Tải bản đầy đủ - 0trang
discover that there are websites containing lists of “famous dead people.”
Ordinary language is ﬁlled with talk that suggests that people go on existing after
they have died (Feldman 1992, 2000, 2001).
Some philosophers, however, insist that these familiar forms of speech should
not be taken seriously. Perhaps common-sense thought about the metaphysics of
death is simply wrong. Furthermore, they point out that there are other familiar
forms of speech that suggest that common-sense metaphysics presupposes the
termination thesis. For example, we often say that we have found “the remains”
of a deceased person, rather than that we have found the person himself (now
dead) (Johansson 2005).
According to the termination thesis, when a person dies he literally ceases to
exist. On one version of this view, at death the person goes out of existence and
a new entity – her corpse – comes into existence in her place (Rosenberg 1998).
Other even more radical views have been defended. Consider a case in which a
person ﬁrst loses her capacity to distinguish herself from others; then falls into a
coma; then lingers near death for a while; then dies and is buried; and then years
later her body disintegrates. Some apparently want to claim that the person went
out of existence when she lost her “ﬁrst-person perspective” (Baker 1999, 2000).
On this view, if a person becomes seriously and permanently mentally deranged,
she is “gone.” Even if her body is walking around, eating and drinking, and
talking, the former person is no longer in existence. Others claim that the person
goes out of existence when she falls into an irreversible coma. Yet others would
say that the person persists through these psychological changes, but goes out of
existence when she dies (Olsen 1997, 2004; Johansson 2005). And yet others
would say that since the person is her body, the thing that was the person goes
on existing as a corpse until that body disintegrates (Feldman 1992, 2000, 2001;
The termination thesis ﬁgures prominently in the debate about the evil of
The Epicurean argument
A substantial portion of the philosophical literature on death is stimulated by
a line of argument due to Epicurus and subsequently defended by Lucretius.
death is nothing to us. For all good and evil consists in sensation, but
death is deprivation of sensation. … So death, the most terrifying of ills, is
nothing to us, since so long as we exist death is not with us; but when death
comes, then we do not exist. It does not then concern either the living or
the dead, since for the former it is not, and the latter are no more.
(Epicurus 1994: 29)
LIFE, DEATH, AND ETHICS
In a related passage Epicurus said that “the wise man” does not fear the
cessation of life. This has been taken to mean that it is irrational to fear
Some have maintained that the argument essentially presupposes the “experience condition” (Silverstein 1980; Rosenbaum 1993). In its simplest form, this
principle asserts that something is an evil for a person only if the person
experiences it. The argument, under this interpretation, might go like this:
(1) No one experiences his own death.
(2) If no one experiences his own death, then his own death is not an evil for
the one who dies.
(3) If his own death is not an evil for the one who dies, then it is irrational for
anyone to fear his own death.
(4) Therefore, it is irrational for anyone to fear his own death.
Critics have been quick to point out a variety of familiar counter-examples to the
experience condition. For example, it might be bad for me to be secretly
betrayed by my supposed friends, even if I am never aware of the betrayal
(Fischer 1997). Thus, premise (2) is objectionable.
A weaker form of the experience condition would not be refuted by examples
of this sort. We might maintain that something can be evil for a person only if
the person could have experienced it. Although I did not actually experience any
acts of betrayal, it seems reasonable to suppose that such acts could have been
experienced. One’s own death, on the other hand, could not have been experienced. This generates a slightly modiﬁed version of the argument:
(1) No one can experience his own death.
(2) If no one can experience his own death, then his own death is not an evil
for the one who dies.
(3) If his own death is not an evil for the one who dies, then it is irrational for
anyone to fear his own death.
(4) Therefore, it is irrational for anyone to fear his own death.
