Contemporary philosophical literature contains two kinds of arguments concerning the morality of abortion. One
family of arguments (see the following three sections) relates to the moral status of the fetus—the question of whether
the fetus has a right to life, is the sort of being it would be seriously wrong to kill, or in other words is a 'person' in
the moral sense. An affirmative answer would support claim (1) in the central pro-life argument, while a negative answer would
support claim (2) in the central pro-choice argument.
Another family of arguments (see the section on Thomson, below) relates to bodily rights—the question of
whether the woman’s bodily rights justify abortion even if the fetus has a right to life. A negative answer would
support claim (2) in the central pro-life argument, while an affirmative answer would support claim (2) in the central pro-choice
Arguments based on criteria for personhood
Mary Anne Warren, in her famous article arguing for the permissibility of abortion, holds that moral opposition to abortion is based on the following argument:
It is wrong to kill innocent human beings.
The fetus is an innocent human being.
Hence it is wrong to kill the fetus.
Warren, however, thinks that 'human being' is used in different senses in (1) and (2). In (1), 'human being'
is used in a moral sense to mean a 'person', a 'full-fledged member of the moral community'. In (2), 'human being' means 'biological
human'. That the fetus is a biologically human organism or animal is uncontroversial, Warren
holds. But it does not follow that the fetus is a person, and it is persons that have rights, such as the right to life.
To help make a distinction between 'person' and 'biological human', Warren notes that we should respect the
lives of highly intelligent aliens, even if they are not biological humans. She thinks there is a cluster of properties
that characterize persons:
consciousness (of objects and events external and/or internal to the being), and in particular the
capacity to feel pain
reasoning (the developed capacity to solve new and relatively complex problems)
self-motivated activity (activity which is relatively independent of either genetic or direct external control)
the capacity to communicate, by whatever means, messages of an indefinite variety of types, that is, not just
with an indefinite number of possible contents, but on indefinitely many possible topics
the presence of self-concepts, and self-awareness, either individual or racial, or both
A person does not have to have each of these, but if something has all five then it definitely is a person whether
it is biologically human or not, while if it has none or perhaps only one then it is not a person, again whether it is biologically
human or not. The fetus has at most one, consciousness (and this only after it becomes susceptible to pain—the timing of which is disputed), and hence is not a person.
Other writers apply similar criteria, concluding that the fetus lacks a right to life because it lacks self-consciousness, or rationality and self-consciousness, or 'certain higher psychological capacities' including 'autonomy'. These writers disagree on precisely which features confer a right to life, but agree those features must be certain developed psychological features which the fetus
Arguments of this sort face two main objections. The comatose patient objection claims that as patients
in a reversible coma do not satisfy Warren's (or other) criteria—they are not conscious, do not communicate,
and so on—therefore they would lack a right to life on her view. One response is that 'although the reversibly comatose lack any conscious mental states,
they do retain all their unconscious [or dispositional] mental states, since the appropriate neurological configurations
are preserved in the brain.' This may allow them to satisfy some of Warren’s criteria.
The infanticide objection points out that infants (indeed up to about one year of age, since it is only
around then that they begin to outstrip the abilities of non-human animals) have only one of Warren’s characteristics—consciousness—and
hence would have to be accounted non-persons on her view; thus her view would permit not only abortion but infanticide. Warren agrees that infants are non-persons (and so killing them is not strictly murder),
but denies that infanticide is generally permissible. For, Warren claims, once a human being is born, there is no longer a conflict between it and the
woman's rights, since the human being can be given up for adoption. Killing such a human being would be wrong, not because it is a person, but because
it would go against the desires of people willing to adopt the infant and to pay to keep the infant alive.
Nonetheless, Warren grants that her argument entails that infanticide would be morally acceptable under some
circumstances, such as those of a desert island. Philosopher Peter Singer similarly concludes that infanticide, particularly of severely disabled infants, is justifiable under certain conditions. And Jeff McMahan grants that under very limited circumstances it may be permissible to kill one
infant to save the lives of several others. Opponents may see these concessions as a reductio ad absurdum of these writers' views; while supporters may see them merely as examples of unpleasant acts being
justified in unusual cases.
The natural capacities view
Some opponents of Warren’s view believe that what matters morally is not that one be actually exhibiting
complex mental qualities of the sort she identifies, but rather that one have in oneself a self-directed genetic propensity
or natural capacity to develop such qualities. In other words, what is crucial is that one be the kind of entity
or substance that, under the right conditions, actively develops itself to the point of exhibiting Warren's
qualities at some point in its life, even if it does not actually exhibit them because of not having developed them
yet (fetus, infant) or having lost them (severe Alzheimer's). Because human beings do have this natural capacity—and indeed have it essentially—therefore (on this view) they essentially have a right to life: they could not possibly
fail to have a right to life. Further, since modern embryology shows (it is said) that the fetus begins to exist at conception and has a natural
capacity for complex mental qualities, therefore the right to life begins at conception.
