticism of
certain uses of hypothetical examples in ethics. But I wish to emphasize that
Nussbaum herself does not argue that such examples are never appropriately
employed in ethics. Thus, Nussbaum might well agree with the overall thesis
for which I am arguing in this essay.
12. Nussbaum, Love’s Knowledge, 46.
13. Ibid., 46–7.
14. Ibid., 47.
15. For a very useful discussion, see Lawrence Blum, “Moral Perception
and Particularity,” Ethics 101 (1991): 701–25; and various essays in Blum, Moral
Perception and Particularity.
16. Barbara Herman discusses these issues within a Kantian framework in
“The Practice of Moral Judgment,” Journal of Philosophy 82 (1985): 414–36.
17. For a similar point, see Michael DePaul, “Argument and Perception: The
Role of Literature in Moral Inquiry,” Journal of Philosophy 85 (1988): 552–65.
18. One might usefully distinguish two sorts of capacities which are rel-
evant here. First, there is the capacity to generate factors which are relevant to
moral assessment; to this capacity the streamlined stories may indeed be rel-
evant. I owe this point to Michael Otsuka. But given a set of factors which an
agent believes may be relevant, there is the capacity to look at a complex and
ambiguous situation and apply the possibly relevant factors in order to identify
the particular factors relevant to the situation. To this capacity the abstract and
schematized examples will be largely irrelevant, and this is the sort of capacity
on which I focus in the text.
19. I should emphasize that this inference is not made by Nussbaum.
20. I am indebted to correspondence with Michael DePaul for this point.
21. For a very illuminating presentation of this sort of problem, see Shelly
Kagan, “The Additive Fallacy,” Ethics 97 (1988): 5–31.
22. In our introductory essay to Ethics: Problems and Principles, Mark Ravizza
and I pose the question of why it should be thought that ethics is relevantly
different from social sciences such as economics or linguistics with regard to the
appropriateness of the use of abstraction. We said:
An economist constructs a model which is really a hypothetical “world.”
This hypothetical world is characterized by the assumptions of the model
and is presumably considerably simpler than our world (or even the part of
our world pertinent to the model). Having constructed this simpler world,
the economist considers its properties and (ideally) generates conclusions.
These conclusions may to some extent illuminate features of our world.
Similarly, when we construct thought-experiments in ethics, we are con-
structing hypothetical worlds (albeit very small ones). These worlds are
characterized by the assumptions of the examples, and they are in many
ways simpler than real-world situations. We then scrutinize the hypothetical
worlds and (ideally) generate conclusions. Again, these conclusions may to
some extent illuminate features of our world. In ethics, hypothetical exam-
ples are very much like the models of economics.
If the techniques of model-building and abstraction are useful and worth-
while in economics, then why should these techniques not yield similar ben-
efits in ethics? (p. 47)
Although, admittedly, the issues are put rather starkly here, I do think the
question deserves serious consideration. For further reflections on these issues,
see Thought Experiments in Science and Philosophy, edited by Tamara Horowitz
and Gerald J. Massey (Lanham, Md., 1991); and Roy Sorenson, Thought
Experiments (Oxford, 1992).
23. Of course, the standard “philosophers’ examples” are not the only sche-
matized cases that are widely used in thinking about moral matters. Consider, for
example, the extensive use of parables and stories in moral education and practical
reflection quite generally, and, in particular, in the teachings and educational prac-
tices of various religions. Immediately, one thinks of the parables and stories of the
Old and New Testaments, Hasidic Tales such as the Tales of Rabbi Nachman, and
the stories of Chuang Tsu. Consider, just for fun, the following example:
Two monks, having taken a strict vow of chastity, approach a stream in which
a woman is drowning. The older monk immediately jumps in, rescues the
woman, and carries her to the shore. Then they continue their walk. Some
time later, the younger monk questions his elder, wondering whether the
action of the older monk violated his sacred vow. The older monk replies,
“See, I carried the woman for a few minutes and left her there by the side of
the stream, but you have carried her with you all this way. . . . ”
Surely, no one would deny the prevalence of such stories in moral education
and practical reflection. Nor would anyone insist that their relevance depends
in any way upon their veracity or richness of detail. If such stories have a place
in practical reflection, why couldn’t philosophers’ streamlined examples also
have a place (although a different place)?
I suppose there could be many purposes of telling the story of the monks.
One point is that one should not take certain rules or principles narrowly or
“literally,” but one should interpret them in light of their purpose. Often such
stories are useful insofar as they remind us of some point we are apt to lose track
of in ordinary life. They put things in unusual and striking ways and can help us
Stories 143
either to see issues in new ways or to remind us of important points. Although

these purposes are slightly different from the purposes of introducing stream-
lined philosophers’ examples, they share certain similarities: they in some sense
involve “meta-reflections” on the first-order considerations of morality, and to
this extent are similar to the purposes of the streamlined examples.
For the above story of the monks, I am very grateful to Mark Ravizza.
Perhaps it has special significance for him, as he told it to me a few days before
he entered a Jesuit novitiate!
24. Richard Rorty argues against the use of schematized philosophers’
examples: Essays on Heidegger and Others: Philosophical Papers, vol. 2 (Cambridge,
1991). In his essay, “Heidegger, Kundera, and Dickens” (66–82), Rorty argues
that such examples are misleading insofar as they (unlike certain examples
from literature) foster the impression that there is an “essence” of morality—a
true answer or set of answers to moral questions that can be codified and pre-
sented in some sort of theory. Rorty’s “postmodernist” anti-theory position
leads to a denial of even the role I have found for schematized abstract phi-
losophers’ examples in ethics; Rorty’s position is considerably stronger than
Nussbaum’s in this respect. Also, it should be noted that Rorty’s objection to
the use of examples is driven by his substantive view about ethics; in this way it
is not a “neutral” objection to a particular methodology.
I believe that certain other objections to the use of streamlined hypothetical
examples are also driven by particular substantive positions in ethics. For exam-
ple, a proponent of virtue ethics might object to such examples on the grounds that
they inappropriately tend to point us toward thinking of general rules as central in
ethics. Similarly, a Kantian might object to such examples to the extent that they
are not rich enough to help us to crystallize the agent’s motive and thus his or her
“maxim.” It is important to see that these sorts of objections are “theory-driven”
(or perhaps in Rorty’s case “anti-theory-driven”); they issue from a particular sub-
stantive view and are to this extent not neutral objections to a methodology. How
such objections are properly viewed is a dialectically delicate and difficult matter.
It must be conceded that it is fair to require that a particular methodology not be
significantly “tilted” toward (or against) a particular approach. But as regards this
latter point, recall that the methodology I have been defending here does not hold
that hypothetical examples should be the only or even the primary considerations
in ethical reflection. It would be difficult to argue that a methodology which sim-
ply finds some room for such examples is significantly biased against particular
approaches (which themselves have some minimal claim to plausibility).
Various friends have read pieces that have become parts of this essay and
have offered encouragement. I am very grateful to them: Howard Wettstein,
Alex Rosenberg, David Glidden, Dwight Furrow, Kerrin McMahan, Larry
Wright, Anthony Brueckner, Gerald Dworkin, Stephen Munzer, John Heil, Andy
Coats, Antonio Rauti, Michael DePaul, and Geoffrey Sayre-McCord. Also, I am
indebted to David Solomon and Lawrence Blum for their interesting discus-
sion at the Riverside Philosophy Conference, 1991, and to the UCLA Law and
Philosophy Discussion Group, especially David Copp, David Dolinko, Michael
Otsuka, and Seana Shiffrin. Finally, I owe many thanks to Mark Ravizza for
the example of the monks, for encouragement, and for numerous helpful and
enjoyable conversations
9
Free Will, Death, and Immortality:

The Role of Narrative
John Martin Fischer
Although the future is uncertain, we can at least be confident

of one thing: the past is always changing.
Slavic Folk Saying1
I. INTRODUCTION
The notion of ‘narrative’ is rich and suggestive, but, at the same
time, vexed. The notion is invoked by philosophers and literary
theorists (and others) in very different ways. Despite the confusions
engendered by its multiple meanings and uses, I believe that the
notion of narrative can be illuminating with respect to issues about
freedom, death, immortality, and the meaning of life. Care must be
taken, however, to distinguish different ideas, and to apply them
appropriately.
In previous work, I have made some tentative and sketchy sugges-
tions.2 I have claimed that the value of acting freely, or acting in such a
way as to be morally responsible, is the value of self-expression. This
value is a kind of aesthetic value, or akin to an aesthetic value. When
I act freely, I ‘make a statement,’ and the value of my free action is the
value of writing a sentence in the book of my life (my narrative), rather
than the value of ‘making a difference’ (of a certain sort) to the world.
I have not suggested that artistic self-expression is the only value, or a
hegemonic one; rather, the suggestion was that the value of free action
is the value, whatever that is, of artistic self-expression. Further, I have
suggested that our lives’ having the signature features of narrative
does not in itself imply that immortality would necessarily be undesir-
able (or even unrecognizably similar to our current lives).3
145
Here I wish to develop and tie together these ideas a bit further
(although, perhaps inevitably, not as fully as I would like). I shall begin
by laying out and discussing relevant aspects of the work of David
Velleman on these topics.4 I shall then argue that the value of our lives
as free creatures is indeed a species of the value of artistic self-expres-
sion. More specifically, I shall contend that acting freely is what makes
us the sort of creatures that live lives that have the characteristic features
of narratives. Acting freely is what changes the depictions of our lives
from mere characterizations to stories (or narratives). Further, I shall
distinguish the claim that our lives can be explained in the distinctive
manner of narrative explanation from the claim that our lives can be
evaluated in the characteristic narrative fashion. Finally, I suggest that
understanding our lives as narratives (in either sense) is compatible
with the possibility that unending life would be attractive to human
beings. Narrativity need not entail the necessity of endings.
II. VELLEMAN ON NARRATIVE VALUE

AND NARRATIVE EXPLANATION
II.1 Velleman on Narrative Value

Velleman argues that the overall value of our life, considered as a whole,
is not determined by a function that merely adds together all of the
levels of momentary well-being in our lives. Additionally, he contends
that our levels of momentary well-being are not simply the marginal
increment in our level of overall welfare or value. The two dimensions
of value—momentary and overall—are not analyzed in terms of each
other or reducible to each other.5
According to Velleman, we care not just about the total amount of
momentary well-being in our lives, or even its temporal ordering; we
care about the story of our lives, the narrative structure of our lives.6
More carefully, the function that specifies the overall value of our lives
takes into account in an important way certain characteristics that
are signature features of what we take to be ‘good’ ” (in some sense)
narratives.
As Velleman concedes, his point is not that we value narratives that
are ‘better stories’ in a sense that would be of interest to potential read-
ers of novels or (say) literary critics or reviewers. A life that makes a
rather boring story might well be preferable, in terms of overall value,
to a life filled with unexpected twists and turns, excitement, titillation,
and tragedy. So exactly what does Velleman mean when he says that
the overall value of our lives is importantly sensitive to narrative con-
tent? What are the signature features of narratives that help to shape
the functions that determine the overall value of our lives?
Free Will, Death, and Immortality: The Role of Narrative 147
I suggest that we begin by considering the examples Velleman

employs. Velleman starts with an example of Michael Slote’s:
A given man may achieve political power and, once in power, do