But further reﬂection makes it clear that even in this weaker form, the
experience condition remains problematic. Part of the trouble emerges when we
reﬂect on fundamental theories of welfare. If a simple form of preferentism is
true, then a person’s welfare is ultimately determined by the extent to which her
desires are satisﬁed. Anything that frustrates a person’s desires would therefore
count as being bad for her. In typical forms, this theory conﬂicts with
the experience condition even in the revised form. Suppose a seafarer desires to
be buried at sea (the burial to take place after he dies, of course!). Suppose that after
he dies, he is buried on land. The imagined form of preferentism implies that
being buried on land was bad for this seafarer, since it served to frustrate one of
his desires. However, since he was dead at the relevant time, he did not and
could not have experienced that burial.
Some theories of welfare imply that the ultimate determinants of a person’s
welfare are things that the person must experience. Sensory hedonism is a
familiar example of such a theory. If this theory is true, then a person’s
welfare is ultimately determined by his own experiences of pleasure and pain.
Such a view would imply that nothing can be an ultimate determinant of a
person’s welfare unless he could experience it. But even this theory leaves room
for unexperienceable evils. Consider a “silent stroke.” By deﬁnition, the
victim of a silent stroke is not aware of the stroke when it occurs. But of course
even a hedonist can acknowledge that the occurrence of the stroke may be a
great evil for the victim. The stroke itself is extrinsically bad for the victim in
virtue of its consequences. If the victim is worse oﬀ in terms of pleasure and pain
as a result of the stroke, then (though he does not experience the stroke
itself) the stroke is still declared to be bad for him. Thus, premise (2) is still open
The experience condition seems, then, to be indefensible (Fischer 1997).
Furthermore, the Epicurean texts do not make any explicit mention of it. Thus,
some philosophers have sought a diﬀerent interpretation of the argument.
Some commentators have understood the argument to turn essentially on
certain features of harm. They have said that something can harm a person only
if it harms him at a time. They take it that Epicurus’s point was that death cannot
harm a person at any time before he dies (since it hasn’t yet happened); nor can
death harm a person at any time after he dies (since he does not exist at those
times). Thus, the argument may look more like this:
(1) His own death cannot harm a person before he dies (because it has not yet
(2) His own death cannot harm a person after he dies (because the victim does
not then exist).
(3) If (1) and (2) are true, then there is no time at which his own death can
harm a person.
(4) If there is no time at which his own death can harm a person, then his own
death is not an evil for the one who dies.
(5) If his own death is not an evil for the one who dies, then it is irrational for
anyone to fear his own death.
(6) Therefore, it is irrational for anyone to fear his own death.
While it is obvious that in some cases a person is harmed because he has
actually received certain evils (on a hedonistic assumption, this would always
ultimately be understood to be some sort of pain), it should also be obvious on
reﬂection that in other cases a person is harmed because he has failed to receive
certain goods. This might happen, for example, in a case in which a person is
LIFE, DEATH, AND ETHICS
harmed because he never got an education, or the opportunity to vote, or the
chance to travel freely. In the case of these “harms of deprivation,” the identiﬁcation of a precise date of the harm seems far more problematic. Precisely
when, we may ask, did the uneducated person suﬀer the harm of not getting an
education? Thus, there is doubt about the notion that every harm must have a
date. So premise (4) is questionable.
Others have in eﬀect rejected premise (2). They have suggested that a person
may suﬀer the harm of early death at many times after he has died. For example,
one may claim that the victim of an early death is harmed at all post-mortem
times when he would have been happy if that early death had not taken place
(Bradley 2004, 2009; Feit 2002). It is reasonable to suppose that if his early death
actually harmed him, then there must be such times.
Another suggestion is that harms of deprivation have a kind of “universal”
occurrence. Suppose a person would have been happier if he had lived longer.
Then since every time is a time when it is correct to say that he would have been
happier if he had lived longer, we can say that every time is a time when he is
harmed by dying young. This yields another way to defend the rejection of premises (1) and (2) (Feldman 2002). Many ﬁnd this implausible (Bradley 2009,
According to a standard version of the “deprivation approach,” a person’s
death may be bad for her, and worthy of fear, because her death deprives her of
all the goods she would have enjoyed if she had lived longer (Nagel 1979).
