The Geography of Genethics and the Ends of Life1 - University of ...

The Geography of Genethics and the Ends of Life 1
Tim Bayne
University of Oxford and St. Catherine's College
Manor Road
Oxford OX1 3UJ
United Kingdom
tim.bayne@gmail.com
This paper is under review. Please contact the author before quoting it.
One should not confuse the craving for life with endorsement of it.
Elias Canetti, The Secret Heart of the Clock
I.
Job, so the story goes, cursed the day of his birth and the night of his conception. While
Job’s resentment at having been brought into existence was directed at God, in
contemporary life such attitudes are directed at one’s parents or the medical
establishment and are often played out in court. In what is known as a wrongful life case,
an individual with a serious genetic disorder argues that her life is not worth living, and
that she was wrongly brought into existence. Understandably, the courts have not
welcomed such cases. As one court complained, “whether it is better never to have been
born at all than to have been born with even gross deficiencies is a mystery more properly
to be left to the philosophers and the theologians.”
I cannot speak for the theologians, but philosophers certainly have not known what to do
with what, following David Heyd (1992), I will call “genethical questions”. Philosophers
1 Thanks to David Archard, David Benatar, Avery Kolers, and Neil Levy for very helpful comments on an earlier draft
of this paper. Earlier versions of this paper were presented to the philosophy departments at Charles Sturt University
(Wagga Wagga) and the University College Cork, and I am grateful to the audiences on those occasions for their
comments.
1

disagree about whether a person can be wronged or benefited by being brought into
existence, and if they can, under what conditions someone is wronged or benefited by
being brought into existence. My main goal in this paper is to clarify the logical geography
of genethics. I proceed in two stages. In the first sections of the paper I argue that
genethical judgments ought to be context-insensitive. That is, the truth-value of genethical
claims does not depend on the context of the attributer of the judgment. I then examine
three genethical models: the no-faults model, the dual benchmark model, and the parity
model. None of the three models is without problems, but it seems to me that the parity
model, according to which our judgments about the value of coming into existence ought
to be constrained by our judgments about the value of death, has the most going for it.
II.
Genethical questions arise in different contexts. One context is Job’s. We can ask whether
actual lives—our own or those of other individuals—are or have been worth living. Call
such contexts retrospective contexts. Genethical questions also occur in prospect. Would-be
procreators ask themselves what kinds of people it is permissible or reasonable to create.
Call such contexts prospective contexts.
It is intuitively plausible to suppose that retrospective and prospective genethics ought to
constrain each other in certain ways. Let us consider some cases first, and then turn to the
question of general principles.
Amy was born with a terrible disease that prevents her from enjoying many of life’s
goods. Amy faces a bleak future. She is in constant pain, and wishes that she had never
been born. Suppose that Amy’s parents deliberately conceived Amy, despite having had
good reason to believe that she would be born with this disease and that it would impair
the quality of her life in the ways that it has. It seems to me that it would be strange to
hold that it is reasonable for Amy to regret having been born and yet think that her
parents did nothing wrong in deliberately having her; after all, they knew that she would
have the kind of life she has, and they now accept that Amy’s regret at having been born
is reasonable. Judging that Amy’s regret at having been born is reasonable ought to go
hand-in-hand with judging that her parents acted impermissibly—or at least
unreasonably—in deliberately conceiving her.
Tom’s parents were in much the same situation as Amy’s parents: they conceived him in
the knowledge that they were almost certainly going to have a blind child. At the time,
they wondered whether they were doing the right thing; many of their friends and family
advised them to adopt a child, or to refrain from having children. But Tom’s parents
reasoned as follows: “Having been born, Tom is likely to be glad to have come into
2

existence. How then could it be wrong for us to bring him into existence?” Let us
suppose—as seems not unreasonable—that Tom is glad to have been brought into
existence. Tom’s parents might appeal to this fact to justify their decision to have him.
And it seems to me that their appeal has some force: if Tom’s endorsement of his life is
reasonable, then there is a prima facie case for thinking that his parents did nothing
wrong in knowingly bringing him into existence. 2
These cases suggest that first-person and third-person genethical judgments ought to
constrain each other: The fact that someone does—or would—regret having come into
existence gives one reason not to bring them into existence, and the fact that someone
does not—or would not—regret coming into existence means that one does not have a
person-regarding reason to avoid bringing them into existence. In the following three
sections I explore (and defend) the notion that retrospective and prospective genethics
ought to constrain each other. I examine three ways in which these two genethical
perspectives might differ from each other, and argue that none of these differences makes
a difference.
III.
One difference between retrospective and prospective genethics concerns the relationship
between the target of the genethical judgment and the person making the judgment (the
attributer). Retrospective genethics can take either a first-person or a third-person form.
We can hold evaluative attitudes to both our own lives and also the lives of others.
Prospective genethics, by contrast, necessarily takes a third-person form: one cannot
wonder whether or not to bring oneself into existence. Might the distinction between firstperson
and third person perspective drive a wedge between retrospective genethics and
prospective genethics?
I think not. It seems implausible to suppose that the truth-value of genethical judgements
should be a function of who is making the judgement. It seems to me that both firstperson
and third-person genethical judgments should be grounded in the same fact: viz.,
the quality of the life in question. If the quality of a person’s life is observer-independent,
it follows that first-person and third-person genethics should proceed in parallel. And it
seems to me that the quality of a person’s life is observer-independent: S’s life cannot be
worth living from A’s perspective but not from B’s perspective. I think this claim holds
2 Note that to say that Tom’s endorsement of his life is reasonable is not to say that people cannot be wrong about the
value of their own lives (to them). Even if tend to over-estimate our own levels of well-being, we nonetheless have
good reason to take seriously people’s judgments concerning their own level of well-being in asking whether or not it
would be reasonable to bring them into existence.
3

true even when B is S, that is, when one of those evaluating the life is also the subject of
the life in question. I do not deny that A and B might disagree about whether or not S’s
life is worth living. Nor do I deny that individuals generally have a privileged epistemic
position when it comes to evaluating whether or not their own lives are worth living. Nor
am I claiming that there is always a fact of the matter as to whether or not a life is worth
living.
A second difference between prospective and retrospective genethics is obviously
temporal. Retrospective judgments are made during or subsequent to the life of the target
subject, whereas prospective genethics arises prior to the birth of the target subject (if,
indeed, there is an actual target). Could temporal differences justify differences between
prospective and retrospective genethical judgments?
It is certainly true that the judgments we make about the quality of our own lives are
temporally sensitive. As Francis Kamm has pointed out, in contemplating our lives in
prospect, we seem to place a lot of weight on the goods and the bads of sensation and
experience, whereas in evaluating our lives retrospectively we emphasize the goods and
bads of achievement over those of experience; we give more weight to what we have done
than what we have suffered (1993: 34). Might these differences drive a wedge between
prospective genethics and retrospective genethics? The perspective of those asking
prospective genethical question is, in some sense, temporally prior to the target of their
deliberations. Any child that would-be procreators decide to produce will, of course, exist
subsequent to their deliberations. So, if future experiences count more than past
experiences, and/or past achievements count more than future achievements, we have the
beginnings of an explanation as to how prospective genethics and retrospective genethics
might dissociate.
Of course, we should ask whether the temporal sensitivity that Kamm has uncovered is
reasonable. It is one thing to claim that we do evaluate our lives in temporally sensitive
ways, it is another thing to claim that such attitudes are reasonable. There is certainly
something puzzling about the notion that the value of one’s life might depend on how far
through it one is. Furthermore, Kamm’s asymmetry seems to have intuitive force only in
the context of first-person judgments, and does not appear to extend to third-person
judgments. It is not clear what to make of this asymmetry between first- and third-person
judgments, but at the very least it suggests that Kamm’s intuition provides no obstacle to
the thought that we should demand constraints between different third-person genethical
judgments.
4