things of great value, after having been in the political wilderness
throughout his earlier career. He may later die while still ‘in harness’
and fully possessed of his powers, at a decent old age. By contrast,
another man may have a meteoric success in youth, attaining the
same office as the first man and also achieving much good; but then
lose power, while still young, never to regain it. Without hearing
anything more, I think our natural, immediate reaction to these
examples would be that the first man was the more fortunate . . . 7
Whereas Slote concludes from this sort of example that we have a
‘time preference for goods that come late in life,’ Velleman concludes
that we care about the narrative content of our lives (and not just the
total levels of momentary well-being, added up or even added up and
weighted according to temporal position). It is a distinctive feature
of narratives that later events can alter the ‘meaning’ or ‘significance’
of earlier events. In this sense narratives can have ‘loops,’ as Derrida
(notoriously, in some quarters) contended. It is not that we can change
the physico-causal past; but we can sometimes change its meaning and
thus its contribution to the value of our lives overall.
Velleman says:
Why would a person care about the placement of momentary goods
on the curve that maps his changing welfare? The answer, I believe,
is that an event’s place in the story of one’s life lends it a meaning
that isn’t entirely determined by its impact on one’s well-being at
the time. A particular electoral victory, providing a particular boost
to one’s current welfare, can mean either that one’s early frustra-
tions were finally over or that one’s subsequent failures were not yet
foreshadowed, that one enjoyed either fleeting good luck or lasting
success—all depending on its placement in the trend of one’s well-
being. And the event’s meaning is what determines its contribution
to the value of one’s life.
. . . The meaning of a benefit depends not only on whether it fol-
lows or precedes hardships but also on the specific narrative rela-
tion between the goods and evils involved. Slote’s politician would
have experienced an improvement in his well-being whether his
years of toil were capped by electoral victory or merely cut short by
his winning the lottery and retiring young. But the contribution of
these alternative benefits to the overall value of his life wouldn’t be
determined entirely by how well-off each would make him from one
moment to the next. Their contribution to his life’s value would also
be determined by the fact that the former would be a well-earned
reward, and would prove his struggles to have been a good invest-
ment, whereas the latter would be a windfall in relation to which his
struggles were superfluous. Thus benefits that would effect equal
improvements in his momentary well-being might contribute differ-
ently to the value of his life, by virtue of lending and borrowing dif-
ferent meanings in exchange with preceding events.8
Velleman contends that it is better to thrive as a result of learning from
one’s misfortunes than simply as a result of (say) winning the lottery or
some other windfall. As Velleman puts it:
A life in which one suffers a misfortune and then learns from it may
find one equally well-off, at each moment, as a life in which one suf-
fers a misfortune and then reads the encyclopedia. But the costs of
the misfortune are merely offset when the value of the latter life is
computed; whereas they are somehow cancelled entirely from the
accounts of the former. Or rather, neither misfortune affects the
value of one’s life just by adding costs and benefits to a cumulative
account. The effect of either misfortune on one’s life is proportionate,
not to its impact on one’s continuing welfare, but to its import for
the story. An edifying misfortune is not just offset but redeemed, by
being given a meaningful place in one’s progress through life.9
Velleman believes that the following pair of stories illustrates the same
point. In both lives your first ten years of marriage are unhappy and
are followed by equal amounts of contentment. But in the first life you
get divorced and consider your first marriage a ‘dead loss’; you just
happen to meet someone else with whom you live happily (ever after!).
In the second life, you learn from the troubles of your first ten years,
and you save the marriage (and live happily ever after). Indeed, in the
second story you think of the initial segment of the marriage as the
‘foundation of your [later] happiness.’ Velleman says that we would
prefer the second life. He says, ‘You can simply think that a dead-end
relationship blots the story of one’s life in a way that marital problems
don’t if they lead to eventual happiness.’10
So it seems that the primary examples suggested by Velleman of
salient features of desirable narratives (in the sense relevant to the func-
tion that determines overall value of a human life) are of hard work
being rewarded and learning from mistakes, rather than simply profit-
ing (comparably) from windfalls. My suggestion is that we interpret
Velleman’s claim that we take narrative content into account in evalu-
ating overall welfare as involving the idea that ‘narrative content’ is a
shorthand for a kind (or perhaps kinds) of relationship among events
of which his cases are instances. ‘Narrative content’ is not explained
generally, but it involves the sort of relationships that are present in his
cases and other, similar cases. When this sort of temporally extended
pattern of events occurs, the kind of ‘exchange of meaning’ that is char-

acteristic of narratives can take place.11
Velleman applies his claim (that human beings are distinctive insofar
as the overall value of our lives is partly a function of narrative content)
to issues pertaining to the badness of death. A decision about physician-
assisted suicide depends at least in part on whether the individual’s life
would on balance be worth continuing (on some appropriate metric).
Velleman says:
The choice between heroic medical treatment and passive eutha-
nasia is therefore frequently said to require so-called quality-of-life
considerations. Whether days should be added or subtracted from
a patient’s life is to be judged, according to the prevalent view, by
whether the days in question would be spent in a state of well-being
or hardship.
In my view, however, deciding when to die is not (despite the
familiar saying) like deciding when to cash in one’s chips—not, that
is, a decision to be based on the incremental gains and losses that one
stands to accumulate by staying in the game. It is rather like decid-
ing when and how to end a story, a decision that cannot be dictated
by considerations of momentary well-being. Hence a person may
rationally be willing to die though he can look forward to a few more
good weeks or months; and a person may rationally be unwilling to
die even though he can look forward only to continued adversity.
The rationality of the patient’s attitude depends on whether an ear-
lier or later death would make a better ending to his life story.12
II.1a Some Reflections

The idea of employing the notion of narrative as a guide in potential
cases of physician-assisted suicide is fascinating and suggestive. I think
it has limitations, though. We saw above that the pertinent notion of
‘better ending to his life story’ cannot be the notion of ‘better’ impor-
tant to a reader of a work of fiction or a literary critic. But we do not
have a general account of the relevant notion of narrative content and
thus of ‘better story’ or ‘better ending of the life-story.’ We do have a
set of examples, and we can intuitively recognize similar sequences of
events that have the requisite ‘narrative’ characteristics. But I think that
this will only provide guidance in some cases—by no means all.
Certainly we can say that the death of someone in the prime of her
life, with many ongoing professional and personal projects underway,
and no significant health problems, would be a bad ending to her story,
in the relevant sense. Similarly, we can presumably say of an elderly
woman who has had a full, rich life and has completed all her important
life-projects, has said goodbye to her loved ones and friends and made
arrangements for the disposition of her estate, and is in significant pain,

that it would not be a bad thing for her to die; death now rather than
later would be a better ending to her life story. Velleman’s notion of nar-
rative content arguably does help to sort through such examples.
But there will be many other cases in which it simply does not give
sufficiently definite guidance. What about a person who has completed
all his life-projects, but still to some extent enjoys simply watching cer-
tain television programs or reading the newspaper? He is physically
confined to a bed, but is not in intolerable or significant pain. He knows
however that he has a condition that will slowly incapacitate him fur-
ther, and he has decided that now is the time to die. Would death now
rather than in six months or a year or two years be a better ending to
his story? Are his interests in television and the newspaper sufficient
to say that death now would rob him of a ‘better’ ending to his story?
What if he did not have such interests, or any interests? That is, what
if he were largely indifferent to everything, including television and
the newspaper and even news of his family, and so forth? Would this
make a difference, and, if so, would the narrative account suggested by
Velleman entail (or help to explain) the difference in question?
The ideas that people should thrive because of their efforts, or in
virtue of learning from their mistakes, do not provide sufficient guid-
ance about the sorts of cases just sketched. Further, it is not clear how
to extrapolate or extend the ideas to provide definite guidance in these
and a whole range of cases. I would think that many (although cer-
tainly not all) cases of potential physician-assisted euthanasia pertain
to elderly individuals who have completed the major projects of their
lives, at least so far as ‘projects’ are understood in a fairly standard way.
It is already the case that either they have learned from their mistakes
or not, either they have flourished because of their efforts or not, and so
forth. These considerations don’t seem to help with the decision about
when exactly the decision to terminate life should be made, if at all. Of
course, we could try to interpret ‘narrative content’ in such a way as to
apply to anyone with ongoing interests of any kind, but this seems a
stretch; here considerations of quality of life appear to be driving our
judgments, rather than considerations of narrative content or structure.
Thus, whereas Velleman’s idea is highly suggestive, it can give only
a partial guide to the difficult decisions concerning physician-assisted
suicide.13
Velleman also applies his insights about the significance of narrative
content to the question of the distinctive badness of the death of a per-
son. A human being can see his life as an extended sequence over time;
also, he can see it as a story. A human being can thus care about the nar-
rative structure of his life. According to Velleman, a nonhuman animal,
such as a cow, cannot conceive of itself as a continuing entity—cannot
even take an extended temporal perspective at all. What the animal
cannot conceive it cannot care about. As Velleman puts it, ‘ . . . a person

can care about what his life story is like, and a premature death can
spoil the story of his life. Hence death can harm a person but it cannot
harm a cow.’14
Velleman believes that, because a nonhuman animal cannot take
a temporally extended perspective, it is incoherent to say that one
sequence of moments can be better for a cow than another sequence of
moments. He says, ‘For a lower animal, then, momentary well-being
fails not only of additivity but of cumulability by any algorithm at all.
Consequently, the totality of this subject’s life simply has no value for
him, because he cannot care about it as such . . . ’15
It is not clear to me that value (and harm) should be tied so closely
to a subject’s ability to conceive of the relevant thing. I do not know
how to adjudicate this sort of issue. But I would simply point out that
even if we granted that a cow’s death deprives the cow of a sequence
of future experiences (which on balance are positive), this is not the
distinctive sort of value of which death can deprive a human being.
That is, death can be bad for a person in a special way: it can rob a per-
son of a good ending to his life-story, whereas it cannot so rob a cow.
The lives of human beings (or, more carefully, persons) have a dimen-
sion of value over and above the accumulation of momentary well-
being, whereas at best this sort of accumulation exhausts the value of
a cow’s life. Thus death can be bad for a person in a distinctive way,
even if we grant that there is a cumulative dimension to the value of a
nonhuman animal’s life.
Note that Velleman contends that the capacity to adopt a temporal
perspective from which one can see an extended period of time is nec-
essary for the capacity to care about the story of one’s life. Thus, on
Velleman’s view, such a capacity is a necessary condition for the overall
value of humans’ lives being affected by narrative considerations.16 I do
not however believe that the capacity to take this sort of perspective is
sufficient. Consider a human being who is being thoroughly controlled
by remote-control direct stimulation of the brain, so that his choices
and actions are intuitively not his own free choices and actions. This
individual may well have the normal capacity to take a more expansive
temporal perspective, but the fact that he never acts freely inclines me
to say that his overall level of well-being is to be determined by sim-
ply adding his momentary welfare values. For an individual who does
not act freely, there is no separate dimension of value—the narrative
dimension—irreducible to the intertemporal aggregation of momen-
tary welfare. Such an individual would be ‘in between’ mere animals
(on Velleman’s account) and ordinary human beings (and other per-
sons). That is, he would have the capacity to take the relevant tempo-
ral perspective, and yet not act freely. Thus, unlike mere animals (on
Velleman’s account), his momentary welfare values could be added
together; but unlike ordinary human beings, he would not have a sepa-
rate narrative dimension of value.
If I am correct about this supposition, then acting freely is the ingre-
dient that gives us the distinctively narrative dimension of value.
Acting freely is the feature which, when added to the others (including
the capacity to take the appropriate temporal point of view and thus to
care about one’s story), transforms us into creatures whose lives can be
evaluated by reference to salient sorts of narrative relationships.17
It is typically held that only a free creature—only a creature capa-
ble of acting freely—can have a meaningful life. It emerges now that
acting freely is the specific ingredient that endows our lives with the
distinctively narrative dimension of value. Only a creature who can act
freely can affect the ‘meanings’ of past events by virtue of affecting the
narrative relationships among various events in his life. Death robs a
person of something especially important—the capacity to continue to
lead a meaningful life. That is, it can now be seen that it robs a person
of the capacity to continue to lead a life with a narrative dimension of
value, and thus with a specific sort of meaning. It is sometimes said
that death is bad because it deprives us of possible future goods. On
the view I have sketched, it is also bad because it can deprive us of the
possibility of changing the narrative meaning of the past.
On this sort of view (to which I am attracted), one can disagree with
Velleman’s claim that death cannot be bad for a cow, but still maintain
that the death of a person is bad in a special way. Death arguably can be
bad for the cow insofar as it deprives the cow of future pleasures (plea-
sures that would be part of an extended sequence that overall has con-
siderably more pleasure than pain). But death can be bad for a person
insofar as it deprives him of past goods as well as future goods: it cuts off
the accumulation of momentary well-being, and it can prevent us from
writing a better ending to our story (and thus vindicating our pasts).18
I follow Carl Ginet in holding that our freedom is the freedom to
add to the given past, holding the laws of nature fixed.19 This sort of
‘fixity-of-the past’ view pertains to the physico-causal events and fea-
tures of the past, and not to their ‘meanings.’ This constraint applies to
both the notion of ‘freedom to do otherwise,’ and the notion of ‘acting
freely.’ Whereas we cannot go backward in physico-causal space-time
and ‘change the past,’ we can readily go backward in narrative space-
time. Whereas it is a constraint on our freedom that the physico-causal
past be fixed, and that our actions be extensions of the given (physico-
causal) past, it is precisely our capacity to act freely that provides the
ingredient that allows for backward travel in narrative space-time.20
II.2. Velleman on Narrative Explanation

