www.richmond-philosophy.net - © COPYRIGHT 2002-10 RUTC - Designed and maintained by Mat Carmody.

Is Peter Singer's Utilitarian Argument about Abortion Tenable?

Keith Crome


My aim in this essay is to examine Peter Singer's views concerning the morality of abortion, advanced in his Practical Ethics. I shall show that Singer's argument is not tenable, not because it is rationally unacceptable, i.e. self-contradictory or incoherent, but rather, because of the very rationality of Singer's position. In this respect, what I argue is that Singer's claims suppose an exclusively rational - and thus abstract - notion of ethical agency. However, before advancing this criticism, and in order to do so, I shall begin by explaining Singer's argument, itself established on the basis of what he perceives to be a fundamental weakness in the terms in which the debate over abortion is conventionally staged.

1. The Conventional Parameters of the Dispute over Abortion

According to Singer the issue of abortion is currently one of the most bitterly disputed of all ethical issues. The debate has been long-running and 'neither side has had much success in altering the opinions of its opponents'.[1] The common opinion is, then, that this is a moral problem without solution. As Don Marquis has argued, the debate over abortion is capable of generating opposed but 'apparently defensible syllogisms', and consequently something of 'a stand-off' between the disputing parties results.[2] For Singer, however, far from being the sign of a genuinely insoluble dilemma, the lack of consensus over the moral status of abortion is simply an indication that the problem has been wrongly posed. Thus, Singer claims 'there is a clear-cut answer' to the question of the moral permissibility of abortion and those who think there is not 'are simply mistaken' (PE 137).

Typically, Singer says, the dispute about abortion 'is often taken to be a dispute about when a human life begins' (PE 138). In order to establish why this is, and to demonstrate what he sees as the wrongness of this common approach to the problem, Singer formalizes the common argument against abortion as a syllogism:

First premise: It is wrong to kill an innocent human being. Second premise: A human foetus is an innocent human being. Conclusion: Therefore it is wrong to kill a human foetus. (PE 138)

It is because it seems incontrovertibly true that killing an innocent human being is wrong, that the usual response of critics of this argument is to attempt to deny its second premise, arguing that the human foetus is not a human being. As a result the debate around abortion has found its centre of gravity at this point, the argument focussing on whether, or at what point, the foetus can be said to be a human being and accorded the same protection against being killed as other human beings.

However, as Singer points out, the problem with this is that the development from fertilized egg to child is continuous. Consequently, it is difficult for those who want to defend abortion to establish a morally significant dividing line between the earliest stages of life when abortion would be permissible and the point at which that life turns into a properly human life, when it would not. Singer shows this difficulty in relation to four possible points at which human life is commonly suggested to emerge: birth, viability, quickening and the onset of consciousness. I should like to quickly outline the arguments that Singer makes regarding each one of these points.

Of the first, birth, Singer notes that we are sympathetically inclined towards the child that has been born, which we can 'see, hear and cuddle' (PE 138) and thus prone to morally favour it over the unborn foetus. However, for Singer, irrespective of our emotional sympathies - or prejudices - birth is incapable of marking that point at which one can say a being may or may not be killed. Not only is it certain that, in any given case, irrespective of whether the foetus/baby is in the womb or outside of it, it is the same being, but without doubt the child that is one week old is little different from one that is a week away from birth, and a prematurely born child is less developed - and thus less human - than one that is nearing the end of its normal term. Consequently, using birth to mark the point at which human life can be said to emerge and killing the child immoral is, for Singer, arbitrary. As Singer says, 'the location of a being - inside or outside the womb - should not make that much difference to the wrongness of killing it' (PE 139).