A final point of contrast between retrospective and prospective genethics concerns the
modal status of the respective judgments. In retrospective genethics the attributer and
target of attribution are always world-mates – that is, they exist in the same possible
world. This is not true of prospective genethics. Sometimes the attributer(s) and target
will be world-mates, as when prospective procreators become actual procreators. In other
situations the attributer(s) and target are not world-mates, as when prospective
procreators decide against procreating.
Might the modal differences between retrospective and prospective genethics provide us
with a wedge between them? I think not, for modal differences locate the distinction in the
wrong place. Rather than separating retrospective from prospective genethics, they would
distinguish retrospective contexts in which the attributer(s) and target are world-mates
from retrospective contexts in which the attributer(s) and target are not world-mates. This
would be a strange result. Whether or not a certain genethical decision is reasonable ought
not to depend on whether it is carried out!
IV.
In the previous sections I explored the notion that genethical judgments ought to be
context-invariant. In the remainder of this chapter I argue for a second kind of constraint
on genethics, viz., that there ought to be parity between our genethical judgments and our
judgments about the value of death—what I will call mortethics.
I divide genethical models into three groups: no-faults models, parity models, and dualbenchmark
models. No-faults genethicists claim that coming into existence is not properly
subject to moral evaluation, at least so far as the interests of the person that is brought into
existence are concerned. No-faults genethicists do not deny that it can be wrong to bring
someone into existence, for procreation can have adverse effects on the interests of thirdparties,
but they hold that it cannot be wrong on account of the well-being of the person
who is brought into existence. A no-faults genethicist denies that it can be reasonable to
have any evaluative attitude—either regret or satisfaction—towards having come into
existence.
Parity theorists and dual-benchmark theorists share the view that it can be wrong to bring
someone into existence on account of their well-being, and that it can be reasonable to
regret having been created. The two positions are distinguished by where they put the
genethical benchmark. To a first approximation, parity theorists insist that the level of
well-being that determines whether or not a life is worth starting is identical to that which
determines whether or not it is worth continuing. (I explain below why this is true only to
5

a first approximation.) The parity theorist insists on parity between genethics and
mortethics. Dual benchmark theorists reject any such parity. They hold that our genethical
judgments need not—and in fact should not—be brought into line with our judgments
about the kinds of lives that are worth sustaining. The dual benchmark theorist posits one
benchmark for genethics and another for mortethics. In aretic terms, the level of wellbeing
that makes a life worth creating is different from that which makes a life worth continuing.
In principle, there are two ways of being a dual benchmark theorist. On the one hand, one
could hold that the life-worth-starting threshold is lower than the life-worth-continuing
threshold. Although this version of the view is coherent, it is deeply implausible. To the
best of my knowledge, dual benchmark theorists are united in insisting that the lifeworth-starting
benchmark is higher than the life-worth-continuing benchmark. As we will
see, some dual benchmark theorists hold that the life-worth-starting benchmark is so high
as to be practically unobtainable.
V.
The leading proponent of no-faults genethicism is David Heyd. In Genethics Heyd argues
that coming into existence can be neither a harm nor a benefit. “It is equally meaningless
to resent our parents for having been born unhappy as it is to be grateful for having been
born happy” (1992: 109). Heyd thinks there is something logically problematic with the
idea that one could harm or benefit someone by bringing them into existence.
No-faults genethics flies in the face of certain robust intuitions. Whether or not we think
Job was justified in cursing the day of his birth, his actions certainly seem to have been
coherent. It seems to be possible for someone to rationally regret having come into
existence. What exactly is the problem meant to be?
The problem, says Heyd, is with the notion that it is possible to be better off in a world in
which one does not exist than one is in a world in which one does exist (1992: 122).
Indeed, the problem is with the notion that one could have any level of well-being— any
level of ‘offness’—in worlds from which one is absent. There might be an impersonal
sense of goodness in which worlds containing Job exists might be better than worlds in
which he fails to exist, but, according to Heyd, it makes no sense to say that such worlds
are better for Job. For a world to be good or bad for one, one must exist within it. Heyd
concludes: “to be cannot in itself be either good or bad, a subject of duty or prohibition, a
right or wrong” (1992: 124).
These claims should have a familiar sound to those acquainted with the Epicurean view of
death. Epicureans hold that it is irrational to attach any moral or prudential value to one’s
6

own death; one can be neither harmed nor benefited by death. Because death brings nonexistence,
the states of being dead and alive are incommensurable with respect to levels of
welfare. The dead are neither better off than the living, nor are they worse off; they simply
aren’t off at all.
This suggests that an Epicurean approach to mortethics goes hand-in-hand with a
Heydian approach to genethics. If the transition from existence to non-existence can be
bad (or good), why can’t the transition from non-existence to existence also be bad (or
good)? If the event of one’s death can be morally evaluated on account of the harm or
benefits it brings with it, why can’t the event of one’s birth?
Heyd, however, rejects the Epicurean view of death. He presents two arguments for
thinking that there is an asymmetry between coming into existence and going out of
existence. The first is that death has a subject, “an identifiable individual whose life is cut
short contrary to his or her interests” (Heyd 1992: 123). By contrast, there is no identifiable
individual who is brought from non-existence into existence: identity is always
subsequent to existence. We can say of someone who died that she was deprived of her
future, but we cannot say of a possible person who was not brought into existence that she
was deprived of existence, for only in the former case do we have an actual subject of
harm.
It is of course true that although the event of death brings about non-existence, death itself
always involves a subject who is harmed. And it is also true that possible people who are
not brought into existence are not deprived of existence. Nonetheless, Heyd’s response
leaves open the possibility that genethical judgments concerning people who have been
(or will be) born are coherent, for in such cases we do have an actual subject to which
harm (or benefit) can be ascribed. But to say this is to depart from the no-faults view, at
least to some degree. Heyd’s view, remember, is that “to be cannot in itself be either good
or bad, a subject of duty or prohibition, a right or wrong” (1992: 124).
Of course, no one is (directly) harmed by the decision not to instantiate a good life; and,
correlatively, no one is (directly) benefited by the decision not to instantiate a bad life. So
here there is an asymmetry between genethical decisions and mortethical decisions: with
respect to death, someone is (directly) affected no matter what we decide to do, whereas
whether or not someone is (directly) affected by our decisions concerning birth depends
on what we decide to do. But none of this undermines the point that if one can be harmed
or benefited by going out of existence then one can also be harmed or benefited by being
created.
7