In a fascinating recent article, David Velleman has sought to give an
account of narrative explanation (which, of course, is distinct from nar-
rative value).21 As Velleman puts it, ‘A story does more than recount
events; it recounts events in a way that renders them intelligible, thus
conveying not just information but also understanding.’22 Velleman
seeks to describe the distinctive explanatory force of narrative, and to
distinguish narrative explanation from explanation in the social and
natural sciences.
Velleman says:
This question arises for various disciplines in which narrative
comes into play. For historians, it is the question whether narrating
historical events conveys understanding over and above that con-
veyed by subsuming the same events under the generalizations of
economics, political science, or sociology. For clinical psychologists,
it is the question whether fitting symptomatic behaviours into a life-
story adds to the understanding gained by fitting them into diagnos-
tic categories. Even the police or the jury must ask themselves what
sort of explanatory value there is in a suspect’s giving his alibi in the
form of a story.23
In providing his account of the distinctive potency of narrative explana-
tion, Velleman builds on the work of Louis Mink and W.B. Gallie.24 Both
of these theorists emphasize the importance of characterizing events in
terms of their relations to outcomes or ‘endings’. Velleman says:
A narrative must move forward not only in the sense of telling one
event after another but also in the sense of approaching or at least
seeming to approach some conclusion to those events, some termi-
nus, finish, or closure.
Here I should elaborate on a point . . . about the difference between
narrative and the artistic genres that employ it. A novel or a theater
piece need not reach a conclusion or even seem to approach one. But
a novel or a theatre piece need not be a work of narrative, either; it
may be a work of narrative only in parts, or it may be ‘of’ narrative
only in the sense of commenting on the requirements of narrative
only by pointedly defying them. A bad story can make for a great
novel (though perhaps not the sort of great novel that one likes to
read). The necessity of an ending is not inherent in the aesthetics of
the novel or play but in the nature of storytelling, a form of discourse
that a novel or play need not employ.25
On Velleman’s view, a narrative explains by allowing the audience
to assimilate the events in the story to a familiar emotional pattern or
‘cadence.’ He says:
A story therefore enables its audience to assimilate events, not to
familiar patterns of how things happen, but rather to familiar pat-
terns of how things feel. These patterns are not themselves stored
in discursive form, as scenarios or stories; they are stored rather in

experiential, proprioceptive, and kinesthetic memory—as we might
say in the muscle-memory of the heart. Although the audience may
have no discursive memory of events such as those of the story, it
nevertheless has an experience of deja senti, because its emotional
sensibility naturally follows the ups and downs of the story, just as a
muscle naturally follows the cycle of tension and release.
What’s more, the emotion that resolves a narrative cadence tends
to subsume the emotions that preceded it; the triumph felt at a happy
ending is the triumph of ambitions realized and anxieties allayed;
the grief felt at a tragic ending is the grief of hopes dashed or loves
denied. Hence the conclusory emotion in a narrative cadence embod-
ies not just how the audience feels about the ending; it embodies
how the audience feels, at the ending, about the whole story.26
So, on Velleman’s account, narrative explains by getting the audience
to understand the relevant events because it ‘knows how they feel,’ in the
sense that it experiences them as leading it through a natural emotional
sequence, and also because ‘it knows how it feels about them,’ in the sense
that it arrives at a stable attitude toward them overall.27
II.2a Some Reflections

Above I distinguished two notions of ‘better story’ and, derivatively, ‘bet-
ter ending of the life-story.’ One notion is, as I put it, especially relevant
to potential readers of the story or critics, such as reviewers. Employing
this notion, one might say that a certain story is better insofar as the
ending is unpredictable, or there are unexpected and exciting twists and
turns of plot, and so forth. It is clear that it cannot be this notion that is
pertinent (say) to questions about physician-assisted suicide. Rather, a
second notion is pertinent, and we can get a grasp on it by considering
salient examples from what are considered ‘good’ (in the relevant sense)
life-stories: persons thriving because of their efforts (and not merely
because of unexpected windfalls), persons learning from their mistakes
and thus flourishing (rather than deeming their mistakes ‘dead-weight
losses’ or flourishing by lucky accidents), and so forth. It is this second
notion that is relevant to questions about physician-assisted suicide.
In his essay on narrative explanation, Velleman says:
. . . the question how storytelling conveys understanding is insepa-
rable from the question what makes for a good story. Of course, a
good story can be good in many accidental respects, ranging from
the elegance of its diction to the personal attractions of its characters.
But what makes a story good specifically as a story—what makes it
a good example of storytelling, or narrative—is its excellence at a
particular way of organizing events into an intelligible whole.28
I am not sure whether Velleman intends to be making the following

sort of suggestion, or whether he would agree entirely, but I would
recommend that we interpret Velleman’s remarks as suggesting a way
of giving more concrete content to the first notion of ‘better story.’ That
is, I suggest that ‘good story,’ in the sense relevant to potential readers
and reviewers, is goodness in presenting a sequence of events in such
a way as to have emotional resonance (i.e., to map onto the natural
cadence of our emotions in a natural way). The various devices of plot
and literary style can be seen as ways of seeking to achieve a reso-
nance with our natural emotional cadences—to achieve ‘deja senti’ in
various ways. Some such devices work considerably better than others
(for individual readers and also groups of readers). The classic literary
(as well as musical) forms can be considered ‘tried and true’ ways of
evoking emotional resonances. (It then becomes the challenge of the
author—or composer—to work creatively and in original ways within
the structure of these forms, or perhaps to transgress the boundaries
set by these forms, so as to achieve emotional resonance.) Now I cer-
tainly doubt whether the notion of ‘good story’ of interest to readers
and reviewers of literature can be reduced to a single idea, but this sug-
gestion can perhaps illuminate at least a central aspect of the notion.29
Velleman says:
Any sequence of events, no matter how improbable, can provide
material for storytelling if it completes an emotional cadence. Twins
separated at birth are ideal protagonists for a story even if their
eventual reunion is a fluke. A discovery due to serendipity, a trag-
edy narrowly averted by dumb luck, a mundane act that unforesee-
ably becomes the last in a life accidentally cut short—these are the
stuff not only of literary storytelling but of legend, gossip, and other
forms of everyday narrative. Whether a winning lottery ticket or a
fatal housefire makes enough of a story to be featured on the local
news depends, not on whether its causes can be told, but rather on
whether the surrounding circumstances will call up feelings that can
be brought to some resolution by this inexplicable stroke of good or
bad fortune. So long as we feel an anxiety relieved or a hope dashed,
we have the sense of hearing a story, even if we have no idea why
events took the relevant turn.30
How exactly do these remarks fit with the point made above that the
narrative value of a life in which efforts are rewarded is greater than
that of a life in which one thrives by serendipity or lucky windfall?
We have to distinguish between narrative understanding and narrative
value. We can achieve narrative understanding, and we could even say
that the story is a good story, when we are told of someone flourishing
by winning the lottery. Depending on the details of the presentation, the
sequence of events can resonate with an emotional cadence—the story
can depict the sequence of events in such a way that it feels familiar—it
feels like a recognizable human drama. Upon hearing the presentation,
we are inclined to say, in our hearts, ‘Ah yes . . . ’ But it does not follow
that such a story depicts a life with great narrative value. So, for exam-
ple, such a story does not depict a life with more narrative value than
the story of a life of flourishing as a result of lessons learned or simply
hard work. A better story in the telling need not make for a story of a
life with more value. An ending that is ‘better’ in the sense of narrative
understanding need not be ‘better’ in the sense of narrative value.
Return to Velleman’s claim, ‘ . . . the question how storytelling con-
veys understanding is inseparable from the question what makes for a
good story.’ The question of how storytelling conveys understanding is
inseparable from what makes for a good story, in the sense that is rel-
evant to readers and reviewers of literature. Further, it seems that only
a life that can be explained in the distinctively narrative way is capable
of having the characteristic narrative value—is capable of being evalu-
ated through a function that is sensitive to the sorts of structural rela-
tionships sketched above. So the possibility of narrative explanation is
inseparable from the possibility of narrative evaluation; being a nar-
rative is a necessary condition for having narrative value. Of course,
it does not follow from anything said thus far that the better a story
is in the telling—the greater its virtues along the dimension pertinent
to narrative understanding—the higher the value of the function that
determines overall value of the life will be (other things equal).
As I said in the introductory section of this essay, I have suggested in
previous work that the value of acting freely is the value (whatever that
is) of writing a sentence in the narrative of one’s life. On some approaches,
acting out of character is inconsistent with the idea of one’s life’s having
narrative structure (and admitting of narrative explanation). But I do not
employ the idea of narrative in a way that would rule out free action
that is ‘out of character.’ The situation here is a bit like the possibility
of narrative explanation of flourishing as a result of a windfall, such as
winning the lottery. If the storyteller is adept, the story will resonate; that
is, the structure of the story will map onto the emotional memory of the
‘listener’ or ‘reader.’ As above (in the case of succeeding as a result of a
lucky accident), the story of action out of character can be told in such a
way that it feels familiar—it feels like a recognizable human drama. It all
depends on the way the story is told, and it is a delicate matter to tell the
story in such a way as to elicit an emotional response of the proper sort.
Certain ways of telling the story will result in puzzlement, whereas more
skillful storytelling will evoke that heartfelt, ‘Ah yes . . . ,’ indicating an
isomorphism with a human emotional cadence. One can freely act out of
character, even on an approach that invokes the importance of narrative
explanation, and identifies the value of acting freely with the value of a
certain sort of aesthetic activity (defined in terms of narrative).
On certain views of moral responsibility, one is morally responsible

insofar as one expresses an enduring ‘character trait.’ This approach fol-
lows David Hume. But the basic problem with the Humean approach is
that one can act freely, and be morally responsible, even when one acts
out of character. The Humean may try to find some more ‘complicated’
character trait that is expressed in the behavior, but I believe that this
trail leads inevitably to an unacceptable etiolation of the ordinary notion
of ‘character trait.’ On this approach, it will turn out in the end that ‘char-
acter traits’ do not play any substantive, independent explanatory role.
Despite the considerable attractions of a Humean approach to such mat-
ters, I believe that the possibility of acting out of character—and doing
so freely—constitutes a significant and, indeed, insuperable problem.
On my view of acting freely and moral responsibility, one need not
be expressing any sort of enduring character trait, or expressing any
sort of ‘commitment’ or ‘positive evaluation’ of the relevant behavior.
Rather, one is expressing oneself in the sense of writing a sentence in
the narrative of one’s life. If the story of one’s past, and one’s behavior,
resonates appropriately, then one is acting freely and is morally respon-
sible for the behavior in question. Again, it all depends on whether the
storytelling is adept. In some cases the story can be told in such a way
that the behavior under consideration is ‘prefigured,’ even if it is out
of character. When the behavior is in this way prefigured, the story
resonates with the listener (or ‘audience’) in the indicated way, and the
agent is morally responsible.31
III. NARRATIVE, ENDINGS, AND IMMORTALITY

On the view of narrative explanation under consideration here, a nar-
rative must have an ending. If our lives are narratives, or have the dis-
tinctive structure of narrative, then they must have endings. On this
view, we cannot be immortal (insofar as our lives are narratives or have
narrative value), if our lives are indeed narratives. To imagine immortal
human life is to imagine human life devoid of an essential or at least
very important characteristic: having narrative structure and thus a
distinctive dimension of value.
I think that, strictly speaking, this is correct. If a narrative must have
an ending, then it is clear that our lives cannot have the sort of mean-
ing that involves taking a retrospective perspective on its totality, as
it were, and assigning a meaning that reflects the overall arc of the
lifestory. But an immortal life could have something very much like
narrative meaning, strictly conceived; or, perhaps better, the relevant
temporal stages or parts could be explained in the distinctive and strict
narrative manner, and they could have the signature features of narra-
tive value. Whereas the life as a whole could not be considered a nar-
rative, the parts could be, and this would seem to render immortal life
recognizably like our current human lives and also potentially desir-
able (in a distinctive way).32 The literary analogue for such a life is
not the novel, but perhaps a collection of short stories. Of course, the
collection needs to be infinitely large, and the short stories need to be
interconnected, with the same character appearing as the protagonist.
Perhaps a better analogy would be a series of novels with the same
protagonist, like a mystery series with the same detective. Over time
the detective’s character may change, but the changes can be organic;
they need not be discontinuous.33
Another analogy would be a ‘serial’ novel or even a television series.
Our lives may seem to be soap operas at times! Or perhaps they are
‘sitcoms’; if so, mine is in certain respects (apart from pecuniary consid-
erations) more like Larry David’s on HBO’s marvelous series, Curb Your
Enthusiasm, than Ward Cleaver’s on Leave it to Beaver. Now, of course,
all sitcoms do eventually get cancelled—Leave it to Beaver can only be
seen in syndicated reruns. But, as far as I can see, it is not part of their
distinguishing features or essence that they be so (cruelly) terminated.
A serial novel, a television series, a series of mystery novels with the
same detective, a collection of short stories—they all have parts that can
be explained and evaluated in the characteristic narrative fashion.34
If I consider an apparently possible immortal human life, I can see
parts of it as having the distinctive features of narrative understanding
and value. Parts of the life can be explained in such a way as to achieve
emotional resonance and resolution. But since the life as a whole has
no ending, there is no possibility of achieving distinctively narrative
understanding of the whole life. And thus there is no final answer, as it
were, giving the narrative value of the life as a whole. Whereas this is
indeed the case, I do not think that it renders immortal life unintelligi-
ble or unrecognizably human; nor do I think it makes it impossible for a
human being to find such a life valuable and desirable. (After all, to find
such a life potentially desirable does not require rank-ordering it even
ordinally against other such lives, or against finite human lives. And
even in a finite life, there is the problem of combining the two irreduc-
ible dimensions of value into some overall value score—the cumulative
measure of momentary well-being and the narrative dimension.)
Someone might say that there is a big difference—a big and crucial
difference—between merely (!) very, very long life and infinitely long
life. It might be suggested that, although we can ‘get our minds around’
long life, it is a mistake to suppose that infinitely long life is relevantly
similar to merely very long, finite life. The infinite, it might be said, is
just fundamentally different, and thus fundamentally mysterious (and
not presumably amenable to evaluation in the relevant ways).
This raises difficult and obscure questions. Consider, first, the putative
Divine Attributes, such as omnipotence, omniscience, moral perfection,
and so forth. It is sometimes claimed that we cannot understand these
properties, since they are fundamentally different from the bounded