One diﬃculty for those who adopt the deprivation approach concerns the
rationality of emotions. If we focus on “positive” harms, such as suﬀerings of
pain, we might think that there is no problem about the rationality of our
fear or hatred of harms. Thus, it might be quite rational for a certain prisoner to
fear being sent to a foreign country to be tortured. After all, it might turn
out to be a horrible experience. Some deprivations seem to be suitable objects
for fear. For example, if someone threatens to steal my car, or to kidnap my
child, then I may reasonably fear the threatened loss. But other instances of
deprivation seem troublesome. Suppose I know quite well that I will not be
granted magical powers, such as the power to ﬂy. Suppose I believe (correctly, let
us imagine) that as a result of not being able to ﬂy, I will miss out on a remarkable sequence of very enjoyable experiences that I would have had if I had been
granted the power to ﬂy. Thus, not being granted the power to ﬂy deprives me
of some great goods. Yet it would be completely irrational for me to fear this
deprivation, or to allow myself to fall into despair because of it. The question,
then, is this: How are we to distinguish between (a) deprivations (such as those
resulting from early death) that seem to merit a strong emotional response and
(b) deprivations (such as those resulting from the failure to be given magical
powers) that do not seem to merit a strong emotional response? Why is fear
rational in one case but not in the other (Draper 1999, 2004)? This puzzle has not
yet been satisfactorily solved.
Lucretius and the mirror of time
In a widely quoted passage in On the Nature of Things, Lucretius says:
Look back at the eternity that passed before we were born, and mark
how utterly it counts to us as nothing. This is a mirror that Nature holds
up to us, in which we may see the time that shall be after we are dead. Is
there anything terrifying in the sight – anything depressing – anything
that is not more restful than the soundest sleep?
It seems pretty clear that Lucretius means to be presenting an argument here;
and it seems pretty clear that the argument somehow involves the notion that
there is nothing terrifying about the long stretch of non-existence that a person
“suﬀered” before she was born. It also involves the notion that the long stretch
of non-existence that a person will suﬀer after she dies is somehow the “mirror
image” of that prenatal time. The conclusion is not precisely stated, but some
have taken it to be this: it is irrational for a person to fear her own death.
Let us assume that the argument can be understood in this way:
(1) It would be irrational for a person to be emotionally upset about her
(2) If it would be irrational for a person to be emotionally upset about her
prenatal non-existence, then it would be irrational for a person to be
emotionally upset about her post-mortem non-existence.
(3) Therefore, it would be irrational for a person to be emotionally upset
about her post-mortem non-existence.
Replies to this argument take several forms. Some have (in eﬀect) rejected (2).
They have said that post-mortem times are not the mirror image of prenatal times.
The failure of symmetry (it has been alleged) makes it irrational to fear prenatal
non-existence, but allows it to be rational to fear post-mortem non-existence.
One alleged diﬀerence between these two periods of time might be this: whereas
those post-mortem times are times at which a person might have existed (after
all, the person could have survived his disease; he could have died later), the
prenatal times are not times at which the person might have existed (because he
could not have been born earlier than he in fact was born) (Nagel 1979).
This sort of reply seems to turn on indefensible metaphysical assumptions
about time and personal identity. As several critics have pointed out, even if a
person essentially arose from the union of a speciﬁc sperm and egg, still it is
metaphysically possible that that sperm and egg could have combined at an earlier date. This is especially clear in cases involving frozen sperm and egg that are
chosen for in vitro fertilization. Such fertilizations may occur earlier or later, at
LIFE, DEATH, AND ETHICS
the discretion of parents and doctors. If earlier fertilization had happened, the
person would have come into existence earlier (Feldman 1991: 222; Breuckner
and Fischer 1993: 215).
Others (e.g. Parﬁt 1984) have pointed out that we seem to care deeply about future
harms and beneﬁts, but to be nearly indiﬀerent to past harms and beneﬁts. The
“amnesiac patient” example illustrates this. Suppose you wake up in a hospital.