Heyd’s second argument for an asymmetry between genethics and mortethics invokes a
desire-based account of the badness of death. “[Categorical] desires make us not only
wish to go on living (not to commit suicide), but also to be happy that we did not die two
years ago. But they cannot make us happy to have been born at all, since had we not been
born there would not have been any such categorical desires” (1992: 123). I am not
convinced that one can account for the harm of death by appeal to desire, but let me
waive that concern here, for Heyd’s argument fails even if we accept a desire-based
account of death’s badness. If frustrated desire is able to explain death’s badness then it
can also explain the badness of having come into existence. It is true that I wouldn’t have
had any categorical desires had I not been born, but it is also true that I wouldn’t have had
any categorical desires now had I died two years ago. Heyd seems to confuse the claim
that an actual person might wish to not have been born with the claim that a non-actual
person might have had wishes. 3 The latter claim is incoherent, but the former claim is not.
If it is coherent to prefer a future X over future Y on account of the fact that fewer of one’s
categorical desires are left unsatisfied in X than in Y, then it is coherent to regret that the
actual world is one in which one has many unsatisfied categorical desires (and few
satisfied ones), rather than a world in which one doesn’t (and didn’t) exist and thus has no
unsatisfied categorical desires (or, of course, satisfied ones).
Bernard Williams once remarked that “while [a man] can think egoistically of what it
would be for him to live longer or less long, he cannot think egoistically of what it would
be for him never to have existed at all” (Williams 1973: 87). Can one think egoistically of
what it would be for the world to carry on without one? I think so. One can certainly wish
to avoid a future of unrelieved suffering, and in this sense one can have an egoistic
preference for a future that lacks oneself over a future that includes one. And if that is so, I
don’t see why one can’t wish—on egoistic grounds—that one had never come into
existence.
I have not attempted to tackle no-faults genethicism head-on. Instead, I have argued that
no-faults genethicists are committed to an Epicurean view of death. Epicureans, no doubt,
will be nonplussed by this approach, but true Epicureans are few on the ground. Coupled
with its intrinsic counter-intuitiveness, it seems to me that this result gives us good reason
to look elsewhere for an acceptable genethical model. At the very least, this result
provides us with a useful constraint on genethical theory building.
3 Holtug 2001 also makes this point.
8

VI.
Unlike no-faults genethicists, dual benchmark genethicists allow that an individual can be
harmed (or, perhaps, benefited) by being created. Dual benchmark genethicists also accept
that it can be reasonable for a person doomed to a truly miserable life to regret having
been born. But how miserable is too miserable? According to the parity theorist, a life is
too miserable to be created when, and only when, it is the kind of life that someone with it
would be better off dead. There is a single threshold for lives that are worth creating and
lives that are worth continuing.
Dual benchmark accounts reject this view. Here is Benatar’s statement of the dual
benchmark position (which he endorses):
The judgment that a disability is so bad that it makes life not worth continuing is
usually made at a much higher threshold than the judgment that a disability is
sufficiently bad to make life not worth beginning. That is to say, if a life is not worth
continuing, a fortiori it is not worth beginning. It does not follow, however, that if a
life is worth continuing that it is worth beginning or that if it is not worth beginning
that it would not be worth continuing. For instance, while most people think that
living life without a limb does not make life so bad that it is worth ending, most (of
the same) people also think that it is better not to bring into existence somebody who
will lack a limb. We require stronger justification for ending a life than for not
starting one. (Benatar 2000: 176-177; see also Cohen 1997, Peters 1989)
Benatar talks of lives starting and continuing, which is a very natural way to speak. But on
a four-dimensional account of lives it is also problematic: four dimensional objects or
events do not start or continue, they just are. But this problem can be easily dealt with: the
relevant question is what the overall (four-dimensional) shape of a life should look like in
order for it to be permissible to actualize.
But what is that shape? Where is the life-worth-starting benchmark to be set, and what
reasons can the dual benchmark theorist give for setting the benchmark where they do
and not somewhere else?
In general, dual-benchmark theorists have done little to address these questions. The
following quotations are typical of the dual benchmark approach:
Since we ought to try to provide every child with at least a normal opportunity for a
good life, and since we do not harm possible people if we prevent them from
existing, we ought to try to prevent the birth of those with a significant risk of living
worse than normal lives. (Purdy 1995: 302)
9

I assume … that there is a strong moral obligation to prevent preventable harm and
suffering and that this obligation applies equally to curing disease and injury and to
preventing the avoidable creation of people who will have disease or injury. (Harris
2000: 31)
These claims are highly problematic. Purdy appeals to an unanalyzed notion of normality.
Depending on exactly how this notion is cashed out, her comments may well imply that a
significant number of the world’s children should not have been brought into existence.
Harris’s remarks lead in the same direction, for most of us will meet with some form of
disease or injury during the course of our lives. Dual Benchmark accounts are motivated
by the common-sense intuition that it can be wrong to create someone with a severe
handicap, but in their attempt to preserve this intuition they run the danger of flouting
what is arguably a more robust intuition, viz., that the vast majority of children are not
wronged by being brought into existence.
Some dual benchmark theorists are quite happy to reject this intuition. As we will see
(section XII), David Benatar holds that no life that contains some pain manages to meet
the threshold on lives that are worth starting. On his view, most—if not all—individuals
are wrongly brought into existence. Few dual benchmark theorist will want to follow
Benatar’s lead, but those who want to set the life-worth-starting threshold lower will need
to defend their position. Purdy states that the interest we have in being free from disease
or special limitation “is sufficiently compelling in some cases to justify the judgment that
reproducing would be wrong” (Purdy 1995: 307). But when are such interests “sufficiently
compelling”, and what makes them sufficiently compelling in those cases and not others?
The dual benchmark theorist might appeal to some form of species norm in setting the
life-worth-starting benchmark. (This might have been what Purdy had in mind in
suggesting that we ought to try to prevent the birth of those with a significant risk of
living worse than normal lives.) But such an approach seems to me to be rather
implausible. Species norms change over time; certainly average levels of human well
being have increased significantly over the centuries. Relational properties such as “being
above (or below) the norm of well-being for one’s species” might have an indirect bearing
on genethical issues, for one’s level of wellbeing is not unrelated to one’s conception of
how well one is doing relative to certain norms, but it seems implausible to suppose that
they should play a direct role in determining whether or not one has been wronged by
being brought into existence.
It might be argued that the parity theorist (who has a single benchmark account) fares no
better in providing a motivated account of where to draw the life-worth-living
10

