The Ethics of Killing: Problems at the Margins of Life 
by Jeff McMahan.
Oxford, 554 pp., £35, February 2002, 0 19 507998 1
Show More
Show More

Two-thirds of the way through this dense, involved and exhausting book, its author acknowledges that his views about the nature of persons have the following implication. Suppose that a woman, without family or friends, dies giving birth to a healthy infant. At the same hospital there are three five-year-old children who will die if they do not receive organ transplants, and the newborn has exactly the right tissue type. If Jeff McMahan’s theory is right, it is morally permissible to ‘sacrifice’ the orphaned infant in order to save the other three children.

We can hold off for a moment on the question of why his theory has this implication. The simple fact that it does, as he points out, appears to be a reductio ad absurdum of his position. And McMahan has the grace to confess that he ‘cannot embrace’ this implication ‘without significant misgivings and considerable unease’. But he embraces it nevertheless, since he continues his examination of abortion and euthanasia for a further 150 pages, still drawing on the same account of the nature of persons that led to the apparent reductio in the first place – as if simply acknowledging its existence constituted a sufficient settling of accounts with it.

Reduction to absurdity is, of course, a fate that any philosophical theory might meet – although it is relatively rare for a theorist simultaneously to underline and ignore this flaw in his progeny. But such split-mindedness is peculiarly disturbing when the theory concerns itself with morality, since in such a context a reductio argument confronts us with moral absurdity – or, to put it more bluntly, with the morally intolerable, the morally unthinkable. Elizabeth Anscombe once said that anyone who thought in advance that it was open to question whether an action such as procuring the judicial execution of an innocent person should be entirely excluded from consideration showed a corrupt mind. She thereby expressed a (highly controversial but hardly unintelligible) fear not only of evil actions, but of thinking evil thoughts – a fear of the dulling and degrading of moral sensibility that such acts of contemplation can encourage and express. The analogy with McMahan’s moral absurdity is not exact. A morally innocent victim is certainly central to his cautionary tale; but while (to his credit) he cannot bear explicitly to accept this evil consequence of his theory, he also cannot bear to reject the theory on its account. On the contrary, he acts as if his claim to be a serious moral thinker would be more severely damaged if he took such evil consequences as sufficient reason to abandon his theoretical endeavour. Being a fearless thinker matters more, it seems, than avoiding morally fearful thoughts.

In this respect, his work is representative of the mainstream of contemporary practical ethics in the Anglo-American philosophical world. Take the fact that the moral intolerability of his theory is revealed by invoking an imaginary case. The Ethics of Killing is full to bursting with such thought-experiments, or ‘cases’ as McMahan calls them. Their deployment is a central technique in the modern moral philosopher’s tool-kit; it is even the way some philosophers achieve a kind of immortality – as with Judith Jarvis Thomson and her ‘case’ of the woman to whom a violinist is hooked up for life support, a case designed to illustrate a certain argument for abortion (McMahan’s critical discussion of it is the best I have seen).

The Ethics of Killing puts this technique to more systematic and imaginative use than any other book I know of. McMahan’s cases even get a separate, and very useful, index – although its enigmatic entries (‘The Pipe Sealer’, ‘The Accidental Nudge’, ‘The Whole-Body Transplant’) irresistibly reminded me of the ‘Concordance of Nicola Six’s Kisses’ in Martin Amis’s London Fields (‘The Rosebud’, ‘Clash of the Incisors’, ‘The Turning Diesel’). However, inventing, varying and examining such cases requires and encourages the exercise of a certain kind of imagination – the kind evoked by the label ‘thought-experiments’. These cases are devised rather than created, modified rather than rewritten, analysed rather than entered into; they invite precisely the epithets with which McMahan’s book has been greeted – ‘novel and ingenious’, Peter Singer tells us on the dust-jacket, rather than, say, ‘wise and insightful’.

McMahan’s ‘Sacrificial Newborn’, with which we began (my title, not his; strangely, it doesn’t appear in his index), is a case in point. Its sole rationale is to present us with a lightly clothed calculation. Eliminate anyone with whom the newborn might have a human relationship (since their distress might complicate the sums), and whistle up enough older children to outnumber our orphan (thereby forcing us to acknowledge that three is at least three times greater than one). It’s an arithmetical tale, morality by numbers – and the simpler the texture, the clearer the point.

But what if our concern were not clarity but understanding, or an engagement of our moral imagination with something resembling the texture and complexity of human reality? Not only is medical unlikelihood bypassed for McMahan’s purposes (no tissue-typing problems); we hear nothing of the family and friends of the three five-year-olds. This is presumably because their obvious joy at the redemption of their children would simply shift the balance of calculation even further in the same direction. But what if the mother of one of these children discovers the source of her daughter’s new organs? Would her joy be untainted by this knowledge? Is it obvious that she, or any of the parents involved, would regard it as legitimate for a healthy orphan to be thus abused? Is it obvious that any of them, or indeed any of the hospital staff, would not feel an obligation to the memory of the newborn’s dead mother that might make them hesitate over its ‘sacrifice’?