and finite analogues in human power, knowledge, and goodness. The
standard reply, of course, is that the Divine Attributes, although differ-
ent from their finite analogues, are to be understood along the lines of
those attributes; the relationship between the infinite and finite here is
one of analogy.
Similarly, the arithmetical rules that define certain operations, such
as addition, subtraction, multiplication, and division are different for
infinite and finite numbers. It is sometimes supposed that, because the
arithmetics are different, there must be a difference in kind between
finite and infinite magnitudes (and thus between finite and immortal
lives). Given this difference in kind, it may not even be possible to grasp
or comprehend an infinitely long human life. But, again, the reply
should be that, although the rules pertaining to the relevant mathemati-
cal operations are different in finite and infinite numbers, they are not
totally dissimilar. Indeed, there are sufficient similarities to suggest that
the relationship between these sets of rules is one of analogy.
But Erich Reck has raised an interesting question here:
. . . one can now start to play with other mathematical facts about the
infinite to create some curious questions: Suppose, e.g., that every
hundred years I have one bad day (for whatever reason). Then over
the span of an infinite life (say a life that is omega days long—the
smallest infinite), the number of bad days will still be infinite. In
fact, the number of bad days overall will be the same as the number
of good days (both omega). Or to modify the example a bit: sup-
pose you have a bad day after ten days, then the next after 100 days,
then the next after 1000 days; i.e., the intervals between bad days
get longer and longer. Overall, there will still be infinitely many bad
days, thus the same (infinite) number as the good days. Or to turn
the example around: imagine that after ten days you have your first
good day of your life, then the next good day after 100 days, then
the next good day after 1000 days, etc. Overall you will still have
the same (infinite) number of good and bad days. These kinds of
examples are used in the literature on the infinite to illustrate how
different it is from the finite.35
In my view, the difficulties to which Reck brings our attention per-
tain to any attempt to quantify and aggregate welfare in the way typi-
cally employed by the function that determines overall welfare, given
information about momentary welfare. These phenomena need to be
addressed. But they do not pertain to the second dimension of value,
the narrative dimension, which is irreducible to the first. The narrative
dimension of value has to do with the structural relationships among
events; it does not involve any sort of addition or mere quantitative
analysis of good and bad experiences, good and bad days, and so forth.
Here then is a way in which the narrative dimension of value is better

able to address certain features of immortal life.
Similarly, consider a point raised by Borges in his short story, The
Immortal, as described by Gustavo Llarull:
. . . the immortals are fundamentally devoted to intellectual pursuits;
their physical needs are very easily satisfied (they can go without
food for days) and sensual pleasures do not entice them. They are
immersed in thought for entire weeks, and they barely talk to each
other. In this context, Borges presents an interesting episode: one of
the immortals falls from a cliff.
Although the other immortals know about this accident, they go
to his aid months later. Borges suggests that what from our finite view
is surely deemed as cruelty or negligence may simply be the natural
expression of attitudes that have changed in virtue of the correla-
tive change in time patterns. In an average life span of 70 years, tak-
ing 15 minutes to help someone in need may be appropriate . . . Now,
the same proportion, for a life span of 7,000 years, yields 25 hours!
Needless to say, the idea of having to wait 25 hours, or more, to be
aided by my fellow immortals, when they are fully aware that I am
in need of help, sounds repulsive to say the least . . . 36
As Llarull points out, Borges’ description need not be the reality in
an immortal life. Llarull says, ‘ . . . “negative” or pain-related episodes,
interactions or events would not lose their urgency [in an immortal
life].’37 It is clearly a mistake to think of immortal human life along the
lines of an anatomical ‘horizontal (or longitudinal) explosion’ of ordi-
nary, finite human life, keeping the temporal proportions fixed, as it
were. Although there are perfectly good welfare-based reasons not to
take the ‘longitudinal explosion’ model seriously, one can also invoke
narrative relationships. Pain must be responded to as soon as possible,
other things equal, and this fact would still obtain, in an immortal exis-
tence. It is a mistake to take a purely quantitative approach to envisag-
ing and evaluating the relationships between finite and infinite human
existence.38
IV. CONCLUSION
I have in a tentative way explored the interconnections among narra-
tive explanation, narrative value, free will, and immortality. I have built
on the fascinating and suggestive work of David Velleman. I have sug-
gested that our acting freely is what gives our lives a distinctive kind of
value—narrative value. Free Will, then, is connected to the capacity to
lead a meaningful life in a quite specific way: it is the ingredient which,
when added to others, endows us with a meaning over and above the
cumulative value derived from adding together levels of momentary
welfare. In acting freely, we are writing a sentence in the story of our
lives, and the value of acting freely is thus a species of the value of artistic
creativity or self-expression (understood appropriately). Finally, I have
suggested that the fact that our lives are stories need not entail that they
have endings, or that immortality would necessarily be unimaginable
or essentially different from ordinary, finite human life. Yes, a certain
sort of narrative understanding of our lives as a whole would be impos-
sible in the context of immortality; but much of what we care about,
and value, in our stories might remain.39
NOTES
1. It is well known that it is difficult to predict the future. A colleague of
mine once pointed out that the situation is even worse with respect to the past:
it is, he said, impossible to predict the past.
2. John Martin Fischer, ‘Responsibility and Self-Expression,’ The Journal
of Ethics, Vol. 3, No. 4 (1999): 277–297.
3. John Martin Fischer, ‘Epicureanism About Death and Immortality,’
forthcoming, The Journal of Ethics (Chapter 7).
4. J. David Velleman, ‘Well-Being and Time,’ Pacific Philosophical Quarterly
72 (1991), pp. 48–77.
5. Ibid.
6. Cashiers in California have asked me, ‘How’s your day going so far?’
Also, I’ve been implored to ‘Have a great rest-of-your-day!’ Of course, these
sorts of thoughts, from people one does not really know, are superficial and
slightly irritating. They can also reflect a mistaken view about value—or, per-
haps, a focus on only one dimension of value.
China’s Chairman Mao was asked what he thought of the French Revolution.
He reportedly replied, ‘It is too early to tell.’ This clearly goes to the opposite
extreme. Aristotle more moderately urged us ‘to call no man happy, until that
man is dead,’ alluding to an old adage going back to Herodotus’ tale about
King Croesus. The adage was ubiquitous in the 5th century BCE, made espe-
cially so by Sophocles, whose play, Women of Trachis, begins, ‘There is an old
saying that no man is blessed until his final day.’ (Heracles dies at the end of the
play, after entering as hero at the start.)
7. J. David Velleman, ‘Well-Being and Time, p. 51; the quotation is from
Michael Slote, ‘Goods and Lives,’ in his Goods and Virtues (Oxford: Clarendon
Press, 1983), pp. 23–24.
8. Ibid., pp. 53–4.
9. Ibid., pp. 54–5.
10. Ibid., p. 55.
11. Whereas Velleman’s contention here is appealing, it is tantalizingly
underdeveloped. One problem is that there is a limited number of examples
actually offered; also, as I point out in the text, Velleman does not offer any
general formulaic explanation of narrative value. Perhaps not surprisingly,
then, various philosophers have suggested to me that there are other poten-
tial explanations for our intuitions or judgments about various lives that are
at least as plausible (as ‘narrative value’). Thaddeus Metz has suggested to me
that we can account for the judgments to which Velleman draws our attention
by keeping in mind that stretches of one’s life (or activities in these stretches)
can have instrumental value as well as intrinsic value; thus, even a utilitarian
can account for the greater value of certain lives by pointing out that stretches
in certain lives (but not others) can have instrumental value. (For a discus-
sion, see Thaddeus Metz, ‘Utilitarianism and the Meaning of Life,’ Utilitas 15
(2003), pp. 50–70.) David Hershenov has suggested that the judgments can be
explained not so much by invoking narrative value, but by the moral value
of staying married, the desire for (or value of) moral self-improvement, and
so forth. (A skeptic about the invocation of the notion of narrative value here,
such as Hershenov, might then seek to construct examples of lives that lack the
putatively relevant features—moral self-improvement, staying married, and so
forth—to see if the relevant evaluations of the life stay the same.) Andrews
Reath has suggested that what is significant about the cases under discussion is
that they indicate that there are irreducibly relational goods; but this does not in
itself entail anything about specifically ‘narrative’ value.
These matters require much more attention than I can give here. I should
point out that I myself have some doubts about how to evaluate the various
scenarios. For example, I do not think that it is somehow better or more valuable
that a depressed individual pull out of his depression by pure strength of will,
or a regimen of psychotherapy, or any other sort of extended and perhaps ardu-
ous set of activities and reflections, rather than by taking an antidepressant. Of
course, if the chances of recurrence of the depression are greater if one simply
takes the medication, then that counts against it. But I do not share the intuition
that there is something less valuable—that it is somehow ‘cheating’—to use the
antidepressant (successfully). And it is not exactly clear to me how to distinguish
this case, or class of cases, from those discussed by Velleman. I hope to turn to
these issues in future work. Here I shall not press the worries, and take it, as a
working hypothesis, that Velleman’s intuitions or judgments about scenarios are
plausible, and that his invocation of narrative value is explanatorily helpful.
12. J. David Velleman, ‘Well-Being and Time,’ p. 62.
13. This is not necessarily a criticism of Velleman, as it is unclear that he
envisaged his account as providing guidance in all cases of potential physician-
assisted suicide.
14. J. David Velleman, ‘Well-Being and Time,’ p. 71.
15. Ibid., p. 71.
16. For skepticism about the necessity of this sort of capacity for the distinc-
tive notion of ‘valuing’ (as opposed to merely preferring) and also for autonomy,
see Agnieska Jaworska, ‘Respecting the Margins of Agency: Alzheimer’s Patients
and the Capacity to Value,’ Philosophy and Public Affairs 29 (1999), pp. 105–38.
17. I do not know how to prove my supposition, and I recognize that it would
be good to have more to say here. I base the claim on consideration of contexts
in which we would normally suppose that there is narrative value and thus the
ability of the relevant agent to affect the meanings of past events, but in which it
is explicitly understood that the agent is manipulated in such a way as not to be
acting freely; in this range of thought experiments, my intuition is that the agent
cannot affect the meanings of the past events, and that this is precisely because
he does not act freely. I hope to be able to justify this intuition (at least to some
extent) in future work. See Chapter 1 for further defense of my suggestion.
18. Of course, the past goods in question are not ‘experiential’ goods.
19. Carl Ginet, On Action (Cambridge, U.K.: Cambridge University Press,