Grounding the right to life in essential natural capacities rather than accidental developed capacities
is said to have several advantages. As developed capacities are on a continuum, admitting of greater and lesser degrees—some,
for example, are more rational and self-conscious than others—therefore: (1) the 'developed capacities’ view must
arbitrarily select some particular degree of development as the cut-off point for the right to life—whereas the
'natural capacities' view is non-arbitrary; (2) those whose capacities are more developed would have more of
a right to life on the 'developed capacities' view—whereas the 'natural capacities' view entails we all have an equal
right to life; and (3) the continuum of developed capacities makes the exact point at which personhood ensues vague, and human
beings around that point, say between one and two years of age, will have a shadowy or indeterminate moral status—whereas there is no such indeterminacy on the 'natural capacities'
Some defenders of Warren-style arguments grant that these problems have not yet been fully solved, but reply that the 'natural capacities' view fares no better. It is argued, for example, that as
human beings vary significantly in their natural cognitive capacities (some are naturally more intelligent than others), and as one can imagine a series or spectrum of species with
gradually diminishing natural capacities (for example, a series from humans down to amoebae with only the slightest differences in natural capacities between each successive
species), therefore the problems of arbitrariness and inequality will apply equally to the 'natural capacities' view. In other words, there is a continuum not only of developed but of natural capacities, and
so the 'natural capacities' view will inevitably face these problems as well.
These problems aside, some critics reject the 'natural capacities' view on the basis that it takes mere species
membership or genetic potential as a basis for respect (in essence a charge of speciesism), or because it entails that anencephalic infants and the irreversibly comatose have a full right to life. Moreover, as with Marquis’s argument (see below), some theories of personal identity would support the view that the fetus will never itself develop complex mental
qualities (rather, it will simply give rise to a distinct substance or entity that will have these qualities), in which
case the 'natural capacities' argument would fail.
The deprivation argument
A seminal essay by Don Marquis argues that abortion is wrong because it deprives the fetus of a valuable future. Marquis begins by arguing that what makes it wrong to kill a normal adult human being is the fact
that the killing inflicts a terrible harm on the victim. The harm consists in the fact that ‘when I die, I am deprived
of all of the value of my future’: I am deprived of all the valuable ‘experiences, activities, projects, and enjoyments’
that I would otherwise have had. Thus, if a being has a highly valuable future ahead of it—a ‘future like ours’—then
killing that being would be seriously harmful and hence seriously wrong. But then, as a standard fetus does have a highly valuable future, killing a standard fetus
is seriously wrong. And so ‘the overwhelming majority of deliberate abortions are seriously immoral’—‘in
the same moral category as killing an innocent adult human being’.
A consequence of this argument is that abortion is wrong in all the cases where killing a child or adult with
the same sort of future as the fetus would be wrong. So for example, if involuntary euthanasia of patients with a future filled with intense physical pain is morally acceptable, aborting
fetuses whose future is filled with intense physical pain will also be morally acceptable. But it would not do, for example,
to invoke the fact that some fetus's future would involve such things as being raised by an unloving family, since we do not
take it to be acceptable to kill a five-year-old just because her future involves being raised by an unloving family. Similarly,
killing a child or adult may be permissible in exceptional circumstances such as self-defense or (perhaps) capital punishment; but these are irrelevant to standard abortions.
Marquis’s argument has prompted several objections. The contraception objection claims that if
Marquis’s argument is correct, then, since sperm and ova (or perhaps a sperm and ovum jointly) have a future like ours,
contraception would be as wrong as murder; but as this conclusion is (it is said) absurd—even
those who believe contraception is wrong do not believe it is as wrong as murder—the argument must be unsound. One response is that neither the sperm, nor the egg, nor any particular sperm-egg combination, will ever itself
live out a valuable future: what will later have valuable experiences, activities, projects, and enjoyments is a new entity,
a new organism, that will come into existence at or near conception; and it is this entity, not the sperm or
egg or any sperm-egg combination, that has a future like ours.
As this response makes clear, Marquis's argument requires that what will later have valuable experiences and
activities is the same entity, the same biological organism, as the fetus. The identity objection rejects this assumption. On certain theories of personal identity (generally motivated by thought experiments involving brain or cerebrum transplants), each of us is not a biological organism but rather an embodied mind
or a person (in John Locke’s sense) that comes into existence when the brain gives rise to certain developed
psychological capacities. If either of these views is correct, Marquis’s argument will fail; for the fetus (at least
the early fetus, lacking the relevant psychological capacities) would not itself have a future of value, but would
merely have the potential to give rise to a different entity, an embodied mind or a person, that would have
a future of value. The success of Marquis’s argument thus depends on one’s favored account of personal identity.