enchmark. Not so. She can invoke the sorts of reasons we appeal to in evaluating the
rationality of suicide or euthanasia—roughly, a life is worth living (starting or continuing)
if there is reason to think that its goods will defeat its harms. We might call this the
baseline (or neutrality) intuition. Obviously, fleshing this thought out would require an
account of goods and harms, not to mention an account of how goods might defeat (or
outweigh, balance) harms. My point here is simply that the parity theorist has a principled
account of where to place the life-worth-starting benchmark, even if there is much work to
be done to develop the account. 4
VII.
Even if it is unclear whether the dual benchmark theorist is to place the life-worth-starting
benchmark, there does seem to be something to the thought that there is a gap between
lives that are worth starting and those that are worth continuing. As Benatar points out,
the fact that we judge that some people have been wrongly brought into existence even
though we do not judge that those same people would be better off dead seems to show
that there is a margin between the life-worth-starting benchmark and the life-worthcontinuing
benchmark. Call this the marginal life intuition.
How robust is the marginal life intuition? I know of no firm data that addresses this
question, but I suspect that it is not particularly robust. Some people think that it is wrong
to knowing bring a deaf child into existence; others regard such a decision unproblematic.
Some people share Benatar’s view that it is wrong to knowingly create someone who will
lack a limb; others find such attitudes puzzling. Not only do people differ in their views of
such cases, it is relatively easy to shift people’s responses to marginal life cases by altering
the contextual frame within which the case is presented. (Would the deliberate creation of
a person without a limb be wrong in a world in which everyone lacked a limb?) Perhaps
the marginal life intuition does not run as deep as dual benchmark theorists suggest.
Secondly, there is good reason to suspect that the marginal life intuition is easily confused
with other intuitions—intuitions that are not inconsistent with the parity approach. In
some situations, potential procreators are forced to chose between one of two potential
persons. Now, where one of the two persons (say, A) is thought to be more likely to have
4 Although the parity thesis and the baseline intuition hang together very nicely, the proponent of the parity account is
not committed to endorsing the baseline intuition. A parity theorist could reject the idea that the life-worthstarting/continuing
benchmark should be set at the level of neutral wellbeing; perhaps the only kinds of lives that are
worth starting and continuing are ones in which life’s goods exceed its bads. I’m not myself much drawn to this view,
but perhaps it is defensible.
11

a higher level of well-being than the other, many have the intuition that the potential
procreator ought to create A rather than B (Parfit 1984; see Belshaw 2003 for a contrary
view). Perhaps we confuse the thought that B has a marginal life—that is, a life not worth
starting—with the thought that S should have created A rather than B. But these are two
quite different thoughts, and the latter does not entail the former. Of course, this point
does not provide a full explanation of the marginal life intuition, for the intuition seems to
retain some force where the decision facing a procreator is between having a certain child
and having no child.
Another intuition with which the marginal life intuition might be confused is that
someone might be unfortunate to have been born with a certain condition. The parity
theorist denies that Amy has been wronged by having been brought into existence, but he
need not deny that she is unfortunate to have been born without a limb. It seems to me
that we regard those born without a limb as unfortunate to have been born without a
limb, and not—as Benatar suggests—as unfortunate to have been born at all.
A third response is the most important: third-person marginal-life judgments are not
reflected by first-person marginal-life judgments. Consider the very example that Benatar
uses: it is better not to bring into existence somebody who will lack a limb. Whatever
intuitive support this claim might have, there is no doubt that it has less intuitive support
than the claim that someone who was born without a limb should regret having been
born. (Indeed, I suspect that most of us think that, ceteris paribus, it is not reasonable for
those born without limbs to regret having been born.) Perhaps those born without limbs
often regret having been born without limbs, but I doubt that they tend to regret having
been born. (And if they regret having been born, it’s not on account of the fact that they
were born without a limb.) Faced with a choice between having been born without a limb
and not having been born at all, such individuals would generally choose to have been
born without a limb.
But if cases that generate third-person marginal-life intuitions do not generate first-person
marginal-life intuitions, then we are faced with a difficulty. I argued earlier that firstperson
and third-person genethical judgments should be brought into alignment:
judgments as to whether or not it is reasonable for a person to regret having been born
should be tied to judgments concerning whether or not they have been harmed by being
born. The fact that we don’t have first-person marginal-life intuitions in the missing limb
case—and indeed think that such intuitions would be unreasonable—suggests that there
would be something problematic about third-person marginal-life intuitions in such cases.
12

I have argued that it is not at all clear that we do have third-person marginal-life
intuitions, but even if we do, perhaps we should reject them.
As we have seen, Benatar will argue that we should bring our first-person and thirdperson
genethical intuitions into alignment by revising our first-person intuitions rather
than our third-person intuitions. He holds that those born without limbs should regret
having been born because we should all regret having been born: we are all engaged in
mass deception as to how wonderful things are for us (Benatar 1997: 353; 2006; chapter
three). Perhaps, as Elias Canetti suggested in The Secret Heart of the Clock, we confuse the
craving for life with the endorsement of it. This view deserves serious consideration, but I
don’t have the space to engage with it here. Suffice it to say that I find it uncompelling.
We might, on occasions, be wrong about how well things are going for us, but it seems
highly implausible to me to suppose that we are so fundamentally mistaken as all that.
For what it’s worth, the fact that we do continue to bring children into the world seems to
me to provide some support for the view that we do think that life ought not to be
regretted.
VIII.
Are there any cases that generate a reasonable first-person marginal-life intuition? Are
there any situations in which it is reasonable for a person to think that although their life
(as a whole) was not worth starting it was worth continuing?
Consider Saul Smilansky’s (1997) case of the concentration camp survivor who, reflecting
upon his life near its end, may feel that it was worthwhile. “He might resent anyone
implying otherwise. But when remembering the awful years of the war, his physical and
psychological suffering, the loss of his first wife and child, and all the other relatives and
friends, he may also think that having been spared the suffering, in not having been born,
might have been preferable” (Smilansky 1997: 243).
Let us call this person “Theo”. Theo thinks that not having been born “might have been
preferable”. In representing Theo’s thoughts in this way, I take it that Smilansky’s view is
that Theo thinks that it would have been preferable for him to have not been born; Theo
himself would have been better off not having come into existence. How can we reconcile
this thought with Theo’s conviction that his life has been worthwhile? How could it be
reasonable for Theo to believe that his life (as a whole) has been worthwhile, if he also
thinks that he would have been better off having never been born?
One thing that Theo might mean in describing his life as worthwhile is that he has made
the best of it that he could, given his circumstances. The life he has lived is better than
13