Your memory is a blank; you have no idea how you got there. The nurse informs
you that either you had a very painful operation last night but have forgotten it as a
result of some amnesia-inducing drugs, or else you will be getting a somewhat
less painful operation later today. If you have the operation later today, you will get
an amnesia-inducing drug and as a result you won’t remember the pain afterwards. It seems perfectly natural for you to prefer that the painful operation be
in the past and forgotten. This suggests that, other things being equal, we prefer
that our pains be in the past – “over and done with.” A corresponding thought
experiment would suggest that we prefer that our pleasures be in the future.
These thoughts suggest a diﬀerent way in which the future and the past are not
mirror images of each other. We fear dying prematurely because we care about
future pleasures. Failing to receive those pleasures seems to be contrary to our
welfare. On the other hand, we don’t care about having been born late because
we don’t care about any pleasures that we lost as a result of being born later than
we might have been (Fischer 2006a, b; Breuckner and Fischer 1986, 1993).
Critics may acknowledge that in fact we do have the temporal bias that Parﬁt
describes; but they may go on to say that this does not constitute a refutation of
Lucretius’s point. The point, after all, concerned the rationality of our attitudes.
Lucretius seems to have claimed that in light of the metaphysical symmetry of
past and future, there is no justiﬁcation for our asymmetrical attitudes. He
reminded us that it would be irrational to fear past non-existence. He wanted to
know how, given the metaphysical symmetry of past and future, it could
be rational to fear future non-existence while it is irrational to fear past nonexistence. Merely pointing out that in fact we do have these asymmetrical attitudes seems not to be a reply at all (Kaufman 2009/1999).
A diﬀerent sort of reply to Lucretius would involve rejecting premise (1). Perhaps
we can agree that most of us are indiﬀerent to the pleasures we lost as a result of being
born too late. But we might claim that there would be nothing irrational about it if
someone were distressed about these lost pleasures. Perhaps the bias toward the
future is common but unjustiﬁed. Maybe it would be reasonable to be unhappy
about all lost pleasures, whether past or future (see also McMahan 2006).
The conquest of death
Whether the fear of death is rational or not, it appears that this is a fear that
people have experienced throughout history. Some religious traditions seem
almost to be built upon this fear, since those traditions claim that true believers
will gain a chance to evade death. For some it seems obvious that an eternal life
would be vastly preferable to a short life followed by an endless stretch of nonexistence.
Of course, no reasonable person would prefer an eternal life of suﬀering in
hell. But what shall we say about an eternal life ﬁlled with typical earthly pleasures and pains? Would this be better than a ﬁnite life followed by nonexistence? Those who fear and hate death apparently would prefer to go on living
their ordinary lives no matter how old they may become. But is this preference
Bernard Williams (1973) claimed that it is not obvious that eternal earthly life
is an attractive prospect. After a suﬃciently long time, the victim of eternal life
would become bored. He would have done everything that a person with his
tastes and preferences could possibly enjoy doing. It would be no fun to have to
do it again. On the other hand, if he were to change his tastes and preferences,
further enjoyment might be possible, but then it would be unclear whether the
original person had survived. For with the imagined dramatic change in tastes
and preferences, the previous person would have been transformed into a
new person. Thus, if a given person retains his identity long enough, he will
eventually become bored. Eternal life for a persisting person is thus seen to be
Critics have pointed out two main problems with Williams’ argument. First, it
appears that Williams has underestimated the individual’s capacity for growth.
Surely a person can retain his identity while gradually coming to enjoy new
experiences (Fischer 1994; Rosenberg 2006). Second, there seem to be some
pleasures that enhance the value of life even when they have been repeated
thousands of times. Surely a person might enjoy (for example) a fresh sexual
encounter every day forever. Thus, while Williams may be right to point out that
not every eternal life would be desirable, he seems to have gone too far when he
claimed that every such life would have to become painfully boring.
The ethics of killing
Suppose we think, following Epicurus, that early death cannot harm a person.
Suppose we think, following Lucretius, that early death is no more to be feared than
late birth. Then we seem to put ourselves into an odd position with respect to killing.
For if we can kill a person painlessly, and if his death will not make others
unhappy, it is not clear why there would be anything wrong with killing him.