These questions may seem entirely unfair to McMahan, whose purpose is not to produce a gripping and gritty exercise in literary realism, but rather to make a theoretical point – indeed, a point against his own theory that my questions simply underline. Certain possible complications in the telling of the tale are excised simply because they are irrelevant to the issue at hand, which is the relative strength of the moral claims made on us by human infants and young children. It is to this matter alone that McMahan wants us to direct our moral intuitions, and his case is constructed so as to clear the scene of any other potentially polluting concerns.

This kind of disagreement about the value of thought-experiments in ethics was once raised from a different perspective by Carol Gilligan. She presented groups of students with a problem case: would it be morally permissible for an impoverished person to steal medicine from a chemist in order to save the life of her sick child? The men argued with one another about the immorality of theft and the sanctity of human life; the women inquired: ‘Why doesn’t she ask the chemist to give her the medicine she needs?’ Some will take it that the women had simply misunderstood the point of the problem case, and of problem cases in general; for to include such a dialogue between chemist and parent is to change the case, and thus to avoid responding to the particular issue of principle it was intended to abstract from the complexity of reality. Others will take it that the men had been distracted by matters of principle from attending properly to the concrete reality of moral experience and the possibilities of human fellowship.

The difficulties I raised in the case of the Sacrificial Newborn are analogous. First, its extreme condensation forces its deviser to take certain moral valuations for granted; hence McMahan treats it as obvious that the relatives of the sick five-year-olds will rejoice in their salvation, and that the death of the newborn’s mother removes her wishes and concerns entirely from the equation – hardly self-evident assumptions from many moral perspectives. And this raises the general difficulty of separating the value-neutral facts of a case from the moral intuitions we are supposed to bring to bear on it. If our differing intuitions lead us to contest more or less every step in the construction of the case, the idea that a value-neutral account is available even in principle will seem deeply questionable. Second, thought-experiments in ethics presuppose that we can get clearer about what we think on a single, specific moral issue by abstracting it from the complex web of interrelated matters of fact and of valuation within which we usually encounter and respond to it. But what if the issue means what it does to us, has the moral significance it has for us, precisely because of its place in that complex web? If so, to abstract it from that context is to ask us to think about something else altogether – something other than the issue that interested us in the first place; it is, in effect, to change the subject.

The general structure of McMahan’s book raises another version of precisely this question. His treatment of the specific issues of abortion, infanticide, euthanasia and suicide (in Parts Four and Five) applies conclusions reached by his prior discussion of the nature of death and killing (in Parts Two and Three), which are in turn guided by conclusions reached in his opening account of the nature of persons. This approach appears so obviously right to McMahan that he barely pauses to explain it; but presumably the line of thought goes like this. Only if we understand the nature of persons can we understand what it is (and why it is bad) for them to die, and only then can we understand what is bad about causing them to die by killing them, and only then can we understand what, if anything, is bad about the specific forms of killing known as abortion and euthanasia.

This certainly sounds sensible: let’s settle the more general issue before moving to the more particular; or, rather, let’s get the rational foundations of our moral thinking in place before we engage in specific moral disputes. But, for McMahan, these rational foundations are not themselves moral; clarifying the nature of persons is treated in the first part of his book as a metaphysical rather than an ethical issue. He aims, before doing anything else, to establish the essential nature of a certain kind of entity, and to do so in ways which, while they may (and indeed, must) have consequences for our moral thinking, are themselves essentially ethically neutral.

His account of personhood builds on one of the most influential contemporary pieces of metaphysical analysis in this area – that of Derek Parfit. Parfit claims that, in normal circumstances, personal identity (what makes me the same person across time) is a matter of the holding of certain relations of psychological continuity and connectedness – relations of memory, desire, intentions and their fulfilment, and so on. McMahan revises this account so that psychological continuity is treated more broadly, and as a matter of degree; but most fundamentally, he argues (against Parfit’s original position, but with reference to his most famous thought-experiment, that of a Star Trek-style teletransporter) that identity is preserved only if the relevant psychological continuity is grounded in the physical and functional continuity of enough of those areas of the person’s brain in which consciousness is realised. Hence, according to this ‘Embodied Mind’ account, a person in the early stages of Alzheimer’s has reason to be egoistically concerned about what may happen to his body even in the final phases in which the mental life associated with his body will no longer be even weakly psychologically connected from day to day.