The second stage sometimes used to mark the point at which a human life properly begins is that of viability. If birth is the time at which the child actually begins to exist outside of the womb, viability is the time at which the foetus is deemed capable of so doing. Thus, if birth seems an utterly arbitrary point at which to distinguish between a being with the right not to be killed and one lacking such a right, viability is less obviously so. On the one hand, and as Singer points out, using such a marker as the point at which a human life begins undoes the patently unjustifiable discrimination between the prematurely born child and the viable foetus at the same stage of development. On the other hand, the stage at which independent life is possible is (perhaps a little unclearly) connected to our understanding that a properly human life is a life that is individuated. Because the viable foetus can be deemed potentially, if not actually, an individual life, then it at this point, it is claimed, that it should be accorded the same rights that all individual humans enjoy.

However, as Singer notes, one important objection to making viability the cut-off point is that viability is in no way a fixed, unalterable stage of pregnancy. Rather, it varies according to the state of, and access to, medical technology. Not only has the stage of viability has been pushed back by medical advances, but in a city with the most up-to-date medical facilities a foetus born six months prematurely would have a fair chance of survival whereas one born in a remote village in New Guinea, say, would not. Is it then the case, Singer asks, that we would say it was morally right in the past to have an abortion at six months and now morally wrong? Furthermore, would we say it is morally wrong for a woman whilst living in New York to abort at six months, but morally acceptable for her to do so if she flew to a village in New Guinea where the foetus would not be viable? 'The trip', he continues, 'does not change the nature of the foetus, so why should it remove its claim to life?' (PE 140). If, in response to such apparent absurdities, it is argued that it is the foetus' dependence on the mother for survival that means that it has no right to life independently of her wishes, then it would seem to follow that any human being dependent on another for survival would also forgo their own right to life. However, as Singer points out, we do not think that is the case, so 'it is not plausible to suggest that the dependence of the nonviable foetus on its mother gives her the right to kill it; and if dependence does not justify making viability the dividing line, it is hard to see what does' (PE 141).

Of the third point said to mark a morally significant distinction between the foetus and a properly human life, namely quickening, Singer is rapidly dismissive. Quickening is the time when the mother first feels the foetus move, and in Catholic theology it is held to indicate the moment at which the foetus is besouled. For Singer, this is nothing but 'an outmoded piece of superstition' (PE 141). Not only is it the case that the mother's perception of the foetus' movement has been shown to be antedated by non-perceptible movements, but, Singer argues, 'the capacity for physical movement - or the lack of it - has nothing to do with the seriousness of one's claim for continued life' (PE 141), for we do not posit the lack of the ability to move as grounds for denying paralysed people such a right.

Finally, Singer considers the onset of consciousness as having a claim to mark a morally significant dividing line between foetal life and a properly human life. Singer allows that consciousness - or awareness - and along with it the capacity to feel pleasure and pain, are of real moral significance, for it is, he argues, ultimately on the basis of the pain or pleasure that they cause that our actions must be judged. Nevertheless, Singer notes that neither opponents nor supporters of the right to abortion have made much appeal to consciousness. This, he argues, is because those who oppose abortion 'really want to uphold the right to life of the human being from conception, irrespective of whether it is conscious or not' (PE 142), whilst those who favour abortion, recognize it as a risky strategy, since evidence points to the existence of consciousness as early as the seventh week after conception and a similarly early date for the capacity to feel pain.

All in all, then, the search for a morally significant dividing line between human life and mere foetal life, a dividing line that would legitimate abortion, has not been fruitful. Lacking such a line it is necessary either to assimilate the moral status and rights of the child to that of the foetus or to assimilate those of the foetus to that of the child. Since no-one wants to allow parents the right to terminate the life of their children on request, it seems that the only defensible position is to accord to the foetus the rights and protections that are accorded to the child.

Despite this, Singer concludes that it would nevertheless be premature to assume that the case against abortion is sound. This is not because no argument can ever definitively rule out the possibility of establishing a morally relevant dividing line between foetal life and human life, but because it is necessary, according to Singer, to look at the problem in a different way. It is to this different way of posing the moral problem of abortion that I now want to turn.