But this result is in direct conﬂict with a deeply entrenched component of
common-sense morality. We all ﬁnd something intuitively plausible about the
idea that killing another person is morally impermissible. Of course, we recognize that there are some circumstances in which it may be morally permissible to
LIFE, DEATH, AND ETHICS
kill another person – as, for example, in cases of justiﬁed self-defense when there
is no other way to save oneself from a vicious criminal assault. Some philosophers maintain that while it is generally wrong to kill others, it may be morally
permissible to kill oneself in order to avoid unremitting, pointless suﬀering.
Others seem to think that it may be morally permissible to kill (or to allow to
die) a person who has fallen into a horrible state in which continued existence is
of great disvalue. This is especially compelling in the case of someone who
repeatedly and emphatically has asked for help in ending his life. There are also
questions about killing enemy combatants in a justiﬁed war and questions about
executing convicted murderers, rapists, and traitors. But in the standard case
not involving any of these exceptions, it seems wrong to kill people. This gives
rise to a profound question: Precisely when (if ever) is it morally permissible to
kill a person? To answer this question, it seems that we must answer a yet more
fundamental question: When it is morally wrong to kill a person, precisely why
is it wrong?
Some philosophers proceed as if they think that these questions about the
morality of killing can be studied more-or-less in isolation, but a more reasonable assumption is that one’s view about the morality of killing should be a
corollary of one’s view about the morality of behavior in general. If we don’t
know what makes actions in general wrong, it’s hard to see how we can know
what makes acts of killing wrong.
Act consequentialism of the simplest variety is the view that an action is
morally right (permissible) if and only if no alternative would lead to a better
outcome for all aﬀected. This theory oﬀers a simple explanation of the wrongness of killing, when it is wrong. According to the theory, it is wrong to kill
someone (when it is wrong) precisely because killing him leads to an outcome
that is on the whole worse than the outcome of some alternative course of
behavior. Familiar examples involving killing have been thought to prove this
view untenable. Consider a case in which several decent people are in need of
organs for transplantation. If they get those organs, they and their friends will be
happier. Suppose a healthy but not very happy person is available. Suppose he
has very few friends. If he is killed and his organs are transplanted several others
will be happy. Very few will be made unhappy. In this case, act consequentialism
seems to imply that it would be morally permissible (even obligatory) to kill the
innocent victim so as to keep several others alive. This implication is widely seen
as repugnant, and so the act consequentialist approach has been rejected.
More extreme objections have been raised. It has been thought, for example,
that act consequentialism implies that there is a moral obligation to kill any
person whose unhappiness is dragging down the collective welfare level. It may
appear that the community as a whole would be better oﬀ with such people out
of the way, provided that it is possible to kill them painlessly (Henson 1971).
The theory also seems to imply, somewhat implausibly, that we have a moral
obligation to bring new people into existence whenever the community as a
whole will be better oﬀ with such people in existence. This obligation would
persist even in cases in which the introduction of the new people would decrease
the welfare of everyone already in existence, provided that the total welfare of
the community as a whole (including the new people) would be increased. In
response to cases of this sort, some have been moved to consider “personaﬀecting” forms of consequentialism. On these theories, the utility of an action is
determined by the pleasures and pains of people who would exist whether the
action is performed or not. Thus, the hypothetical pleasures of people who
would be brought into existence if the action were performed, but who otherwise would never exist, would be disregarded. This sort of view faces new diﬃculties of its own. It seems to permit the creation of a new person who
would lead a miserable life, provided that his introduction would make already
existing people happier. Other versions of the person-aﬀecting form of consequentialism have been proposed, but all are currently controversial (Arrhenius
2003; Roberts 2003).
Advocates of the deprivation approach may claim that acts of killing are
wrong (when they are wrong) because they inﬂict serious harms of deprivation on
their victims. This view is open to many objections. Suppose an innocent victim
is being brutally attacked by a vicious killer. Suppose the victim can defend himself
only by killing his attacker. But suppose that the attacker would be harmed by
being thus killed. Surely in such a case it would be morally permissible for the
victim to defend himself, even though such defense would impose a serious
harm (early death) on the attacker. Harm to the one killed cannot be the sole
determinant of the normative status of acts of killing.