1990). See, also, John Martin Fischer, The Metaphysics of Free Will: An Essay on
Control (Cambridge, Mass.: Blackwell Publishers, 1994).
20. We now have available to us a kind of ‘alternative’ explanation—no
doubt, not the intended explanation—of the putative fact alluded to in footnote
1 above—that it is impossible to predict the past. If we can go backward in
narrative space-time by freely acting in the future, and this free action cannot be
predicted, then we also cannot predict what the narrative past will have been.
21. J. David Velleman, ‘Narrative Explanation,’ Philosophical Review 112
(2003), pp. 1–26.
22. Ibid., p. 1.
23. Ibid., p. 1.
24. Louis Mink, ‘Philosophical Analysis and Historical Understanding,’ in
Brian Fay, Eugene. O. Golob, and Richard T. Vann, eds., Historical Understanding
(Ithaca: Cornell University Press, 1987), pp. 118–46; and E.B. Gallie, Philosophy
and the Historical Understanding (London: Chatto and Windus, 1964).
25. J. David Velleman, ‘Narrative Explanation,’ p. 10.
26. Ibid., p. 19.
27. Ibid., p. 19.
28. Ibid., p. 1.
29. To add just one dimension of complexity: note that a typical literary
work or musical composition will not simply evoke one emotion, but many
(sometimes colliding) emotions.
30. J. David Velleman, ‘Narrative Explanation,’ pp. 6–7.
31. I can only gesture briefly at an account of these matters here. Note that
my sketch here may suggest an unintuitive ‘relativization’ of moral responsi-
bility to a particular bit of story-telling. I do not know whether this is genu-
inely problematic; if so, one could define acting freely and moral responsibility
in terms of the ‘availability’ (in some sense) of a story of the right sort.
32. It is a staple of ‘postmodern’ critiques of ‘narrative’ that they posit some
sort of overarching or ‘totalizing’ meaning. On my account of the narrative struc-
ture of an immortal life, there is no single ‘grand narrative’ and thus no ‘totalizing
meaning’; and yet there remains much to which a postmodernist could object,
given that the sub-parts of the immortal life are taken to have suitable meanings.
33. In Amanda Cross’s (Carolyn Heilbrunn’s) delightful series of academic
mystery novels, Kate Fansler (English Professor and detective extraordinaire)
changes noticeably but understandably over time. As she ages, Kate is no lon-
ger the hard-drinking detective who solved the murder of the first tenured
woman professor at Harvard in Death in A Tenured Position. In the later works,
she worries about her excessive indulgence in alcohol, and seeks moderation in
this and other pleasures.
The story of Heilbrun’s own life, as opposed to that of her fictional profes-
sor/detective, ended rather abruptly. The Los Angeles Times obituary (October
15, 2003), says:
Carolyn G. Heilbrun, a distinguished feminist scholar who illuminated the
female experience through erudite reinterpretations of classic English litera-
ture and in literate mystery novels written under the name Amanda Cross,
was found dead in her New York City apartment Friday after an apparent
suicide. She was 77.
The pioneering feminist critic had decided years earlier that she would
end her life by the age of 70 to avoid the inevitable deterioration of age, but
she later explained that she had let the deadline pass when her 60s proved
deeply satisfying.
In The Last Gift of Time: Life Beyond Sixty, a book published the year she
turned 71, she said she would decide each day whether to keep on living.
Her son Robert told the New York Times last week that she had not been ill
when she decided to kill herself.
It is almost as if Heilbrun always knew the end of her life-story, but she
couldn’t wait for the organic development of the plot. In any case, the details
of her reasons for choosing to commit suicide when she did may remain (not
inappropriately) a mystery.
34. In his classic essay, ‘The Makropulos Case: Reflections on the Tedium of
Immortality,’ Bernard Williams argues that an immortal life would be essen-
tially meaningless and unattractive: Bernard Williams, Problems of the Self
(Cambridge, U.K.: Cambridge University Press, 1973), pp. 82–100, reprinted in
John Martin Fischer, ed., The Metaphysics of Death (Stanford: Stanford University
Press, 1993), pp. 73–92. In chapter 6, I argue against Williams’ view. I suggest
that the sort of immortality suggested in the text of the current essay—along the
lines of a character in a series of novels or a continuing television series—could
be appealing.
35. Erich Reck, personal correspondence.
36. Gustavo Llarull, ‘The Problem of Immortality: A Response to Williams,’
unpublished manuscript, University of California, Riverside department of
philosophy.
37. Ibid.
38. The mistake of ignoring the crucial narrative structure of life, even infi-
nite life, is also found in this passage from Victor Frankl: ‘What would our
lives be like if they were not finite in time, but infinite? If we were immortal,
we could legitimately postpone every action forever. It would be of no conse-
quence whether or not we did a thing now; every act might just as well be done
tomorrow or the day after or a year from now or ten years hence. But in the face
of death as absolute finis to our future and boundary to our possibilities, we
are under the imperative of utilizing our lifetimes to the utmost, not letting the
singular opportunities—whose “finite” sum constitutes the whole of life—pass
by unused.’ (The Doctor and the Soul [New York: Alfred Knopf, 1957], p. 73.)
39. I am grateful for thoughtful comments by Erich Reck, Gustavo Llarull,
David Hershenov, David Glidden, Ward Jones, and Thaddeus Metz. I thank
Gustavo Llarull for permission to cite his unpublished work. I have pre-
sented versions of this essay at the Winter Colloquium of the Department of
Comparative Languages and Literatures, University of California, Riverside
(2005), and the Free Will Workshop at the University of California, Riverside
(2005). On these occasions many people gave me very insightful comments,
including: Gary Watson, Andrews Reath, Nathan Placencia, Chris Yeomens,
Neal Tognazzini, Thomas Scanlon, and Lisa Raphals.
10
Stories and the Meaning of Life
John Martin Fischer
I. INTRODUCTION: A FRAMEWORK
FOR MORAL RESPONSIBILITY
In various places, I (sometimes in collaboration with Mark Ravizza)
have sought to present the elements of a theoretical “framework for
moral responsibility.”1 Here I shall begin by sketching the framework
in order to give the background for a more detailed development of the
idea that the value of acting so as to be morally responsible is the value
of a certain distinctive kind of self-expression.
The overall framework for moral responsibility includes (at least) the
following elements: a distinction between the concept of moral respon-
sibility and its conditions of application; a distinction between “regula-
tive” and “guidance” control; an argument that guidance control, and
not regulative control, is the “freedom-relevant” condition linked to
moral responsibility; an account of guidance control in terms of mecha-
nism ownership and moderate reasons-responsiveness; an argument
that guidance control, so construed, is compatible with causal deter-
minism; and an account of the value of moral responsibility (in terms
of self-expression).
There are various plausible ways of specifying the concept of
moral responsibility, including the “moral ledger view,” the “fitting-
ness-of-providing an explanation” view, and the “Strawsonian view,”
which involves aptness for the “reactive attitudes” (resentment,
indignation, gratitude, love, and so forth). I do not take an official
stand as to the proper analysis of our concept of moral responsibil-
ity; perhaps there is no single correct answer here, and our concept
involves elements of the various suggestions. Even so, I find it help-
ful and instructive to take as a working hypothesis some version of
the Strawsonian account of the concept of moral responsibility.
But under what conditions does the concept apply? I here follow
Aristotle: an agent must meet both some sort of “epistemic” condition
and a “freedom-relevant” condition. This tracks Aristotle’s claim that an
agent fails to act voluntarily to the extent that he acts from ignorance or
force. My primary focus has been on the freedom-relevant condition.
165
I offer plausibility arguments for the contention that “acting freely”

(an “actual-sequence” sort of freedom) plays the role of the freedom-
relevant condition; put in other words, guidance control, and not regu-
lative control, is the freedom-relevant condition associated with moral
responsibility. On my view, then, an agent may be morally responsible
but never have had genuine metaphysical access to alternative possibil-
ities—he may never have had “freedom to do otherwise.” These plausi-
bility arguments employ the thought-experiments that originated with
John Locke and have been dubbed, “Frankfurt-Style Examples,” after
Harry Frankfurt. Frankfurt has argued that such cases—involving a sig-
nature sort of pre-emptive overdetermination—imply the falsity of the
Principle of Alternative Possibilities (the principle that states that free-
dom to do otherwise is a necessary condition of moral responsibility).
In my view, the moral of the Frankfurt Stories—and it is interest-
ing that these hypothetical examples are frequently referred to as “sto-
ries”—is that acting freely (and being morally responsible) is a matter
of how the actual sequence unfolds, not whether the agent has genu-
ine metaphysical access to alternative sequences. Although I find the
thought-experiments instructive and illuminating, I believe that there
are other dialectical routes to the same conclusion, and that the thought-
experiments should form part of an overall strategy of argumentation.2
Even if one doesn’t find the Frankfurt-Examples convincing, this should
not in itself issue in a rejection of the basic conclusion about the relation-
ship between moral responsibility and freedom to do otherwise. This
point resonates with my general methodological approach, according to
which stories play a key role, but not the sole or exclusive role in analy-
sis of various phenomena. Thus, the role of Frankfurt-stories is like the
role of hypothetical examples in ethics.
Especially with my co-author, Mark Ravizza, I have sought to
sketch an account of guidance control. On our approach, guidance
control involves two chief elements: mechanism ownership and rea-
sons-responsiveness. An agent exhibits guidance control of an action
insofar as the action issues from the agent’s own, “moderately reasons-
responsive” mechanism. Our more detailed accounts of both elements
have elicited worries and objections; particularly contentious have
been the claim that mechanism ownership involves a certain sort of
“subjective” condition, and that reactivity to reasons is “all-of-a piece.”
I have sought to defend the fundamental ideas where possible. Also,
in recent work I have pointed out that the basic elements of the frame-
work for moral responsibility can be preserved while adjusting the spe-
cific details; I have argued that I can still accomplish everything I had
hoped to accomplish by offering a framework for moral responsibil-
ity even without a commitment to a strong subjectivity or to the con-
tention that “reactivity-is-all-of-a piece.”3 That is, I can accept slightly
adjusted accounts of the fundamental elements of the framework while
Stories and the Meaning of Life 167
still maintaining that moral responsibility does not require regula-

tive control, that it is fundamentally a historical notion, and that it is
compatible with causal determinism. This is important because some
philosophers have apparently dismissed the view because they have
found the specific subjective view or the view that reactivity is all-of-a
piece troubling.
Finally, I have suggested that the value placed on acting freely and
being such as to be held morally accountable is the value of a certain
distinctive kind of self-expression. When we act freely, we express our-
selves in a way that is perhaps a form of artistic creativity (or akin to
such self-expression). What matters is not that we make a difference to
the world, but that we make a certain kind of statement. In acting freely,
we make it the case that our lives have a narrative dimension of value;
thus, acting freely is the ingredient which, when added to other fea-
tures, endows our lives with a meaning beyond the simple addition of
momentary episodes of welfare, or even the addition of such episodes,
weighted for their temporal location. Unlike mere non-human animals,
our lives are stories in a strict sense, and they can have a distinctive
kind of meaning—narrative meaning.4
II. FREEDOM AND THE VARIETIES OF VALUE:

A PROPOSAL
I believe that we do indeed value self-expression, and that the value of
acting freely (and thus being morally responsible agents) is a species
of the value of self-expression. I am inclined to think of this as a sort
of “artistic” self-expression, but if it is, it may be a context, which is
unique or sui generis (in a way I shall explain below). Why exactly do
we value this sort of self-expression? This is a hard question. I think it
is connected to questions about the meaning of our lives.
As I said above, our lives are stories, whereas the lives of rats and even
cats are not. Certainly, one can tell the “story” (speaking loosely) of a rock
or a rat or a cat, but these accounts are not “stories” in a strict sense. They
are not narratives. Of course, it is an interesting and vexed question just
what has to be added to a mere account or chronicle of events to get to a
“story” (strictly speaking) or a narrative; I am inclined to accept the bare
bones at least of the suggestion of David Velleman that the necessary
additional ingredient is some distinctive sort of understanding provided
by the account.5 Further, insofar as we act freely our lives have a narra-
tive dimension of value. Along this dimension, one does not simply add
together momentary levels of well-being, and the meanings or values
of events depend on certain distinctive relationships with other events.
Velleman calls these “dramatic” or “narrative” relationships.
When I act freely, I write a sentence in the story of my life; that is,
the account of my life is strictly speaking a story (rather than a mere
chronicle of events), and my life has a narrative dimension of value.

Insofar as a story or narrative is by definition a work of art, I am an
artist, when I act freely, and I am inclined to say that the value of my
free action is the value of artistic self-expression.6 Note that it does not
follow from the fact that the value of our free action is a species of the
value of artistic self-expression that it is a kind of aesthetic value; rather,
the point is that the value of our free action is a species of the value of
artistic self-expression, whatever that value is. I take it that artistic self-
expression has value from the perspective of human flourishing or
“doing well”; in a broad sense (that encompasses “goodness” as well as
“rightness),” this is arguably a moral value.
Thus, when I act freely, I am by definition engaging in artistic or aes-
thetic activity, and this activity can have value for us—a value that is
distinct from the value of the product. For example, we might plausibly
think that it is valuable—from the perspective of living well or flour-
ishing—to engage in artistic activity; this value is conceptually distinct
from any sort of evaluation (aesthetic or even moral) of the product—
the sculpture or painting or novel. Similarly, on my view, the activity of
my acting freely is a species of artistic activity, and thus it has whatever
value we place on this sort of activity. Of course, it does not follow from
the fact that my acting freely is a species of artistic activity that the
value of the product—the life or perhaps life-story—is solely or primar-
ily aesthetic. (It would be natural to suppose that, because the activity is
by its nature aesthetic, the value we place on the product must be a spe-
cies of aesthetic value; but this is a mistake.) My activity in acting freely
is a kind of artistic self-expression, and it thus has the value we place on
such aesthetic activity. But the product—my life or life-story—typically
is evaluated primarily from the perspective of prudence and morality,
although it can also be evaluated from the aesthetic perspective.
Of course, a work of art such as a sculpture can be evaluated along
various dimensions, including both aesthetic and moral. (Here I am
talking about the product of artistic activity—the sculpture.) Surely,
evaluation is purpose-dependent, and we might have various purposes
in evaluating a particular sculpture. We might, for example, want to
know whether it is composed of materials that are environmentally
sensitive or scarce, or that could only be procured by exploiting people,
and so forth; such purposes would give rise to a moral evaluation of
the sculpture. But perhaps typically we would want to know whether
the sculpture is aesthetically pleasing, and thus our “primary” way
of evaluating it would be aesthetic. That is, given the range of human
purposes, perhaps the “typical” or “primary” sort of evaluation of a
sculpture would be in terms of aesthetic dimensions. Here it might be
said that there is a “match” between the sort of activity that issues in the
product—artistic or aesthetic activity—and the “typical” or “primary”
mode of evaluation of the product.
A life, or a life-story, can be evaluated along different dimensions,