2. Singer's Alternative

As we have already seen, according to Singer the common argument against abortion can be formalized in the following way:

First premise: It is wrong to kill an innocent human being. Second premise: A human foetus is an innocent human being. Conclusion: Therefore it is wrong to kill a human foetus.

As we have also seen, attempts to undo this argument have sought to invalidate the second premise, seeking to show that the foetus is not a human being. However, such attempts seem at best inconclusive, at worst unconvincing.

Now, if the second premise has frequently come under scrutiny, the first has, by and large, been ignored because there is 'widespread acceptance of the doctrine of the sanctity of life' (PE 150). It is this widespread acceptance that Singer seeks to challenge. According to him, the weakness of the first premise lies in the ambiguity of the term 'human', and it is through this ambiguity that the syllogism works. As he points out, the term 'human' can, and often does in moral discussions, mean either 'a member of the species Homo sapiens' or 'being a person' (PE 150). It is the play between these two senses that renders the syllogism acceptable. When the senses are distinguished, the argument is immediately falsified. On the one hand, if 'human' means 'a person', i.e. someone who is 'rational or self-conscious' (PE 150), then whilst there might be some virtue to the first premise, the second is immediately false. On the other hand, if 'human' means 'member of the species Homo sapiens', then, the second premise is trivially true, but the first is false, for 'whether a being is or is not a member of our species is, in itself, no more morally relevant to the wrongness of killing it than whether it is or is not a member of our race' (PE 150).

Following the recognition of this ambiguity, it is necessary, Singer argues, to refocus the entire debate over abortion. Rather than accept that species membership automatically suffices to establish certain rights for the foetus - 'a legacy of religious doctrines that even those opposed to abortion hesitate to [openly] bring into the debate' (PE 150) - we must look at the actual characteristics that the foetus possesses which are morally significant in terms of deciding whether or not it is acceptable to kill it. As a utilitarian, for Singer those characteristics are ultimately related to the capacity to feel pain, such as 'rationality, self-consciousness, awareness' (PE 151). Until such capacities exist then abortion can be said to terminate a life that is of no intrinsic value. Afterwards, it is a matter of determining to what degree such capacities exist and of weighing the interests of those parties affected by the abortion (principally the mother and the foetus) according to a utilitarian calculus. Thus, according to Singer:

When the foetus may be conscious, though not self-conscious, abortion should not be taken lightly (if a woman ever does take abortion lightly). But a woman's serious interests would normally override the rudimentary interests even of a conscious foetus. (PE 151)

As Singer recognises, one possible weakness of such an argument is that it takes into account only the actual characteristics of the foetus. It could be objected that what is at issue in relation to a foetus is not simply its actual characteristics but also, and above all, its potential characteristics. Whilst the foetus may actually lack rationality, consciousness, the capacity to feel or to suffer in the same degree as a properly human person, it nevertheless will, ordinarily, come to have such characteristics. Thus, it is because of its potential that it is necessary to accord to the foetus the same right against being killed as is accorded to an ordinary human.

However, in relation to such an objection, Singer observes first that there is no general rule that allows us to infer that a potential x has the same value as an actual x. Rather, according to Singer, many examples illustrate the contrary: 'to pull out a sprouting acorn is not the same as cutting down a venerable oak. To drop a live chicken into a pot of boiling water would be much worse than doing the same to an egg' (PE 153). Lacking a general rule, it is necessary to seek a specific reason for making such an inference in the case of potential persons. Generally, Singer argues, if we value the life of a person more highly than the life of other animals, and accord it greater protection, it is for reasons connected with the 'capacity to see oneself as a continuing mental subject [...] as a distinct entity with a past and a future' (PE 153). Such reasons, however, do not apply to beings that do not and never have seen themselves in such a way, and thus 'the mere potential for becoming a person does not count against killing' (PE 154).