various lives that he might have lived (lives that began the same as his actual life, but
ended quite differently), even though he regards each of these lives as having been worse
(for him) than non-existence. He has made the best of the difficult times through which he
lived, but he would have been better off not having had to live through such times. This
interpretation of his remarks is clearly consistent with the parity account, but it is
probably not what Theo (Smilansky) means by the phrase.
A second interpretation of what Theo means is that although the various parts of his life
have been good for Theo, his life as a whole has not been. In fact, Bigelow et al use the
example of a concentration camp survivor to make the point that the value of one’s life as
a whole need not be determined in any straightforward way by the value of its parts: “A
person who has had an unspeakably horrible stretch in his life (such as three years in a
concentration camp) might be considered to have led an unfortunate life however good
the rest of his life may have been” (1990: 123). Perhaps a period of utter horror and
brutality can ensure that a person would have been better off not having been born, no
matter how much fortune they meet with in the rest of their life. But if this is how Theo
thinks of his life, it is puzzling why he would describe it as having been worthwhile.
Again, we seem to be unable to reconcile Theo’s regret at having been born with the fact
that he regards his life as a whole as having been worthwhile. He may, of course, endorse
a particular time-slice of his life, such as that which he now enjoys, but to do this is not to
endorse one’s life as a whole. 5
Where does this leave the argument from marginal life cases? I have suggested that on
reflection the marginal life gap is a lot more tenuous and narrower than dual-benchmark
theorists such as Benatar suggest. It is, of course, true that we require stronger justification
for ending another person’s life than for not starting it, but because this claim involves
complex considerations concerning the ethics of killing it should not be confused with the
marginal life intuition. It is one thing to claim that a person’s life is not worth living, it is
quite another to claim that it would be permissible to kill them.
5 Kamm points out that death is an insult to an existing entity—it is a manifestation of the person’s vulnerability. Total
non-existence, on the other hand, is not an insult, for it occurs to no actual entity (Kamm 1993: 42, 64). This, she
suggests, might make it comprehensible why someone might prefer never having come into existence to even a good
mortal life. One might regard the insult of death—independently of what it deprives one of—as an intrinsic bad that is
best avoided if possible. The attitude that Kamm has outlined does not seem to be what Smilansky has in mind, but that
does not detract from its interest. I don’t myself think that the insult factor of death should be given much weight in this
context, but perhaps it should not be dismissed.
14

IX.
I turn now to more general arguments in favour of the dual benchmark view. Unlike the
argument from marginal lives, these considerations support extreme versions of the dual
benchmark position, according to which procreation is generally—if not universally—
morally problematic.
I begin with Shiffrin’s (1999) work. Shiffrin argues that procreation is not a morally
straightforward activity, but one that faces “difficult justificatory hurdles.” In theory, she
holds, all children have causes of action for wrongful life suits. Shiffrin’s argument turns
on the claim that it is impermissible to harm someone without their consent, even when
the action that causes the harm brings about benefits that can be expected to outweigh the
harms it causes. Since even the best of lives involve serious burdens, harms and risks; and
since one cannot secure a person’s consent before bringing them into existence; it follows
that procreation is almost always morally problematic, and its imposer may be justifiably
held responsible for its harmful results. In every case, Shiffrin claims, voluntary
procreation “involves a person imposing a risk upon another where the imposition is not
necessitated by the need to avert greater harm” (1999: 139).
Shiffrin’s argument is primarily addressed to prospective genethics; it is less clear what
her view of retrospective genethics is (or should be). Shiffrin does not argue that most of
us should regret having been born—indeed, the thrust of her argument is that one can be
wronged by an event even when one has no cause to regret it. One might think that this in
itself is problematic: as I have argued, there is something rather puzzling in the thought
that bringing S into existence could be deeply morally problematic although S has no
reason to regret having been brought into existence. But I’ll leave this point aside here;
instead, I want to focus on Shiffrin’s rejection of parity between third-person starting-life
and ending-life decisions. 6
Suppose that you are deliberating about whether or not to preserve the life of a patient
who is temporarily comatose. The patient is suffering a life-threatening illness, but if
treated can be expected to make a full recovery. It seems clear that in such a case it is not
only permissible to save the patient’s life but obligatory. Prima facie, however, Shiffrin’s
position would seem to entail that we ought to let the patient die. After all, saving her life
exposes her to serious future harms and burdens that she would otherwise avoid, and
such harms are unconsented. Call this the comatose patient case.
6 Shiffrin takes no position on how first-person genethical attitudes ought to relate to third-person genethical attitudes,
or indeed, on how first-person genethical attitudes ought to relate to first-person mortethical attitudes.
15

Shiffrin might respond by arguing that although saving the patient’s life exposes them to
future harms (sickness, suffering, loneliness, etc), it also saves them from a greater harm,
namely death. But on what grounds can Shiffrin argue that death is a harm? (Indeed, on
what grounds can she argue that killing someone without their consent is wrong, given
that it prevents unconsented harms?) Shiffrin does not provide a full answer to this
question, but she suggests that death is bad because it interferes with the exercise of
agency. “By constraining the duration and possible contents of the person’s life, [death]
forces a particular end to the person – making her with respect to that significant aspect of
her life merely passive” (1999: 124)
At best, this account of death’s badness might account for our intuitions concerning the
deaths of autonomous, self-conscious individuals. It is harder to see how it might account
for the badness of the death of infants, the severely mentally retarded, the senile, and
others who lack full agency over their own lives. Yet the fact that comatose patient might
happen to be a young child seems not to change our intuitions concerning what we ought
to do. Most of us have the intuition that we should save the comatose patient’s life, even if
she is only a week old. And in saving this infant’s life we impose (or at least allow)
roughly the same amount of unconsented harm that was imposed on the child by creating
her.
Of course, Shiffrin might simply bite the bullet and argue that death harms an individual
only to the extent that it impairs her agency over her life. The deaths of those individuals
who lack significant agency over their lives, such as young children and the intellectually
disabled, might be a tragedy for someone, but not for them. I think this would be a tough
line to adopt, and I won’t pursue it any further.
Another response that Shiffrin might have to the comatose patient objection would be to
supplement her account of the wrongness of death by appeal to the notion of desert in
order to capture the badness of infant death. Perhaps the death of an infant (and perhaps
even a fetus) is bad for them because it deprives them of the goods of life that they
deserve. The infant has made an investment in its life, and depriving it of a return on its
investment would be to harm it. (By contrast, potential individuals have made no
investments in their life, and thus are not owed anything.) There is much that is
problematic about the idea that desert might ground the badness of infant death, but I
won’t pursue that issue here (see McMahan 2002: 168). The central point to note here is
that someone sympathetic to Shiffrin’s approach will demand that desert-based reasons
for thinking that an infant is harmed by death must be weighed against the fact that
saving the infant’s life will expose them to the unconsented harms and burdens of life –
16