including aesthetic, moral, prudential, and so forth. A life or life-
story—considered as a product of artistic self-expression—can cer-
tainly be evaluated aesthetically. It is perhaps most natural to think
about stretches of a life or parts of a life here. For example, athletes are
sometimes thought to retire too late; this seems especially prevalent in
boxing, and it tends to tarnish or diminish the athlete’s entire career. I
also believe that certain athletes and coaches have retired “too early”—
before reaching the heights they could have reached, or perhaps cut-
ting off a string of accomplishments before its “natural” or “optimum”
conclusion. (I believe that the great coach of the San Francisco 49ers,
Bill Walsh, retired too early—and I believe he thought this too in retro-
spect.) No doubt, certain academics—even philosophers—have retired
too early; but surely none—especially no philosophers—have retired
too late!
I take it that these considerations are essentially “aesthetic,” although
they surely are related to (or perhaps involve) moral dimensions. They
have to do with the trajectory or arc of one’s career. Obviously, we can
evaluate the trajectory of several of one’s projects in addition to one’s
career, and we also are deeply concerned with the arc of one’s rela-
tionships with others. Sometimes a relationship maintains its integrity,
whereas in other cases a relationship might end abruptly or slowly
unravel in an unlovely way.
Although we certainly engage in aesthetic evaluation of aspects or
phases of our lives as well as our lives as a whole, I think it is undeni-
able that the typical” or “primary” modes of evaluation (given the stan-
dard human purposes and interests) are prudential and moral. Thus I
suggest that if acting freely is indeed a species of artistic self-expression,
this context is sui generis: it involves a kind of activity—artistic or aes-
thetic activity—whose product is not typically or primarily evaluated
aesthetically. That is, when we act freely, we tell a story that is most
naturally—given the intrinsic nature of human activity and a broad
range of human purposes—evaluated in terms of moral and prudential
considerations, even though the nature of the activity is artistic. Here
then we have perhaps identified what is special and unique about the
context of free action: whereas there is a match between the nature of
artistic activity and the primary mode of evaluation of (say) a sculp-
ture, there is a striking discrepancy between the nature of free human
activity (arstistic) and the value of the activity; whereas its nature is a
species of aesthetic activity, the typical or primary modes of evaluating
the product are prudential and moral.
It is perhaps not surprising that the two elements of this idea about
the value of acting freely correspond in some rough way to the two
components of the analysis of guidance control. The value of acting
freely, on the account I have sketched, is the value (whatever kind of
value that is) of artistic self-expression whose product is typically eval-

uated in terms of prudence and morality. Guidance control is analyzed
in terms of two components: ownership and reasons-responsiveness.
Ownership posits a special relationship to the self, and this corresponds
to the value of artistic self-expression. And reasons-responsiveness per-
tains to the salient dimensions of assessment of human lives in terms
of morality and prudence. Although the relationships here are perhaps
less tight than one might like, it is appropriate that there be echoes of
the elements of the account of the value of acting freely in the analysis of
acting freely. This is one way in which parts of the overall framework
for moral responsibility form a unified, cohesive whole.
III. IMPLICATIONS AND REFINEMENTS
III.1. Historical Narratives

I have contended that our exercising our capacity for acting freely trans-
forms our lives into genuine stories and endows us with the narrative
dimension of value. Eric Schwitzgebel has pointed out however that
one can write “narratives” that are not most appropriately thought of
as works of art. If, say, one writes the history of a family or region, this
sort of “narrative” may seem to have the characteristics I have pointed
to (including the non-additive dimension of value), but we would not
typically say that such a narrative is a piece of art, that writing it is
a species of aesthetic activity, and that its value is a kind of aesthetic
value. One might then worry that simply having the characteristics of
being a narrative and having the narrative dimension of value doesn’t
make our lives “works of art” in the relevant sense, or our free actions
a species of artistic self-expression, and so forth.7
Note however that such narratives—the history of a region or fam-
ily—do not themselves possess the narrative dimension of value. That
is, we would presumably judge them by their accuracy in depicting the
events in question; it would not be a further virtue of such an account
that it depicted (perhaps falsely) individuals learning from prior mis-
takes, flourishing as a result of prior efforts (rather than mere luck),
and so forth. Certainly, these features would endow the lives depicted
with additional value—narrative value, but they would not endow the
narrative itself with such value. Similarly, although the lives depicted
would have the additional dimension of value, we would certainly not
attribute greater value to the creator of the (perhaps false) narrative!
So family narratives would typically not be works of art, but this does
not in any way diminish the plausibility of the claim that we are engag-
ing in artistic activity in freely writing the stories of our lives, and that
such free activity helps to endow our life-stories with the distinctive
features of certain works of art. And, of course, as I emphasized above,
the fact that our activity in creating our life-stories is properly consid-
ered a species of aesthetic creativity does not in itself rule out the pos-
sibility that the typical or primary modes of evaluation of the product is
not aesthetic.8
III.2. The Aesthetic Fallacy

My point is that it does not follow from the fact that one’s free activity
is a species of artistic self-expression, that the only or primary or typi-
cal mode of evaluation of the product be aesthetic. Of course, I have not
argued for or established that the nature of the activity and the value
of the product can pull apart in the way I have suggested. My point
here is rather more modest; I wish simply to insist that, insofar as we
distinguish between the activity and its product, there is no entailment
between the nature of the activity and the mode of evaluation of the
product. Given that evaluation is purpose-driven, it is not surprising
that there are various perfectly reasonable modes of evaluation of the
products of artistic activity, and thus that we cannot simply “read off”
the (appropriate) mode of evaluation from the nature of the activity.
I have sought to carve out a space between the aesthetic nature of
our free activity of writing our life-stories and the appropriate modes of
evaluation of those stories. On this picture, it would be a fallacy to infer
from the fact that the essential nature of our free activity is aesthetic
to the proposition that our lives (or life-stories) should be evaluated
exclusively or primarily in terms of aesthetic criteria. I suggest that we
call this problematic move, “The Aesthetic Fallacy.” Some have attrib-
uted to Nietzsche the view that our lives should be evaluated primarily
along aesthetic—specifically, literary—dimensions.9 Philosophers such
as Nehamas emphasize Nietzsche’s “aestheticism” and his view of “life
as literature.” The details of Nietzsche’s view—and even the broad con-
tours of it—are in dispute, and I am certainly not qualified to enter into
exegetical debates. I simply wish to point out that it would be fallacious
to suppose that it follows from the fact that our freely writing the books
of our lives is a species of aesthetic activity that the books themselves
should be evaluated primarily aesthetically.
I am not stating that Nietzsche himself or any of his commentators
are guilty of this sort of fallacy; after all, they might think that it is help-
ful or illuminating or true to say that the primary mode of evaluation
of lives is aesthetic, but not because it follows from the fact that our
free activity exemplifies aesthetic value. I simply want to identify the
Aesthetic Fallacy as such; it highlights a problematic route to the con-
clusion that our lives are to be evaluated primarily as works of art—a
conclusion I do not in any case endorse. Perhaps someone could argue
convincingly that it does indeed follow from the relevant activity’s hav-
ing a certain nature to the product’s primarily having that same sort of
value, or that it follows from the relevant activity’s having this specific
nature—aesthetic—that its product must primarily be evaluated aes-

thetically. In this case what I have called “The Aesthetic Fallacy” would
not in fact be a fallacy. But I am not sanguine about the prospects for
such an argument, especially given the purpose-driven nature of evalua-
tion, and our multifarious purposes.
III.3. Constraining the Plausible Stories

The purpose-relativity of explanation, together with the multiplicity of
human purposes, results in a challenge to the notion that there is some
single “fixed meaning” to our lives—some master narrative that is, as
it were, “given.” Rather, on my account, one’s life-story is evaluated
along different dimensions, given different purposes. But there is an
even more basic way in which I would contest the notion that there
is a single, fixed, or given meaning to our lives. The claim that in act-
ing freely, we write a sentence in the narrative of our lives suggests
that there is a single story of our lives—a given narrative to which we
add sentences. But of course this is an unreasonable picture. Rather, our
bodily movements and behavior are subject to different interpretations
from different perspectives; although we try to tell our own stories,
and we do our best to offer our interpretations, we do not always get
to tell our own stories. That is, at least in certain contexts, our stories
are told by others—our behavior is at least in significant respects inter-
preted by others (and in ways contrary to our own understanding and
preferences).10
Just as I do not suppose that there is a purpose-independent notion
of evaluation or a “fundamental” or “privileged” purpose (or set of
purposes), I do not suppose that there is a hegemonic perspective from
which our stories get told. Thus, I would qualify my somewhat over-
simplified formulation, according to which in acting freely, we write a
sentence in the narrative of our lives. More carefully, perhaps, in acting
freely we constrain the plausible stories of our lives. Whereas various
defensible interpretations will still be possible, our behavior can signifi-
cantly constrain the plausibility of some of these interpretations, thus
limiting the range of reasonable life-stories.
Now perhaps this is somewhat deflating. You had perhaps thought
of yourself as triumphantly writing the sentences in the book of your
life, but now you are merely constraining the admissible stories—cer-
tainly a comedown! This picture seems to diminish an agent’s control,
and, after all, control is supposed to be the basis of moral responsibility
(on my approach). I admit that the simpler formulation is snappier and
a bit more attractive, but the more refined formulation is after all both
humbler and more realistic. Guidance control is indeed the “freedom-
relevant” basis of moral responsibility, and, as such, it is a significant
and robust sort of control. In exercising guidance control, we determine
certain aspects of the unfolding story of the universe; but it would be
manifestly ludicrous to demand that we can determine everything about

this story. I have also argued that it is unreasonable to aspire for what
I have called “Total Control,” the desire for which I have attributed to
“metaphysical megalomania.”11 The realization that we don’t (always)
get to tell our own story—that there are multiple perspectives from
which different interpretations of our behavior are offered, none of
which is privileged “in advance” or “apart from specific contexts”—is
parallel to a humbler, more realistic conception of the sort of control
that grounds our moral responsibility (guidance control).
III.4. Life and Works of Art

I have argued that there is a sense in which our lives can be understood
as stories. But why not think of our lives as poems, or plays, or other sorts
of works of art? I think the question is interesting, and the answer may
help to sharpen the view I am defending. Of course, a poem or a play—
or, for that matter, a piece of music—can tell a story or determine a nar-
rative, just as much as a novel or short story can. Clearly, the epic poems
of Homer tell rich and detailed stories. I would say that the various liter-
ary and artistic forms—poems, short stories, novels, plays, even pieces
of music—can be thought of as various ways of telling stories—various
“vehicles” for storytelling or the determination of narrative content. Our
lives, then, considered as a sequence of behaviors or even bodily move-
ments, could be thought of as a way of telling a story—a certain sort of
vehicle of narrative content (relative, presumably, to a perspective with
a given set of purposes). Strictly speaking, then, our lives are not stories,
but ways of telling stories, or, perhaps more carefully, ways of constrain-
ing admissible narrative content. They are—as with poems or plays or
novels—the vehicles of content, rather than the content itself.
This point is similar to the notion that a sentence (or perhaps a sen-
tence in a context) is the vehicle for content (say, a proposition that is
expressed by the sentence in the context). It is illuminating to distin-
guish the properties of the vehicles of expression of content from those
of content itself; for instance, some philosophers hold that the vehi-
cles of content are structured linguistically, whereas the propositions
expressed—the content itself—is not structured at all or at least not lin-
guistically structured. Of course, this particular claim is contentious;
for our purposes it is enough simply to mark the distinction between
the properties of the vehicles of content and the properties of content.
I suggested above that when I act freely, I write a sentence in the
book of my life. Perhaps this was closer to the truth than I had imag-
ined, since sentences—or their utterance or presentation—are good
candidates for the vehicles of content, rather than content itself—just as
our behavior is the vehicle for telling our stories. (In the more refined
formulation, our behavior would be the vehicle for constraining the
plausible stories—a way of limiting the admissible narrative contents.)
Even if we say that our lives are (simply put) ways of getting to nar-
ratives—or vehicles for telling stories—we might also say that the vehi-
cles themselves share some interesting properties with (say) poems.
Now of course it is difficult to specify exactly what the defining charac-
teristics of poetry are. One idea would be to specify poetry in terms of a
given set of forms; but this would leave out much contemporary, free-
form poetry. Another idea would be to define poetry in terms of the
dominance or hegemony of a particular trope: metaphor. Again, this
would seem problematic, insofar as much poetry does not seem to use
metaphor as the hegemonic trope, and many other literary forms make
heavy use of metaphor. Whereas it is very difficult to give the essence
of poetry, I believe it has to do with economy of expression. On this way of
thinking of poetry, certain lives could certainly be thought to be similar
to poems insofar as they embody a kind of elegance.
IV. CONCLUSION
Perhaps following Nietzsche, various philosophers have accepted
some version of “aestheticism” or the idea that life should be mod-
eled (in some way) on art. Of course, I have interpreted this idea in a
quite specific way, eschewing some of the more extravagant versions of
the doctrine. I have suggested that the value we place on acting freely
is connected in certain ways to the meaning of our lives. One way in
which this is true is that acting freely renders us artists in the sense
adumbrated above; in virtue of acting freely, our lives can be stories
and can have narrative value. We care about acting freely, then, to the
extent that we value engaging in artistic self-expression. Additionally,
our free actions are artistic activity of a specific kind, whereas the prod-
ucts of this aesthetic activity are typically (although not exclusively)
evaluated in moral and prudential ways; our free action is thus a spe-
cial sort of aesthetic activity. It may be that we especially care about the
sort of artistic self-expression whose product is typically evaluated in
moral and prudential terms: although I cannot here argue for this claim,
it seems plausible to me that the aesthetic activity in creating a prod-
uct with deep prudential and moral value is particularly important to
us. I would further suggest that we care especially about products that
are assessed prudentially and morally that come from this particular ave-
nue—artistic self-expression. So the value of the artistic self-expression
is enhanced by issuing in a product that is typically evaluated morally
and prudentially, and the value of such a product is enhanced in virtue
of coming from a distinctive sort of artistic activity. It might be said that
what makes our lives or life-stories so uniquely special and valuable is
that they are in the realm of the moral and prudential, but arrived at via
the aesthetic. The meaning of life, then, occurs at the intersection of the
aesthetic, moral, and prudential.
I think there is an additional feature of this sort of self-expression