The argument from potential is not exhausted, for, as Singer recognises, it can be claimed that in this instance potential is important not for the rights that it endows the foetus with, but because killing a foetus deprives the world of a future rational and self-conscious being. However, for Singer such a way of construing the argument from potential leads to numerous problems and absurdities. First, not all abortions can be said to deprive the world of such beings. To make his point, Singer asks us to imagine a case where a woman learns she is two months pregnant, and that such a pregnancy, preventing her from doing something she wants to do, is inconveniently timed. After completing what she wants to do, which lasts only a few weeks, she intends - and does - become pregnant again. In such circumstances, the world is not deprived of a self-conscious, rational, being: it is only that the entry of such a being into the world is delayed. Yet, as Singer notes, opponents of abortion would consider an abortion under such circumstances morally outrageous. Secondly, construed in this way the argument from potential applies not only to abortion but to contraception, whether it be artificial or natural, and places both on a par with one another, which is again something that most, if not all, opponents of abortion would not want to accept.[3]

3. From Utilitarian Ethics to Virtue Ethics

Doubtless there are many more objections that could be raised against Singer's argument, (some of which he considers and responds to in Practical Ethics). However, I do not wish to consider such objections, nor do I think it is necessary to do so, for as I said in my introduction, I am not concerned to show that Singer's argument is rationally flawed. In fact, the exposition I have given of that argument, the possible objections to it, and Singer's responses to those objections, show that it is logically strong and rationally well-supported. Instead, I want to contend that there is a limit to the role of rationality in relation to morality and thus to any consideration of the morality of abortion. In other words, I will argue that ethical agency involves more than the mere application of reason, that it is also a matter of acting-well, of attaining a fulfilled and happy life (what the virtue ethicist would call a eudaimonic life), and such a life must take account of more than that which can be rationally calculated - it must also include the whole range of our emotional responses to given situations.

In order to show this, let us suppose that Singer has established that it is not immoral to abort a foetus. That granted it follows that in having an abortion a woman is doing nothing morally wrong. However, in itself that does not suffice to establish that she is doing the right thing. It is perfectly consonant with the judgement that abortion is not immoral, that in having an abortion a woman may nonetheless be acting in a way that is vicious rather than virtuous. A simple example will clarify what I mean. If, as Singer claims, prior to the eighteenth week of gestation the foetus, lacking a sufficiently developed cerebral cortex, cannot feel pain, (see PE 164 - 5) then there is no intrinsic reason why killing the foetus is wrong. Indeed, considering Singer's arguments against using the notion of potentiality as a moral criterion in relation to abortion, it is impossible to see why, on these grounds alone, one might want to condemn the woman who, two-months pregnant, has an abortion because her pregnancy is inconveniently timed and prevents her from doing something she wants to do. I am certain that Singer does not want to suggest that the action of such a woman is praiseworthy, but it is impossible to see on what positive grounds he would be able to condemn it. Perhaps he would not want to do so, yet without doubt ordinarily it would be condemned. It is not difficult to see why that is (although it is difficult for Singer to say why): such an action would be condemned not because it is, in itself, morally wrong to abort a foetus at such a stage, but because it would be a vicious and selfish act, and we immediately and emotively sense this to be the case.

As Bernard Williams has pointed out, the utilitarian would have nothing positive to say about such an immediate and emotive response.[4] That is because, for the utilitarian such a response would have no utilitarian basis: feeling this way would not be justified by any assessment of the consequences of the act and hence would simply be wrong-headed and irrational. Yet, it is not that such a response is simply groundless or wrong; rather, it is based on our appreciation of the seriousness of the act of abortion, which is itself bound up with our sense of the affective value of both family relationships, and the events of birth and death. It is, I think, one of the real limitations of utilitarianism that it is unable to recognise the intrinsic value of our feeling this way. In this instance, what our ethical feelings tell us is, I think, that, as Rosalind Hursthouse has argued, the premature termination of a pregnancy is a serious act, connecting with our concerns and thoughts and, just as importantly, our feelings about death, parenthood and family relationships. For this reason:

to disregard this fact, to think of abortion as nothing but the killing of something that does not matter, or as nothing but the exercise of some right one has, or as the incidental means to some desirable state of affairs, is to do something callous and light-minded, the sort of thing that no virtuous and wise person would do.[5]