The interests objection claims that what makes murder wrong is not just the deprivation of a valuable
future, but the deprivation of a future that one has an interest in. The fetus has no conscious interest in its future, and
so (the objection concludes) to kill it is not wrong. The defender of Marquis-style arguments may, however, give the counterexample
of the suicidal teenager who takes no interest in his or her future, but killing whom is nonetheless wrong and murder. If the opponent responds that one can have an interest in one's future without taking
an interest in it, then the defender of the Marquis-style argument can claim that this applies to the fetus. Similarly, if an opponent claims that what is crucial is having a valuable future which one would,
under ideal conditions, desire to preserve (whether or not one does in fact desire to preserve it), then the defender may ask why the fetus would not, under ideal conditions, desire to preserve its
The equality objection claims that Marquis’s argument leads to unacceptable inequalities. If, as Marquis claims, killing is wrong because it deprives the victim of a valuable future, then,
since some futures appear to contain much more value than others—a 9 year old has a much longer future than a 90 year
old, a middle class person’s future has much less gratuitous pain and suffering than someone in extreme
poverty—some killings would turn out to be much more wrong than others. But as
this is strongly counterintuitive (most people believe all killings are equally wrong, other things being equal), Marquis’s
argument must be mistaken. Some writers have concluded that the wrongness of killing arises not from the harm it causes the
victim (since this varies greatly among killings), but from the killing’s violation of the intrinsic worth or personhood
of the victim. However, such accounts may themselves face problems of equality, and so the equality objection may not be decisive against Marquis's argument.
Finally, the psychological connectedness objection claims that a being can be seriously harmed by being
deprived of a valuable future only if there are sufficient psychological connections—sufficient correlations or continuations
of memory, belief, desire and the like—between the being as it is now and the being as it will be when it lives out
the valuable future. As there are few psychological connections between the fetus and its later self, it is concluded
that depriving the fetus of its future does not seriously harm it (and hence is not seriously wrong). A defence of this objection
is likely to rest, as with certain views of personal identity, on thought experiments involving brain or cerebrum swaps; and this may render it implausible to some readers.
The bodily rights argument
In her celebrated article A Defense of Abortion, Judith Jarvis Thomson argues that abortion is in some circumstances permissible even if the fetus has a
right to life. Her central argument involves a famous thought experiment. Imagine, Thomson says, that you wake up in bed next to a famous violinist. He is
unconscious with a fatal kidney ailment; and because only you happen to have the right blood type to help, the Society of
Music Lovers has kidnapped you and plugged your circulatory system into his so that your kidneys can filter poisons from his
blood as well as your own. If he is disconnected from you now, he will die; but in nine months he will recover and can be
safely disconnected. Thomson takes it that you may permissibly unplug yourself from the violinist even though this will kill
him. The right to life, Thomson says, does not entail the right to use another person's body, and so in disconnecting the
violinist you do not violate his right to life but merely deprive him of something—the use of your body—to which
he has no right. Similarly, even if the fetus has a right to life, it does not have a right to use the pregnant woman's body;
and so aborting the fetus is permissible in at least some circumstances.
Critics of this argument generally agree that unplugging the violinist is permissible, but claim there are morally
relevant disanalogies between the violinist scenario and typical cases of abortion. The most common objection is that the
violinist scenario, involving a kidnapping, is analogous only to abortion after rape. In most cases of abortion, it is said, the pregnant woman was not raped but
had intercourse voluntarily, and thus has either tacitly consented to allowing the fetus to use her body (the tacit consent
objection), or else has a duty to sustain the fetus because the woman herself caused the fetus to stand in
need of her body (the responsibility objection). Other common objections turn on the claim that the fetus is the pregnant woman's child whereas
the violinist is a stranger (the stranger versus offspring objection); that abortion kills the fetus whereas unplugging the violinist merely lets him die (the killing
versus letting die objection); or, similarly, that abortion intentionally causes the fetus's death whereas unplugging the violinist
merely causes death as a foreseen but unintended side-effect (the intending versus foreseeing objection; cf the doctrine of double effect).
Defenders of Thomson's argument—most notably David Boonin—reply that the alleged disanalogies between the violinist scenario and typical cases of abortion
do not hold, either because the factors that critics appeal to are not genuinely morally relevant, or because those factors
are morally relevant but do not apply to abortion in the way that critics have claimed. Critics have in turn responded to
Boonin's arguments. Thomson's argument thus remains highly controversial; but arguably it does at least show that the
moral impermissibility of abortion does not obviously and necessarily follow from the claim that the fetus has a right to