Whether this account really treats the mind as embodied is dubious. McMahan seems rather closer to presenting us with a brain-body dualism in place of the religious/Cartesian mind-body dualism he detests (transplant the right part of a person’s brain into a new body, and you preserve his identity). Certainly, when the embodiedness of his favoured brain-parts is thrust on his attention, he entirely avoids its true significance. This becomes most obvious in another of his cases, that of Dicephalus. This is a rare medical condition, which occurs when a human zygote divides incompletely, and results in twins conjoined below the neck. Referring to the real case of Abigail and Brittany Hensel, McMahan asserts with startling certitude that ‘no one doubts that they are separate and distinct little girls.’ Each, he tells us, has her own private mental life and her own character; there are ‘of course’ two personal or biographical lives, although their shared set of organs sustains a single biological life.

McMahan makes no effort whatever to imagine the meaning of his claims to the girls themselves. Will Abigail feel that she has an utterly distinct personal or biographical life from her sister, or her sister from her? Neither can ever play on her own with other children, have a happy or tearful private conversation with her mother, retreat to her room to rage or sulk or think in solitude, or go out alone with a boyfriend. Suppose they were to have a child; as McMahan notes in passing (but only to confirm that their two minds share a single body) it would be the child of both, a child with three parents. What significance might their motherhood have for each of them ‘biographically’, given its rootedness in their common flesh? Would it confirm, modify or undermine their sense of separateness and union? And what significance will their private mental life have to them when it exists only in the context of such enforced, embodied intimacy with one another?

McMahan’s blithe treatment of this ‘case’ fails to appreciate that the nature of their embodied lives cannot be broken down and distributed between biological union and psychological distinctness in any straightforward way. If a sense of the separateness of persons is part of our concept of personhood, and that sense would be radically disrupted in the case of dicephalic twins, then so would our concept of personhood. Our ordinary concept of a person has the sense and the significance it has because it is embedded in the normal circumstances of our embodied lives with others; alter those circumstances, and our ordinary concept will not simply carry over, and hence its structure cannot be illuminated by their study.

It is not as if we develop a concept of a person (say, as a psychologically continuous entity), and then relate to those we identify as persons in ways we judge appropriate to their metaphysical genus, so that those relations might be evaluated for their consistency with our independently given nature. Rather, our concept of a person is constituted by, and finds its life and sense in, the context of the normal forms of our lives with other persons – with embodied, flesh-and-blood creatures inhabiting structures of language and culture. And since those lives have a moral dimension, since their commonality and variety cannot adequately be characterised except in terms which invite ethical questions and issues (as Abigail’s relations to her sister, her parents and her children plainly do), the same is true of our concept of a person. Personhood is not the metaphysical foundation of an interpersonal ethics; it is itself an ethical notion. To attempt to analyse it while remaining morally neutral is bound to produce exactly the air of mad conceptual science found in McMahan’s description of his Dicephalus case.

This pattern of thinking recurs throughout the book. It is central to McMahan’s treatment of killing, and more specifically to his sense of the relative moral status of human and nonhuman animals. Recognising that human interpersonal relations typically exhibit a requirement of mutual respect, McMahan looks for its rational and metaphysical basis. He finds it in the mature human being’s possession of certain psychological capacities – perhaps a rationally guided will, or sophisticated forms of self-consciousness. He then points out that some human beings lack these capacities, and concludes that they must fall below the threshold of respect. Their moral claims on us should be assessed not in terms of inviolable rights but in terms of their interest in continuing to live, and this interest can be compared with (and sometimes traded off against – hence the Sacrificial Newborn) competing interests. More specifically, certain higher nonhuman mammals, being possessed of at least analogous versions of the psychological continuity that grounds the human interest in continuing to live, should be seen as having analogous second-tier moral status to that of human foetuses, infants and the severely disabled.

For McMahan, then, the moral status of human beings is not univocal, and moral status of the second-tier variety at least is not restricted to members of the human species. To deny this second claim is, he tells us, to exhibit speciesism; it is to load moral significance onto a purely biological category. Suppose, he asks us to imagine, we administer genetic therapy to a canine foetus which confers human levels of self-consciousness, rationality and autonomy on it as it develops; it would then have become a person in the morally relevant sense of that term. Why, then, do we have any more reason to administer the therapy to a severely retarded human adult than to a normal, healthy dog?