Now, in what I have just said there is an implicit appeal to eudaimoina, since someone who can manifest such a selfish or callous attitude towards the act of abortion is not, properly speaking, acting in such a way as to lead a happy or fulfilled life. I should like to say something more about the notion of eudaimonia, which is a central - and much debated - term in virtue ethics, to clarify what I think lies at the basis of the judgment concerning the callousness of the woman who have an abortion in such circumstances as those outlined above.

At first, it might seem a provocative and ill-supported claim - a judgment too far - to assert that someone who has such an attitude towards abortion is not acting in a way conducive to a happy or fulfilled life. How, it might be asked, is it possible to make such an evaluation? Is it precisely not the case that in this scenario the woman who has had the abortion, has done so in order to secure for herself a happy, fulfilled life? And, as is sometimes said, is not happiness one of those things that it is impossible for an individual to be mistaken about? If the woman genuinely thinks or feels that she is happy, must it not be allowed that she is happy? Similarly, if she genuinely thinks or feels herself to be fulfilled, must it not be allowed that she is fulfilled?

But this would be to misunderstand what is at issue in the term eudaimoina, a misunderstanding that arises, in part, from the meaning we now commonly give to the terms 'happiness' and 'fulfillment'. In their modern acceptation these terms are taken to denote subjective states or feelings, and this seems to preclude any really meaningful agreement about what might constitute a happy or fulfilled life. Certainly, it has been recognized by proponents of virtue theory as much as by its critics that what constitutes a happy or fulfilled life is indeed difficult to define, but that difficulty - compounded by the terms used to translate it - derives not from the fact that it is impossible define what is at issue meaningfully (because it is entirely subjective), but rather because what is at issue is so meaningful: eudaimoina is a rich, thick, concept.

I am not simply saying - although it is not wrong to say this - that eudaimoina is a complex concept, involving more than one thing, and so difficult to define. This is true of most - perhaps all - philosophical concepts, and if it is worth remarking on in this instance it is because people mistakenly take the difficulty of defining eudaimoina for an impossibility, and wrongly attribute this impossibility to the supposedly subjective status of 'happiness' or 'fulfillment'. However, beyond this, what I am saying is that eudaimoina is a rich, thick, concept because it is a concrete concept, and it is a concrete concept because it refers to an individual's existence - his or her manner of being - as a whole.

What I mean by this can perhaps be best clarified here by a very simple consideration of the virtues, and the connection that obtains between them and eudaimoina. The virtues - courage, kindness, temperance, and so-on - are not simply attributes, that is, they are not simply 'things' that some humans possess and some not, and some possess to a greater extent than others, in the way that certain objects have certain attributes that others do not have, or have to a greater or lesser extent. The virtues are excellences of character that denote dispositions and traits, the habitual inclination to act, react and feel in a certain way. In short, they are possible ways of being, and as such, in the most profound sense, they reveal and define who we are. It is for this reason that we are never indifferent about the virtues: if someone says I have stubby, ugly, fingers or knobbly knees I shrug it off - it might be true, but they are my fingers and knees! However, if someone says I am cowardly or intemperate I am profoundly aggrieved. I cannot be indifferent about the virtues precisely because it is me - my essential manner or mode of being - that is at issue. Now, in so far as there is the virtues are excellences of character that are requisite to an eudaimonic life it is clear that eudaimoina does not denote a particular state of mind or being in the trivial sense of being happy or fulfilled; instead, eudaimoina is a matter of an individual realising through their practical actions and deliberations their excellence as a human and an individual, that is, of fulfilling the highest possibility of their being.