that infants must endure “the fairly substantial amount of pain, suffering, difficulty,
significant disappointment, distress, and significant loss that occur within the typical life”
(1999: 137). Surely these features of life are not deserved.
But even if Shiffrin can account for why death is a harm she still needs to explain why it is
permissible to save the comatose patient. On her view, we would need reasons to think
that allowing death (at this time) would be to expose the patient to greater harms than
those to which she would be exposed were we to save her life. And it’s not clear that we
will always—or even normally—have reason to think that this is so. It seems entirely
possible that the (unconsented) harms we expose her to by saving her life might outweigh
the harm of death, especially given that death is never avoided but only postponed.
Suppose that the patient is now 50, and saving them will give them another 30 or so years
of life. If Shiffrin is to argue that saving the patient is obligatory she must argue that death
at 50 is worse than death at 80, and worse in such a way as to balance the fact that saving
the patient’s life has involved 30 years of unconsented harms. There are a number of
accounts of why, all other things being equal, death at 50 is worse than death at 80, but
each of these accounts turns on the claim that it is a good thing to extend life, even when
the subject in question faces significant unconsented harm. But if such accounts are
acceptable here it is unclear why a similar line of reasoning cannot be used to justify
procreation: it is a good thing to create life, even when doing so involves unconsented
harms.
I suspect that the most plausible response that Shiffrin might make to the comatose
patient objection is to invoke the notion of substituted judgment. Perhaps it is permissible
to save the comatose patients life because we think that were we in the comatose patient’s
position we too would desire that our life be saved. But if it is permissible to employ
substituted judgments here then there seems to be nothing to prevent them from being
employed in genethical contexts. Would-be parents can justify their decision to have a
child on the grounds that were they (or some suitable idealization of them) in the child’s
position, they would consent to having been brought into existence.
X.
As we have noted, David Benatar agues that being brought into existence is not a benefit
but always a harm, and that we each should regret having come into existence (Benatar
1997, 2006). Benatar’s argument for these claims turns on a putative asymmetry between
pains and pleasures, or harms and benefits more generally. Benatar claims that whereas
the absence of pain is itself good, the absence of pleasure is not bad (unless there is
someone for whom this absence is a deprivation).
17

World A
X exists
1) Presence of Pain
(Bad)
2) Presence of Pleasure
(Good)
World B
X never exists
3) Absence of Pain
(Good)
4) Absence of Pleasure
(Not–Bad)
In order to determine the relative advantages and disadvantages of coming into existence
and never coming to be, Benatar says, we need to compare 1) with 3) and 2) with 4). When
we do this, Benatar says, we discover that 3) is better than 1), but 2) is not better than 4):
“the pleasures of existence, although good, are not a real advantage over non-existence,
because the absence of pleasures is not bad” (1997: 348). All things considered, nonexistence
is preferable to existence because there is nothing bad about never coming into
existence but there is (invariably) something bad about coming into existence. 7 Let us call
the asymmetry between 3) and 4) Benatar’s asymmetry.
Benatar’s matrix has several features worth noting. Firstly, there is something intuitively
odd about describing worlds in which S does not exist as ‘good for S’. It is difficult to see
how anything could be good (or bad, for that matter) for someone in worlds in which they
don’t exist. Further, even if worlds in which S doesn’t exist are good for S, it doesn’t
follow that such worlds will be better for S than worlds in which S does exist. Presumably
pleasures (or, more generally, goods) aggregate in some way—generally speaking, a life
with more goods is better than one with fewer goods. But if this is so, then we need to
know how much pain and pleasure S has in A in order to know whether or not A is worse
than for S than B is; in particular, we need to know how much pain is absent from B.
But perhaps the most pressing question concerns Benatar’s asymmetry itself. Is it true?
Benatar defends it on the grounds that it provides the best explanation of four genethical
judgments (2006: 32ff.; see also 2006: 203). Firstly, while there is a duty to avoid bringing
suffering people into existence, there is no duty to bring happy people into being (2006:
7 Benatar view is closely akin to “frustrationism” (see Fehige 1998; Holtug 2001). Frustrationists claim that the value
of satisfying a preference equals that of not having it; that is, it has zero value. By contrast, the frustration of a
preference has negative value. On this conception of value, the best that one could possibly hope to get out of life
would be to break even—which is the value equivalent to not having been born. But since even the best of lives
involves some pain and frustration, one should expect that one will come out of life in the red; that is, with negative
value. So one is almost certainly better off not having been born. Benatar differs from the frustrationaist in taking the
presence of pleasure to be good, whereas the frustrationist views it as being only neutral.
18

32). Secondly, “whereas it is strange (if not incoherent) to give as a reason for having a
child that the child one has will thereby be benefited, it is not strange to cite a potential
child’s interests as a basis for avoiding bringing a child into existence” (2006: 34). Thirdly,
bringing people into existence as well as failing to bring people into existence can be
regretted, but only bringing people into existence can be regretted for the sake of the
person whose existence was contingent on our decision (34). Fourthly, whereas we are
rightly sad for inhabitants of a foreign land whose lives are characterized by suffering,
when we hear that some island is unpopulated we are not similarly sad for the happy
people who, had they existed, would have populated this island (35).
Even if Benatar’s asymmetry does provide the best explanation of these four judgments, it
is far from clear that that would that give us sufficient reason to endorse his position. The
problem, obviously enough, is that we are also strongly committed to the claim that
individuals with normal human lives—not to mention individuals with lives that are
vastly superior to normal human lives—are not harmed by being brought into existence.
No doubt we would be in something of a genethical pickle were it to turn out that these
intuitions are not reconcilable with each other, but why attempt to extricate oneself from
the pickle by embracing the thought that almost everyone is wronged by being brought
into existence? One could reject the four genethical intuitions that Benatar appeals to or,
perhaps, give up on genethical intuition altogether. Of course, each of these moves comes
at some cost, but it is not at all clear to me that this cost is higher than that which Benatar’s
solution incurs. Of course, Benatar himself seems to think that his view is not counterintuitive—“there
is nothing implausible either in the view that coming into existence is
always a harm or in the view that we ought not to have children” (2006: 207)—but it is
clear that Benatar is in a very small minority here, and the vast majority of people do find
these views strongly counter-intuitive. Evaluating the weight of intuition is a tricky
business even in well-understood domains—and this domain most definitely does not
meet that description!—but it seems to me that even in combination the weight of the four
intuitions that Benatar invokes does not equate to that which he is forced to reject. 8
8 At various points in Better Never to Have Been Benatar seeks to downplay the force of intuition. “Intuitions are often
profoundly unreliable, the product of mere prejudice. Views that are taken to be deeply counter-intuitive in one time
and place are often taken to be obviously true in another.” (2006: 203). No doubt there is much truth in all of this, but
there is no reason here to think that genethical intuitions have no evidential force at all. Indeed, Benatar himself appears
to put some stock in genethical intuition, as when (for example) he claims that the judgment that a disability is so bad
that it makes life not worth continuing should be made at a much higher threshold than the judgment that a disability is
sufficiently bad to make life not worth beginning.
19