Some have suggested (perhaps following Kant) that every person, merely in
virtue of the fact that he is a person, deserves a certain sort of moral respect.
Killing anyone is thus morally wrong because it inevitably involves a violation of
that demand for respect.
When formulated in this stark way, this view seems to violate some very
intuitive general principles about moral obligation in general. Consider a case in
which some agent has a choice of killing one unfortunate person or another. He
can’t avoid killing someone or other. The view in question would then imply
that the agent is doomed to do something morally wrong no matter what he
does. Each alternative involves a violation of the alleged demand for respect.
This seems to conﬂict with the well-grounded moral intuition that there is always
something that morality permits. Perhaps morality permits whichever act would
involve the smaller total amount of disrespect for humanity.
In some cases morality seems to require killing. For example, in an otherwise
just war, military decision-makers may correctly see that on balance far more
lives will be saved if a certain target is destroyed. They may foresee that innocent
civilians will die as an undesired and unintended result of their attack. Yet for all
that, it may still be the morally right thing for the military decision-makers to do.
This seems to show that the moral status of an act of killing depends upon
LIFE, DEATH, AND ETHICS
considerations beyond mere harm to the victim or respect for the victim’s
We may imagine an eclectic view. Suppose we say, ﬁrst, that harm to the
victim constitutes one reason to avoid killing someone; but suppose we say, in
addition, that harm to the community as a whole constitutes another reason to
avoid killing someone. And suppose we say, furthermore, that potential violation of the demand for respect of persons is yet a third source of reasons to
avoid killing. Each of these considerations, we may say, constitutes some defeasible reason to avoid killing. Each consideration may thus be seen as helping to
make such acts of killing prima facie wrong. In any speciﬁc case, the overall
moral status of an act of killing would be determined by the weights in that case
of all of these factors (and perhaps others) somehow taken together (McMahan
While this approach may have the virtue of accommodating a variety of
intuitions about the morality of killing, it confronts a serious objection. The
view seems hardly more than a restatement of the problem rather than a solution. It reminds us that we have conﬂicting pre-theoretic intuitions about the
morality of killing. It then tells us that the actual moral status of any act of killing is somehow determined by some unexplained interactions among the factors
that pre-theoretically seemed relevant. This seems to leave us with precisely the
problem with which we started. We still don’t have any general account of what
makes acts of killing all things considered wrong when they are wrong. We still
don’t have a helpful explanation of what makes acts of killing morally right in
those unusual cases where they are morally right (if any).
It seems, then, that moral philosophers have not yet managed to explain clearly
and precisely why it is wrong to kill people in those cases in which it is wrong.
See also Later ancient ethics (Chapter 5); Utilitarianism to Bentham (Chapter 13);
John Stuart Mill (Chapter 16); Consequentialism (Chapter 37); Contemporary
Kantian ethics (Chapter 38); Respect and recognition (Chapter 47); Ideals of
perfection (Chapter 55); Rights (Chapter 56); Ending life (Chapter 60); Population
ethics (Chapter 61); War (Chapter 67).
Arrhenius, Gustaf (2003) “The Person-Aﬀecting Restriction, Comparativism, and the Moral
Status of Potential People,” Ethical Perspectives 10: 185–95.
Baker, Lynne (1999) “What Am I?” Philosophy and Phenomenological Research 59: 151–9.
——(2000) Persons and Bodies: A Constitution View, Cambridge: Cambridge University Press.
Bradley, Ben (2004) “When Is Death Bad for the One Who Dies?,” Noûs 38: 1–28.
——(2009) Well-Being and Death, Oxford: Oxford University Press.
——(Forthcoming) “Eternalism and Death’s Badness,” in Joseph Campbell, Michael
O’Rourke and Harry Silverstein (eds) Time and Identity, Cambridge, MA: MIT Press.
Brueckner, A. L. and Fischer, J. M. (1986) “Why Is Death Bad?,” Philosophical Studies
——(1993) “The Asymmetry of Early Birth and Late Death,” Philosophical Studies 71: 327–31.