I have to admit that this is a very compressed, and tendentious account of virtue ethics, and that is principally because it is not my intention here to offer an account of virtue ethics per se (rather my intention in this essay is simply to show the mistake of trying to give an account of ethics in rational terms). However, in the light of these remarks about both the virtues and their connection with eudaimoina we are, I think, in a better position to appreciate why we would adjudge an abortion undertaken for the reasons given above to be vicious and selfish, and to say that the woman who chose to act in this way would not be happy or fulfilled. That is because an integral part of who we are as individuals, an integral aspect of our individuality, comprises of our relationships with others and in particular familial relationships, our capacity to care for others, to forego certain things for the love of another being, and also our attitude towards birth and death. These are not accidental or arbitrary features of our individual identities, but essential to them, and anyone who treats them trivially manifests a disregard for their own selves, and cannot, in the sense that I have set out above, be said to live a fulfilled or fulfilling life.

Without doubt there is much more that could be said about all of this. Certainly, I do not want to be taken to suggest that abortion is unacceptable; I think - and my argument reflects this - that the virtue or viciousness of the act is only ever circumstantially determinable. If it is the case that arguments over the rightness or wrongness of abortion have not been settled it is not necessarily because the problem has been wrongly posed as Singer suggests, but that an abstract and universal morality is itself undesirable. In relation to this, it is worth making one final point: if we do in fact adjudge that in a certain circumstance the act of abortion was vicious, this does not mean that we should condemn the woman who acted in this way. Rather, if we ourselves are to act in a virtuous and not vicious way, we should ask what we should do to alter the circumstances so that the woman who made such a choice would not be compelled to do so again.

In sum, then, I am quite happy to allow that Singer provides a rationally compelling argument concerning the morality of abortion. Indeed, I have sought to show this, whilst still claiming that ultimately such an argument is not tenable. This is because Singer's position is too rational and abstractive, lacking a concrete appreciation of ethical agency. Such a concrete appreciation is rooted in our apprehension of what constitutes a good, flourishing or eudaimonic life.

Keith Crome
Department of Philosophy
Manchester Metropolitan University


References and Further Reading

By now the classic utilitarian account of the main topics in the field of practical ethics is Peter Singer's Practical Ethics (Cambridge: Cambridge UP, 1993). The chapter on abortion is the focus of this essay.

An outstanding anthology of essays on topics in practical ethics written from various moral standpoints is Ethics in Practice: An Anthology, ed. H. LaFollette (Oxford: Blackwell, 2002).

Bernard Williams has written one of the most penetrating critical accounts of utilitarianism as part of the book Utilitarianism: For and Against by J. J. C. Smart and B. Williams (Cambridge: Cambridge UP, 1973). Parts of Williams' critique have been anthologised (see, for example, B. Williams, 'Utilitarianism and Integrity' in The Right Thing to Do ed. James Rachels (London: McGraw Hill, 2006), pp. 76 - 83).


[1]  Singer, P., 1993, 'Taking Life: The Embryo and the Foetus' in Practical Ethics, (Cambridge: Cambridge UP) p. 135. Hereafter all references to this text will be given as PE followed by the page number and given in brackets in the main text.  [back] 

[2]  Marquis, D., 2002, 'An Argument that Abortion is Wrong' in Ethics in Practice: An Anthology, ed. H. LaFollette (Oxford: Blackwell) p. 85  [back] 

[3]  Here we should note that under the papacy of Benedict XVI, even the Catholic Church is considering revising its strictures on contraception.  [back] 

[4]  See Williams, B. 2006, 'Utilitarianism and Integrity' in The Right Thing to Do ed. J. Rachels (London: McGraw Hill), pp. 76 - 83.  [back] 

[5]  Hursthouse, R., 2002, 'Virtue Theory and Abortion', in Ethics in Practice: An Anthology, ed. H. LaFollette (Oxford: Blackwell), p. 99.  [back]