Let us return to Benatar’s asymmetry: does it provide the best explanation of the four
judgments that he presents? Well, that depends on what precisely the judgments in
question are.
Is it the case that there is a duty to avoid bringing suffering people into existence, but no
duty to bring happy people into being? Well, that depends on just what one means by
‘suffering people’. Almost no one thinks that we have a duty to avoid creating people who
will experience some suffering. If by ‘suffering people’ Benatar means people who
experience only suffering, then he is clearly right: we do have a strong pre-theoretical
commitment to the view that there is a duty to avoid bringing such people into existence.
And, equally, we have a strong (albeit, perhaps, less strong) commitment to the view that
no duty to bring happy people into being. Does Benatar’s asymmetry provide the best
explanation of this judgment? I don’t think so. An even better explanation would appeal
to an asymmetry between good and bad lives rather than to an asymmetry between good
and bad events. The best explanation of our intuition is simply that we think it is good to
avoid a miserable life, but not bad to miss out on a good life. This explanation accounts for
our intuitions, and it does so without the counter-intuitive consequences of Benatar’s
view.
The second judgment that Benatar invokes to support his account is that “whereas it is
strange (if not incoherent) to give as a reason for having a child that the child one has will
thereby be benefited, it is not strange to cite a potential child’s interests as a basis for
avoiding bringing a child into existence” (2006: 34). Again, we do endorse this judgment,
but only up to a point. It doesn’t seem strange to cite a potential child’s overall interests or
well-being as a basis for avoiding bringing it into existence; in particular, it doesn’t seem
strange to think that if the potential child’s (expected) pain would be such as to
overwhelm its (expected) pleasures, then one should not bring it into existence. But it does
seem strange to cite a potential child’s expected pains as a basis for avoiding bringing it
into existence without at the same time being prepared to cite its expected pleasures as a
basis for bringing it into existence. Here too our commonsensical genethical judgments
seem to be grounded in the expected quality of the target individual’s life, not in some
disembodied calculus involving pleasures and pains.
Benatar’s third data point is the judgment that “bringing people into existence as well as
failing to bring people into existence can be regretted, but only bringing people into
existence can be regretted for the sake of the person whose existence was contingent on
our decision” (2006: 34). Again, this judgment does seem to be one that we share, but in
order to explain it we need invoke only an asymmetry between good and bad lives.
Whereas we regret having brought bad lives into existence, we do not regret not having
20

ought good lives into existence. So there is any asymmetry here, but it is not one that
lends any support to Benatar’s asymmetry. The kinds of lives that we regret having
brought into existence are not those of normal human beings, but those in which the good
of life outweighed its bads.
What about the fourth judgment? Here too the objects of our evaluation appear to be lives.
It is true that we are rightly sad for inhabitants of a foreign land whose lives are
characterized by suffering. But it is also true that we are happy for inhabitations of a
foreign land whose lives are characterized by happiness. More particularly, we are sad for
inhabitants of foreign lands whose lives are so characterized by suffering that we think
that they would be better off dead, and we are happy—or at least not sad—for inhabitants
of foreign lands whose lives are such that we do not think that they would be better off
dead.
The moral of the foregoing is a simple one: there is an asymmetry in our genethical
judgments, but we do violence to that asymmetry if we attempt to explain it in terms of
pains and pleasures (or bads and goods) in isolation. The best systematization of these
four judgments is that there is a deep asymmetry between good lives and bad lives. In
response, Benatar might object that we haven’t really explained these four genethical
judgments by invoking an asymmetry between good and bad lives; we have simply
redescribed them. In order to explain my asymmetry (between lives), we need to invoke a
further asymmetry—his asymmetry—between events.
But it seems to me that taking this extra step adds nothing to what we already have, and
incurs costs of its own. It adds nothing, for Benatar provides no explanation of his
asymmetry. (He has no story to tell as to how it could be a good thing for S to avoid pains
by failing to exist without it also being a bad thing for S to avoid pleasures by failing to
exist.) And it incurs additional costs, for it is obviously at odds with our judgment that we
do not harm normal people by bringing them into existence. It is implausible to suppose
that our judgments about the asymmetry between good lives and bad lives could be
explained by appeal to Benatar’s asymmetry, given that Benatar’s asymmetry leads so
directly to conclusions that are so radically at odds with common-sense. Better, I think, to
treat the asymmetry between good lives and bad lives as brute rather than attempt to
explain it by appeal to intuitions that we quite clearly do not have.
XI.
Although Benatar urges us not to procreate, he does not urge us to commit suicide or to
refrain from offering life-saving treatment to those in need of it. I find this asymmetry
21

puzzling. If one could have benefited by not having being born on account of the pain that
one avoids, so too one can benefit from an early death on account of the pain that one
avoids. But if Benatar accepts this point, then he should accept a parallel point about the
evils of death and non-conception: since death can harm a person by depriving them of the
benefits of life, so too non-existence can harm the non-existent by depriving them of the
goods of life.
In response, Benatar claims that his point is not that “it is better never to exist because one
thereby avoids pains, but that it is better never to exist because one avoids pains without
cost. The same is not true of death. Whereas death does spare one of all future pains, it
does involve a cost” (personal communication). There is something to this, but it is also
somewhat mysterious. Death does exact a price, for it robs one of the goods that one
would have enjoyed had one continued to exist (Feldman 1992; McMahan 2002). But there
is also a sense in which non-existence comes at a cost—namely, the goods that one would
have enjoyed had one come into existence. Of course, strictly speaking, this cost is not
incurred by anyone, for only the existent can be robbed. But, if we are speaking strictly,
then we must also note that it is equally true that only the existent can benefit from having
dodged disaster. Insisting on strict speech threatens to undercut the very asymmetry that
Benatar is at pains to defend.
Perhaps these points are most fully appreciated by considering death from the first-person
perspective. Let us suppose that you are contemplating suicide, and that other parties will
not be greatly affected by your decision. In deliberating about whether or not to commit
suicide, it may seem reasonable to adopt a genethical stance towards the remainder of
your life. Call the subject of this life-segment “future-you”. You could bring future-you
into being by deciding not to commit suicide, or you could make it the case that futureyou
never exists by committing suicide. How should you decide? From a certain
perspective, it seems reasonable to decide on the basis of what quality of life twin-you can
be expected to enjoy: if it is good, then deciding in favour of continued existence seems
reasonable, if it is bad, then deciding in favour of suicide seems reasonable. 9 Arguably,
one doesn’t have any obligations to future-you to bring him or her into existence, although
one might have duties not to bring certain types of future selves into existence. Thinking
of oneself as faced with decisions about whether or not to allow (or bring) a future self
into existence brings to the surface deep points of contact between coming into existence
and staying in existence. Our lives are not given once and for all. The question of whether
9 A slightly different way of addressing the question asks what kind of life you as a temporally-extended whole would
enjoy were twin-you to come into existence.
22