(connected to the meaning of our lives) in virtue of which we deem it
valuable. In Richard Taylor’s fascinating essay, “The Meaning of Human
Life,” he argues that a crucial element in our lives’ having the distinctive
sort of meaning they possess is our power of creativity.12 He says:
If you were to learn that the rest of your life would be spent digging an
enormous hole, then it would perhaps be a reassurance of sorts to be
told that you were actually going to enjoy doing it. If, further, you were
born with, or at any early age conditioned to, a strong desire to do this,
then you would not need to have such a task assigned to you—you
would go to great lengths to gain the opportunity and consider your-
self lucky if you got it. And you would someday view the great hole
you had dug with a deep sense of fulfillment. And therein does each of
us find, in varying degrees, the very picture of his or her own life.13
Taylor goes on to say that what transforms our lives into meaningful
lives is creativity:
That one word [“creative”] sums it up, and, if really understood,
discloses entirely what is missing, not only in all the animate and
inanimate existence that surrounds us but in the lives of the vast
majority of human beings. It is also what philosophers have always
sought as godlike or what makes man, in the ancient metaphor, the
image of God. For what is godlike is not blind power, or aimless
knowledge, or unguided reason, but simply creative power. It is the
primary attribute in the very conception of God.14
Roderick Chisholm thought of the agent as a kind of Godlike first
cause. Perhaps we are indeed the images of God, but not in virtue of
having the power to create ex nihilo or to transcend the network of natu-
ral causation or even to make a certain kind of difference to the world.
Perhaps it is simply in virtue of our power of artistic self-expression
that we are creative in a way that renders us images of God, even if pale
images.15
NOTES
1. See, for example: The Metaphysics of Free Will: An Essay on Control (Oxford:
Blackwell Publishers, 1994); (with Mark Ravizza, S.J.) Responsibility and Control:
A Theory of Moral Responsibility (Cambridge, U.K.: Cambridge University Press,
1998); My Way: Essays on Moral Responsibility (New York: Oxford University
Press, 2006); and (with D. Pereboom, R. Kane, and M. Vargas), Four Views on
Free Will (Oxford: Blackwell Publishers, 2007).
2. For other strategies, see R. Jay Wallace, Responsibility and the Moral
Sentiments (Cambridge, Mass.: Harvard University Press, 1994), and Daniel C.
Dennett, Freedom Evolves (New York: Viking, 2003).
3. John Martin Fischer, “Reply: The Free Will Revolution,” part of a book
symposium on John Martin Fischer and Mark Ravizza, Responsibility and
Control: A Theory of Moral Responsibility, Philosophical Explorations Vol. 8, No. 2
(June 2005): 145–56; and “The Free Will Revolution (Continued),” Journal of
Ethics 10 (2006), pp. 315–345.
4. John Martin Fischer, “Responsibility and Self-Expression,” The Journal
of Ethics, Vol. 3, No. 4 (1999): 277–97, reprinted in My Way, pp. 106–23; and
chapter 9.
5. David Velleman, “Narrative Explanation,” Philosophical Review 112 (2003),
pp. 1–26.
6. In my essay, “Responsibility and Self-Expression,” I contended that the
value of free action is identical to the value of a kind of self-expression, but not
necessarily artistic self-expression: p. 117. My view in the text thus represents
a change.
7. See note 6 above. In “Responsibility and Self-Expression,” this worry
prompted me to propose that the value of our free action is the value of self-
expression, but not necessarily artistic self-expression. It is interesting to consider
whether such a view is importantly different from the view I defend here, that the
value of our free action is indeed a species of the value of artistic self-expression.
8. Note that, on my terminology, it is lives, but not narratives, that have
“narrative value.” Of course, narratives (and the novels, plays, and so forth that
express these narratives) can have aesthetic value (and can be evaluated along
other dimensions, including the moral dimension—but narratives (as opposed
to the lives they depict) themselves would not have “narrative value.”
9. Alexander Nehamas, Nietzsche: Life as Literature (Cambridge, Mass.:
Harvard University Press, 1985).
10. Different interpretations of behavior issue from different perspectives;
thus, I may see my behavior as having a certain meaning, as being embedded
in a certain narrative, whereas a third party may interpret it quite differently.
(Obviously, individuals occupying various different perspectives might see my
behavior as being parts of various different narratives.) These claims are about
the interpretation of the behavior and the associated content of the story. The
idea that from my perspective my behavior has a particular meaning has noth-
ing essential to do with how I experience my life. That is, various versions of what
Galen Strawson has called the Narrativity Thesis—a thesis about experience
—are entirely orthogonal to my contentions here in the text, as well as to my
contention that in acting freely, we endow our lives with “narrative value”:
There is widespread agreement that human beings typically see or live or
experience their lives as a narrative or story of some sort, or at least as a
collection of stories. I’ll call this the psychological Narrativity thesis, using the
word “Narrative” with a capital letter to denote a specifically psychologi-
cal property or outlook. The psychological Narrativity thesis is a straightfor-
wardly empirical, despriptive thesis about the way ordinary human beings
actually experience their lives. This is how we are, it says, this is our nature.
The psychological Narrativity thesis is often coupled with a normative the-
sis, which I’ll call the ethical Narrativity thesis. This states that experiencing or
conceiving one’s life as a narrative is a good thing; a richly Narrative outlook
is essential to a well-lived life, to true or full personhood. (Galen Strawson,
“Against Narrativity,” Ratio XVII (4) (2004), pp. 428–52, esp. p. 428.
My theses are about the structure and etiology of value, whereas Strawson’s
theses are about experience.
11. John Martin Fischer, “The Cards That Are Dealt You,” Journal of Ethics
(special issue in honor of Joel Feinberg) Vol. 10, Nos. 1–2 (2006): 107–29; also,
see Fischer et al., 2007.
12. Richard Taylor. 1981. “The Meaning of Human Existence,” in Burton
M. Leiser, ed.,Values in Conflict: Life, Liberty, and the Rule of Law. New York:
MacMillan Publishing. pp. 3–27
13. Ibid, pp. 23–4.
14. Ibid, p. 24.
15. This essay is a substantially revised version of: John Martin Fischer,
“A Reply to Pereboom, Zimmerman, and Smith,” part of a book symposium
on John Martin Fischer, My Way: Essays on Moral Responsibility, Philosophical
Books Vol. 47, No. 3 (2006): 235–44. In her insightful contribution, Angela Smith
encouraged me to acknowledge more explicitly the social dimensions of moral
responsibility. I concede that my development of the account of moral responsi-
bility might suggest an overly individualistic picture, but I do not believe that,
in the end, I am committed to a problematic sort of atomism. I certainly grant
that the statements we care about (in acting freely) are typically parts of “conver-
sations.” In making such statements, we make connections—connections with
other people and even causes “larger than ourselves” that are valuable parts
of meaningful lives. I am inclined to accept the idea that we care about writing
sentences in the books (stories) of our lives (on the simpler formulation), where
these sentences are typically parts of conversations with others we care about.
In thus making a statement, I make a connection. In writing my story, I help to
write our story. My way becomes a part of our way.
I am grateful to Neal Tognazzini for thoughtful comments on a previous
version of this essay. Also, I have learned from conversations with Howard
Wettstein on the topics treated here.
The seminar is nearly over. The video monitors are bland and the surgeons are cleaning
up and filing out into the hallway. Marilena replaces the white cloth on her cadaver’s
face; about half the surgeons do this. She is conscientiously respectful. When I asked her
why the eyes of the dead woman had no pupils, she did not answer, but reached up and
closed the eye-lids. As she slides back her chair, she looks down at the benapkined form
and says, ‘May she rest in peace.’ I hear it as ‘pieces,’ but that’s just me. (Mary Roach,
Stiff: The Curious Lives of Human Cadavers)
Index
A Suits’ versions, 107–108

Ackrill, J.L., 92 Biological methods, immortality.
acting freely, 145–146 See Body transfers
aesthetic value, 145 Birth and death
agent-causation, 20 asymmetry of possibility view
Allen, W., 32 conscious entity, 70
Anderson, P., 101 harm, components of, 70
André, 86, 87, 92 harm of death and continuing
Anscombe, G.E.M., 132, 135, 140 life, 69
Aristotle, 135 Kripkean view, 71
Aristotle’s literature, 135 attitudinal asymmetry, 73
artistic self-expression, 145 deprivation account, 65
Asimov, I., 99–101 Kaufman and Belshaw’s solution
atomistic serial model, 96 asymmetry of possibility, 66–69
Austin, J.L., 48 psychological continuum, 68
selfhood, 67
B thick and thin characteristics, 66–67
Bambrough, R., 92 meaning of, 64
“Banquet Argument” Nozickean spore, 65–66
bounded temporal srtucture, 123 Parfitian general asymmetry, 74
immortal life, 121–122 puzzling aspect of family, 64
Barnes, J., 37 xenophobia, 73
Bear, G., 101 Blum, L., 140, 143
Belshaw, C., 59–61, 62, 66–69, 72, 76 body transfers, 98–99
Belshaw’s conservation claim Borges, 160
asymmetric attitude, 59 Bretall, R., 92
different personality, 59–60 Bricker, P., 35, 103
emotional scars, 60 Broad possibility, notion of
Benford, G., 99, 101 possibility, 42
Betrayal-behind-one’s-back Brueckner, A.L., 5, 8, 9, 23, 48, 49,
Frankfurt-style version, 39–40 73, 143
Hetherington’s critique, 110–111 Buss, S., 125
McMahon’s examples, 106
Nagel’s philosophical C
views, 37–38 ‘cadence,’ 153–154
Nagel’s version, 106 Campbell, J., 126
Nussbaum’s disagreement, 105 Cantor, G., 103
secret betrayal, 109–110 Čapek, K., 79
stroke/accident, 112 Childhood’s End, 95–96
179
180 Index
Chilshom, R., 24 DePaul, M., 141, 143

Chisholm, R., 21, 22, 175 Derrida, J., 140
Chuang Tsu, 142 “Dialectical Stalemate”
Clarke, C., 94, 95, 97 black hole, 8
Cleaver, W., 158 “Consequence Argument,” 112–113
Coats, A., 143 Free Will debates, 116–117
connected-lives serial model, 97 Nagel-type examples, 118
“Consequence Argument,” 113–114 “Principle of the Transfer of
continuity Nonresponsibility,” 115–116
immortality, 16–17 Dickens, 143
level features, 15 Dolinko, D., 143
standard and non-standard Dworkin, G., 46, 143
subclasses, 15
standard evils, 14 E
Copp, D., 49, 143 ‘endlessly fascinating,’ 84
counterfactual intervener, 39 Epicureans, 104
Croesus, 161 Epicurus, 4, 5, 8, 9, 14, 28, 29, 31, 35, 37,
Curl, R., 76, 93 42, 43, 64, 103
ethical reflection, 136
D Experience Requirement I & II, 38
David, L., 24, 158 Experiential blank after death, 28
death
alternative to, 34 F
asymmetric attitudes, 28 Fansler, K., 163
controversial metaphysical Fay, B., 163
claim, 29 Feinberg, J., 46
deprivation of good things, 31 Feldman, F., 22, 64, 76
different cases, 30–31 Fischer, J.M., 9, 23, 24, 27, 37, 46–48,
eternal torment, 27 51, 61, 63, 69, 73, 76, 77, 79, 93,
experiential blank, 28, 33 103, 110, 125–126, 129, 140, 145,
Parfit’s principle, 33 161, 163, 164, 165, 176, 177
unpleasant experiences, 28, 32 Fitzgerald, Z., 23
death, badness and impossibility of Frankfurt, H.G., 6, 7, 9, 13, 18, 19, 23,
experience 24, 40, 47, 125
betrayal-behind-one’s-back, 37 “Frankfurt-type” counterexamples, 105
counterfactual intervener, 39 Frankl, V., 164
deprivation of, 46 Free Will debates, 116–117
Euthyphro-type point, 45 Furrow, D., 143
Experience Requirement
I and II, 38 G
implication, dialectical situation, 44 Gallie, W.B., 153
possibilities, 41–42 Geach, P., 133, 141
posthumous and prenatal Gernsback, H., 98, 101
nonexistence, 45 Gibson, W., 99
Principle of Alternative Possibilities, 47 Ginet, C., 114, 152
space and time analogy, 46 Glannon, W., 45, 47, 48, 125
Values Connect with Feelings Glidden, D., 143, 164
(VCF), 38 Golob, E.O., 163
Dennett, D.C., 24, 175 guidance control elements, 166
Index 181
H baby-switching, 54–55
Haldeman, J., 99 electronic stimulation (“brain
Hallie, P., 140 zap”), 57
Heidegger, 9, 143 Eskimo example, 55–56
Heilbrun, C.G., 163 judgments and preferences, 56
Heil, J., 143 psychological conception, 52
Heinlein, R.A., 94, 101, 135 “thick” and “thin”
Hellenistic philosophy, 103 individuals, 53, 58
Herman, B., 141 Kelper, A., 121
Hershenov, D., 76, 77, 162, 164 Kierkegaard, S., 88, 89, 92
Hetherington, S., 110, Kierkegaard’s spiritual
111, 112, 125 and religious experiences, 89, 91
Homer, 173 King, S., 21, 24, 98
Horowitz, T., 142 Knopf, A., 164
Horwich, P., 101 Kundera, 143
Hume, D., 157
hypothetical examples L
criticism, 132–134 Levain, J., 129–133, 140
preliminary points, 134–135 Lewis, R., 119, 126
reply to criticism, 135–137 Life and death
aesthetic activity, 13
I asymmetry in, 9
immortality attitudes to, 9
atomistic serial model, 96 authorship, God and freedom
body transfers, 98–99 agent-causation, 20
connected-lives serial model, 97 cosmological argument, 20
nonatomistic model, 95 God’s sovereignty, 21
nonbiological methods of, 99–100 brain-injury, 7–8
nonserial atomistic form, 97–98 conscious awareness, 13
problems faced with, 100 continuity
science fiction, 93 immortality, 16–17
taxonomy, 94 level features, 15
tedium of, 79 standard and non-standard
intellectual enquiry, 80 subclasses, 15
“intrinsic content,” 133 standard evils, 14
deprivation, 5
J Dialectical Stalemates, 8
James, H., 135 Epicurus and Lucretius
Jones, W., 7, 162, 164 views, 4
Juan, D., 82 experiential blank, 4
Jules, G., 23 Frankfurt-cases, 7
loss, betrayal, deception, and
K ridicule, 6
Kagan, S., 141 mirror image of time, 4–5
Kamm, F.M., 48, 141 moral responsibility, 10
Kane, R., 175 narrative/story value, 11–12
Kaufman, F., 52–59, 61, 62, 66–73, 76 narrative value, 12
Kaufman’s proposition Principle of Alternative Possibilities, 6
critique of relationship between, 12
182 Index
Life and death (Continued) McKenna, M., 23