Draper, Kai (1999) “Disappointment, Sadness, and Death,”Philosophical Review 108: 387–414.
——(2004) “Epicurean Equanimity Towards Death,” Philosophy and Phenomenological Research
Epicurus (1994) The Epicurus Reader: Selected Writings and Testimonia, trans., ed. Brad Inwood
and L. P. Gerson, Indianapolis, IN: Hackett.
Feit, Neil (2002) “The Time of Death’s Misfortune,” Noûs 36: 359–83.
Feldman, Fred (1991) “Some Puzzles About the Evil of Death,” Philosophical Review 100: 205–27.
——(1992) Confrontations with the Reaper: A Philosophical Study of the Nature and Value of
Death, New York: Oxford University Press.
——(2000) “The Termination Thesis,” in Peter French and Howard Wettstein (eds) Life and
Death: Metaphysics and Ethics, Special issue, Midwest Studies in Philosophy 24: 98–115.
——(2001) “What to Do About Dead People,” in Dan Egonsson, Jonas Josefsson, Björn
Petersson and Toni Ronnow-Rasmussen (eds) Exploring Practical Philosophy: From Action to
Values, Aldershot, UK: Ashgate, pp. 41–58.
Fischer, John (ed.) (1993) The Metaphysics of Death, Stanford, CA: Stanford University Press.
——(1994) “Why Immortality Is Not So Bad,” International Journal of Philosophical Studies 2:
——(1997) “Death, Badness, and the Impossibility of Experience,” Journal of Ethics 1: 341–53.
——(2006a) “Epicureanism about Death and Immortality,” Journal of Ethics 10: 355–81.
——(2006b) “Earlier Birth and Later Death,” in McDaniel et al. 2006, pp. 189–201.
Henson, R. G. (1971) “Utilitarianism and the Wrongness of Killing,” Philosophical Review 80:
Johansson, Jens (2005) Mortal Beings: On the Metaphysics and Value of Death, Stockholm:
Almqvist & Wiksell International.
Kaufman, Frederik (2009/1999) “Pre-Vital and Post-Mortem Nonexistence,” in David Benatar
(ed.) Life, Death, and Meaning: Key Philosophical Readings on the Big Questions, 2nd edn,
Lanham, MD: Rowman & Littleﬁeld, pp. 241–64; originally published in American Philosophical Quarterly 36, no. 1 (January 1999): 1–19.
Lucretius (1965) On the Nature of Things, trans., introduction and notes by Russell Geer,
New York: Bobbs-Merrill.
Mackie, D. (1999) “Personal Identity and Dead People,” Philosophical Studies 95: 219–42.
McDaniel, Kris, Raibley, Jason, Feldman, Richard and Zimmerman, Michael (eds) (2006) The
Good, the Right, Life and Death: Essays in Honor of Fred Feldman, Burlington, VT: Ashgate.
McMahan, Jeﬀ (1988) “Death and the Value of Life,” Ethics 99: 32–61.
——(2002) The Ethics of Killing: Problems at the Margins of Life, New York: Oxford University
——(2006) “The Lucretian Argument,” in McDaniel et al. 2006, pp. 213–26.
Nagel, Thomas (1979) Mortal Questions, New York: Cambridge University Press.
Olsen, E. T. (1997) The Human Animal: Personal Identity without Psychology, New York and
Oxford: Oxford University Press.
——(2004) “Animalism and the Corpse Problem,” Australasian Journal of Philosophy 82:
Parﬁt, Derek (1984) Reasons and Persons, New York: Oxford University Press.
Roberts, Melinda (2003) “Is the Person-Aﬀecting Intuition Paradoxical?,” Theory and Decision
Rosenbaum, Stephen (1993) “How to Be Dead and Not Care: A Defense of Epicurus,”
American Philosophical Quarterly 23: 217–25; repr. in Fischer 1993, pp. 119–34.
Rosenberg, Jay (1998) Thinking Clearly about Death, Indianapolis, IN: Hackett.