to stay in existence always contains within itself a question about whether to bring ‘the
rest of oneself’ into existence.
To conclude: in thinking about whether or not to remain in existence, it is reasonable to
adopt a form of substituted judgment—one adopts the perspective of one’s future self. But
if this is reasonable, why is it not also reasonable to adopt such a perspective with respect
to coming into existence? If, in adopting the perspective of the person who would be
brought into existence as the result of the relevant deliberation, one decides that the
expected goods outweigh the expected bads, then procreation is permissible; if not, then it
is not permissible.
XII.
I turn finally, and all too briefly, to the parity approach. The basic idea behind the
approach is encapsulated in Bernard Williams’s claim that “what [resenting one’s
existence] requires is that the person should prefer not to have existed, and I take it that
this implies thinking that his or her life is not worth living” (Williams 1995: 227).
“Implies” might be too strong a term here, but it does seem reasonable to suppose that
there should be internal constraints between one’s attitude to whether or not one’s life
was worth starting and one’s attitude to whether or not it is worth continuing. This
connection seems to extend not just to one’s own life, but to the living of lives in general: if
one thinks that life could never be so bad as to be not worth continuing, then one should
also think that life could never be so bad as to be not worth starting, and vice-versa. We
can employ genethical intuitions to constrain mortethics, and mortethical intuitions to
constrain genethics.
What grounds the plausibility of the parity thesis is the thought that the value of a life (to
the subject of that life) is fixed by its intrinsic quality—on what it is like to live that life. If
living a life is of no value to the subject of that life, then it is pointless to begin that life for
their sake. But if the living of a life is (or at least has the potential to be) of some (global)
benefit to the subject of that life, then there is some point to starting it—it is a life worth
actualizing.
It is important to bear in mind that by “lives” here I mean entire lives, not temporal
segments thereof. Overlooking this point can lead one astray. For example, one might
worry about the following kind of case. Suppose that a person’s life contains some period
of grotesque suffering. That means that it would have been for him never to have existed
and it would have been better before or during that period for him to die. But once the
low point of his life has passed, it is now the case that it is no longer in his interests to die.
23

So his life is worth continuing (at this point), but it was not worth continuing before and it
was not worth starting. So isn’t this an objection to parity?
But there is no objection here. Parity doesn’t say that if the value of a life as a whole was
not worth starting then there is no point in that life at which it would be worth
continuing. (In a sense, what one “continues” is not one’s life in the relevant sense—lives
in the relevant sense just are.) Parity says only that the kinds of conditions that make lives
not worth creating are just the sorts of conditions that make lives not worth continuing.
Because lives are temporally extended, it is obviously possible for a life to be
characterized by such a condition during some of its temporal segments but not others.
Of course, parity has its limits. As many have noted, commonsense intuition resists the
thought that we have obligation to start even very good lives. Indeed, commonsense
resists the thought that it is a good thing to start good lives. There are circumstances in
which creating people is laudable, but creating people can never be laudable because of
the benefits that it bestows on the person who is created. And here, of course, there is a
deep (and puzzling) asymmetry between genethics and mortethics. Although creating
(good) lives is not per se a good thing to do, extending (good) lives is a good thing to do.
Parity has its limits. Actual people have claims on us that merely potential people do not.
XIII.
Courts have had a number of objections to wrongful life cases. One objection is that such
cases assume that the relative merits of existence and non-existence can be compared, and
that this assumption is problematic. This objection cannot be sustained. Another objection
is that a positive verdict in such cases would commit the court to the position that the
plaintiff would be better off dead. This objection has more merit. This objection isn’t quite
right, but if the parity model is right then something very much like it is. The reason the
conclusion does not strictly follow is that the current level of wellbeing that the plaintiff
currently enjoys might differ in important respects from the quality of their life as a
whole. Perhaps the quality of the person’s life has improved and they currently have a life
that is worth living, even though they shouldn’t have been brought into existence on
account of the torment that they have already endured. But the thought behind the
objection is, I think, basically sound: the kinds of conditions that prevent life from being
worth starting are just those that prevent it from being worth continuing.
I have not attempted to provide a full account of genethics here. Instead, I have attempted
to argue for a certain constraint on such account, viz., that there should be parity of a
certain kind between our attitudes towards coming into existence and going out of
existence. The parity thesis does not determine a particular genethical position, but it does
24

indicate that a number of recent genethical approaches are flawed. And that is not an
insignificant result.
References
Belshaw, C. 2003. More Lives, Better Lives, Ethical Theory and Moral Practice, 6: 127-141.
Benatar, D. 1997. Why it is Better Never to Come into Existence, American Philosophical
Quarterly 34/3: 345-55.
Benatar, D. 2000. The Wrong of Wrongful Life, American Philosophical Quarterly, 37/2: 175-
83.
Benatar, D. 2006. Better Never to Have Been: The Harm of Coming Into Existence. Oxford:
Clarendon Press.
Bigelow, J. Campbell, J., & Pargetter, R., 1990. Death and Well-Being, Pacific Philosophical
Quarterly 71: 119-40.
Cohen, C.B. 1997. The Morality of Knowingly Conceiving Children with Serious
Conditions: An Expanded “Wrongful Life” Standard, in N. Fotion and J. Heller
(eds) Contingent Future Persons, Dordrecht: Kluwer, pp. 27-40.
Feldman, F. 1992. Confrontations with the Reaper, New York: OUP.
J. Harris, 2000. Rights and Reproductive Choice, in J. Harris and S. Holm The Future of
Human Reproduction, Oxford: Clarendon Press.
Heyd, D. 1992. Genethics, Berkeley: University of California Press.
Holtug, N. 2001. On the Value of Coming into Existence, The Journal of Ethics 5: 361-84.
Kamm, F.M. 1993. Morality, Mortality, Oxford: Oxford University Press.
Kavka, G. 1982. The paradox of future individuals, Philosophy and Public Affairs, 11/2: 159-
85.
McMahan, J. 2002. The Ethics of Killing, Oxford: Oxford University Press.
Parfit, D. 1984. Reasons and Persons. Clarendon.
Peters, P.G. Jr., 1989. Protecting the Unconceived: Nonexistence, Avoidability, and
Reproductive Technology, Arizona Law Review 31, pp. 487-548.
25

Purdy, L. 1995. Loving Future People, In J. Callahan (ed) Reproduction, Ethics, and the Law,
Bloomington and Indianapolis: Indiana University Press, pp. 300-27.
Shiffrin, S.V. 1999. Wrongful life, Procreative Responsibility, and the Significance of
Harm, Legal Theory 5: 117-148.
Smilansky, S. 1997. Preferring Not to Have Been Born, Australasian Journal of Philosophy 75:
241-47.
Williams, B. 1973. The Makropulos Case: Reflections on the Tedium of Immortality, in his
Problems of the Self, Cambridge: Cambridge University Press.
Williams, B. 1995. Resenting One’s Own Existence, in his Making Sense of Humanity,
Cambridge: Cambridge University Press.
26