self-expression McMahan, J., 23, 46, 125
arbitrariness, 18 McMahan, K., 143
Close Calls vs. Clear Cases, McMahon, J., 7, 106
17–20 McMahon’s examples, 106
creativity and reflective Methuselah’s children, 95
endorsement, 20 Metz, T., 161, 164
make-a-difference view, 19 Mickey, E., 120
unpleasant experiences, 5 Mink, L., 153, 163
Life, meaning and stories “Modern Moral Philosophy,” 132
creativity, 175 Moody, T., 120
freedom, varieties of value Moral responsibility
artistic/aesthetic value, 168 acting freely, 167
evaluation, 168 framework elements, 165
guidance control components, 170 guidance control, 166–167
life-story, 169 Strawsonian concept, 165
self-expression, 167, 174 Munzer, S., 143
implications and refinements Murphy, G., 23
aesthetic fallacy, 171–172
constraining stories, 172–173 N
metaphysical megalomania, 173 Nachman, R., 142
narrative dimension of values, Nagel’s Euthyphro-type point, 45
170–171 Nagel’s philosophical views, 37–38
poetry, 174 Nagel, T., 5, 6, 8, 22, 23, 28, 33, 34,
vechicle of contents, 173 37–41, 46, 48, 49, 52, 61,
moral responsibility, framework 63–65, 104–107, 111, 116–119,
elements, 165 125, 126
Frankfurt-Style examples, 166 Nagel-type examples, 118
freedom-relevant condition, narrative content, 148
165–166 characteristic features, 146
guidance control elements, Velleman’s notion
166–167 goods and evils relation, 147
narrativity thesis, 176–177 heroic medical treatment
Littlefield, 34, 46 and passive euthanasia, 149
Llarull, G., 160, 164 misfortune effects, 148
Locke, J., 166 physician-assisted
Lucretius, 4, 8, 9, 28, 29, 31, 34, 64, 65, suicide, 149–150
73, 104, 118, 119, 121 narrow possibility, notion of
Lucretius’s “profound insight” possibility, 42, 48
Kelper’s views, 121 Nassbaum, M.C., 11, 104, 112, 116, 117,
views about death, 118–119 118, 122, 124, 127, 134, 135,
vs. Lewis views, 119–120 137, 138, 140, 141
Nehamas, A., 176
M Nietzsche, 171, 174
Mancini, 25 Niven, L., 98
Marshall, B., 63 nonatomistic model, 95
Martin, A., 99 ‘non-disappointing’ self-exhausting
Massey, G.J., 142 pleasures, 85–86
McCaffrey, A., 98 nonserial atomistic form, 97–98
McCain, 7 Nozick, R., 5, 6, 22, 34, 46, 65
Index 183
Nussbaum’s critique, 138 Q

Nussbaum’s reconstruction Quality-of-life, 149
“Frankfurt-type” counterexamples,
105–106 R
main arguement and discussion, Raphals, L., 164
104–105 Rauti, A., 143
Nye, J.L., 98 Ravizza, M., 23, 91, 92, 126, 140, 142,
143, 165, 175, 176
O Reath, A., 162, 164
Obama, 7 Reck, E., 159, 164
Odysseus, 61, 62 relativistic immortality, 99
O’Rourke, M., 126 Rorty, R., 143
Otsuka, M., 48, 49, 141, 143 Rosenbaum, S.E., 38, 40, 42, 43, 47, 48,
Overton, L., 125 118, 125
Rosenberg, A., 143
Rowman, 34, 46
P Rucker, R., 101
Parfit, D., 8, 29, 30, 31, 33–35, 34, 35,
68, 73, 77, 92, 101 S
Paul, J., 34, 46 Sauvage, P., 140
Penner, T., 92 Sayre-McCord, G., 143
Pereboom, D., 175, 177 Scanlon, T., 164
Perry, J., 76 schematic philosophers’ examples, 134
personal identity, psychological Schwartz, S., 75
conception Schwitzgebel, E., 170
asymmetric attitudes, 51–52 Science fiction (SF), 93
Belshaw conservation claim Sedaris, D., 120, 126
asymmetric attitude, 59 self-exhausting vs. repeatable plea-
different personality, 59–60 sures, 88
counterfactuals and possibility self-expression, 145
claims, 61 arbitrariness, 18
deprivation account, death’s Close Calls vs. Clear Cases, 17–20
badness, 51–52 creativity and reflective
Kaufman’s proposition endorsement, 20
critique of, 53–57 make-a-difference, 19
psychological conception, 52 Seranella, B., 24, 25
“thick” and thin individuals, Shelley, M., 98
53, 58 Shier, D., 126
Pettigrove, G., 61, 62, 69, 76 Shiffrin, S., 143
philosophical puzzles, immortality, Silverstein, H.S., 38, 40,
90–91 47, 48, 49, 125
physician-assisted suicide, 149–150 Sklenar, D., 62
Pitcher, G., 92 Slote, M., 147, 161
Placencia, N., 164 Smith, A., 177
Plath, S., 37 Smith, B., 127
Principle of Alternative Possibilities Solomon, D., 140, 143
(PAP), 6 Sorenson, R., 142
“Principle of the Transfer of Speak, D., 9, 51, 69, 76
Nonresponsibility,” 115–116 Spinozistic ideas, 81
Proust, M., 135 Strawson, G., 176, 177
184 Index
Strawson, P., 24 moral responsibility, 157

streamlined examples storytelling, 154–156
criticism, 132–134 virtue and community
first version of distress and animated conversation,
worry, 137–138 131–132
preliminary points, 134–135 hypothetical and streamlined
reply to criticism, 135–137 examples
second version of criticism, 132–134
worry, 138–139 preliminary points, 134–135
Stuart Hampshire’s reply to the criticism, 135–137
formulation, 81 Le Chambon village, 130
Sykes, R., 120, 121
W
T Wallace, R.J., 24
taxonomy, immortality, 94 Walsh, B., 169
Taylor, R., 175, 177 Watson, G., 20, 24, 164
Tognazzini, N., 164, 177 Weirob, G., 76
trans-omega longevity, 103 Weller, T., 140
Wettstein, H.K., 143, 177
U White, 7, 39, 41, 48, 108
unconscious motivations, 81 Widerker, D., 23
Wilde, O., 98
V Wilkes, K.V., 140
Values Connect with Feelings (VCF), 38 Williams, B., 9, 23, 69, 76, 79–84,
Van Inwagen, P., 113–114, 126 88–90, 91, 92, 93, 95, 101,
Van Inwagen’s examples. See 122, 127, 164
“Consequence Argument” Williams’s attractiveness conditions
Vann, R.T., 163 ‘endlessly fascinating,’ 84–85
van Vogt, A.E., 98 immortality, 82–83
Vargas, M., 175 Kierkegaard’s spiritual and religious
Velleman, J.D., 10, 11, 12, 21, 23, 24, experiences, 89–90
146, 147, 148, 150, 152, 153, 154, ‘non-disappointing’ self-exhausting
155, 156, 160, 161, 163, 167, 176 pleasures, 85–86
Velleman’s narrative notion self-exhausting vs. repeatable
momentary well-being levels pleasures, 87–88
acting freely, 152 Wittgenstein, 11
death, 152 Wood, O.P., 92
goods and evils relation, 147 Wright, L., 143
heroic medical treatment
and passive euthanasia, 149 X
misfortune effects, 148 xenophobia, 73
nonhuman animal, 150–151
physician-assisted Y
suicide, 149–152 Yeomens, C., 164
social and natural sciences Yourgrau, P., 48
acting freely, 156
for clinical psychologists, 153 Z
emotional pattern/cadence, 153 Zimmerman, D., 49, 177

John Martin Fischer - Our Stories - Essays On Life, Death, and Free Will-Oxford University Press (2009)

Uploaded by

Copyright:

Available Formats

John Martin Fischer - Our Stories - Essays On Life, Death, and Free Will-Oxford University Press (2009)

Uploaded by

Document Information

Original Title

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

John Martin Fischer - Our Stories - Essays On Life, Death, and Free Will-Oxford University Press (2009)

Uploaded by

Copyright:

Available Formats

What is the book about based on the table of contents?

What is the book about based on the table of contents?

What publishing information is provided about the book?

What publishing information is provided about the book?

Our Stories

Copyright © 2009 by Oxford University Press, Inc.

Published by Oxford University Press, Inc.

Printed in the United States of America

1. Introduction: Meaning in Life and Death 3

I am very grateful to Neal Tognazzini and Patrick Todd for their

Permission to reprint the following articles is hereby acknowledged:

All around me are familiar faces

Let us hope, however, that there is a method to the madness—or at least

dies (insofar as it can involve pain and suffering).1 Sometimes death is

the metaphysics of death.12 It will be helpful to have a version of a

your daughter is trekking in the Himalayas and tragically dies in an

insofar as death deprives us of future pleasures, there is a reason to be

What exactly is a narrative or—in the strict sense—a story? After

certain structural features, or perhaps it is a matter of the relationships

including (but, of course, not limited to) elegance and comprehensive-

Level Two features, such as desirability or goodness become greater.

III. SELF-EXPRESSION: SOME SPECULATIONS

III.1. Close Calls Versus Clear Cases

free and morally responsible; typically, such philosophers will claim

of self-expression. Both parties to the dispute might be understood as

III.2. Authorship, God, and Freedom

of causation or our ordinary explanations is if there were an entity with

and complex, but perhaps it would be worth considering the idea

11. The original presentation is in Harry Frankfurt, “Alternate Possibilities

Why Is Death Bad?

Anthony L. Brueckner and John Martin Fischer

I ﬁnd this in my diary, written twenty and more years ago:

I. WHY IS DEATH BAD?

Consider Parﬁt’s example:

I am in some hospital, to have some kind of surgery. This kind of

Parﬁt’s claim is that it seems to be a deep-seated feature of us that

reasonable to take temporally symmetric attitudes toward regrettable

The plausibility of the suggestion may come from a psychological

II. WHY DEATH IS BAD

Death, Badness, and the

John Martin Fischer

I. NAGEL AND THE CRITIQUE OF EPICURUS

Experience Requirement I (ER I): An individual can be harmed by

II. A DEFENSE OF NAGEL: NAGEL MEETS FRANKFURT

In this Frankfurt-style version of Nagel’s betrayal example, I further

III. CLARIFICATION: TWO KINDS OF POSSIBILITY

IV. IMPLICATION: DEATH’S BADNESS

the inadequacy of ER II, it follows that Rosenbaum’s attempt to provide

unpleasant experience coming from some offensive sensory stimulus).

to a proponent of the existence requirement to explain its basis, if it is

laws of physics) that you experience anything bad as a result of the

Death and the Psychological

John Martin Fischer and Daniel Speak

I. THE PROBLEM: OUR ASYMMETRIC ATTITUDES

One could of course address this apparent problem in various ways.

II. THE PSYCHOLOGICAL CONCEPTION OF THE PERSON

require wholesale changes in the past: various persons existing and

personal life of the Eskimo ‘me’ would consist of completely differ-