This, McMahan tells us, is a serious problem indeed. Trying to convince us, he says: ‘Let us assume, for the sake of the argument, that a dog with human intelligence could have a life that would be well worth living even in a society in which it would be a freak, would have no acceptable mate, and so on. In short, let us put those contingent problems aside.’ Easy for us to do; not so easy for the dog. Would a human being, deprived of any acceptable mate and regarded as a freak by his fellows, be faced with merely contingent problems that would leave his capacity to conceive of himself as a person essentially unaffected? What interpersonal relations (of friendship, family, gossip, common hobbies and interests) would be conceivable for our Superdog? And in their absence, what would the sense be of calling it a person nonetheless? I don’t say that there could be no sense in doing so. I say only that the sense it would have is not the sense it has when human beings acknowledge one another as persons. The forms of embodied common life open to distinctively human creatures provide the context within which our notion of personhood has the sense it has. These forms are not the practical enactment of a logically prior intellectual hypothesis about capacity-possession that might turn out to be metaphysically ungrounded.

In other words, our concept of a person is an outgrowth or aspect of our concept of a human being; and that concept is not merely biological but rather a crystallisation of everything we have made of our distinctive species nature. To see another as a human being is to see her as a fellow-creature – another being whose embodiment embeds her in a distinctive form of common life with language and culture, and whose existence constitutes a particular kind of claim on us. We do not strive (when we do strive) to treat human infants and children, the senile and the severely disabled as fully human because we mistakenly attribute capacities to them that they lack, or because we are blind to the merely biological significance of a species boundary. We do it (when we do) because they are fellow human beings, embodied creatures who will come to share, or have already shared, in our common life, or whose inability to do so is a result of the shocks and ills to which all human flesh and blood is heir – because there but for the grace of God go I.

This supposedly speciesist conception of humans as fellow-creatures in fact provides a far more satisfying and powerful way of reconceiving our relations with nonhuman animals than anything McMahan’s argument provides. For nonhuman animals, too, are our fellow creatures in a different but related sense. Their embodied existence, and hence their form of life, is different; but in certain cases, the human and the nonhuman forms of creaturely existence can overlap, interact, even offer companionship to one another, and in many cases, nonhuman animals can be seen as sharing a common fate with us. They too are needy, dependent, subject to birth, sexuality and death, vulnerable to pain and fear. Seen in this light, the worst excesses of factory farming stand out clearly enough without further metaphysical analysis.

Such light is most clearly and sustainedly cast in the work of poets, writers and thinkers – the ethical significance of mortality is a literary and philosophical perennial. But McMahan has a tin ear for such things. Indeed, his discussion – at the outset of his examination of death – of the treatment of mortality in Western culture is the one point in this enormous book when his equanimity deserts him, to inadvertently comic effect. For what he finds in this tradition is ‘a dreary record of evasion and sophistry’. Socrates ‘prates’ about obtaining a great good in the other world while drinking his hemlock (no chance of irony here, just hypocrisy); Schopenhauer’s attempt to reconcile us to our place in the unending cycle of life is ‘wringing a meagre drop of consolation from the rags of bad argument’; and Tolstoy – well, ‘The Death of Ivan Ilych’ is of course a masterpiece, but it is also a ‘conspicuous piece of evasion’. Ivan’s deathbed vision is a deus ex machina which irritatingly leaves us to guess at his reassuring illumination. When one of Tolstoy’s biographers tells McMahan that it is the acceptance of his own mortality that brings Ivan peace, he is unsatisfied. ‘Not only is there no textual support for this conjecture, but insofar as Ivan Ilych accepts mortality, this is not an explanation of his finding peace but is itself precisely what requires explanation.’

I suppose we shouldn’t be surprised at this degree of incomprehension, since Tolstoy’s story precisely articulates its acknowledgment of mortality through Ivan’s gradual acceptance of himself as an unexceptional human being, as confronting a fate common to all human creatures. That acceptance begins when he is himself accepted as a fellow-creature by his servant, Gerasim (‘we shall all of us die, so why should I grudge a little trouble?’ Gerasim remarks, as he massages his master’s aching legs), and so begins to accept Gerasim as his fellow.

McMahan won’t find a sentence in which Ivan says ‘I must accept my mortality!’; but every word in Tolstoy’s text works to articulate such a vision. There are no premises, independently established, allowing us to advance further hypotheses or to draw novel and ingenious conclusions. Rather, Tolstoy gives us a way of seeing ourselves as mortal creatures subject to a common and uncanny fate, but capable nevertheless of doing the right thing. Here at least we might find some genuine sustenance for our moral imaginations; we might even be brought thereby to consider drawing in those on the margins of life rather than confirming their marginalisation as a requirement of reason.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 24 No. 17 · 5 September 2002

In June 1944 the chance of war has made a young British officer part of an Italian resistance formation. When an Allied arms drop is signalled he is in charge of its reception. Arriving at the dropping zone he meets a group of Italians led by his second in command, an Italian officer, who takes him aside and says that there is a man too many in the group – a man unknown to anyone. The British officer interrogates the man – he is in his late teens – who is unable to explain his presence. It would be impossible to keep a prisoner secure. The drop is imminent. The British officer consults with his Italian colleague who says: You are in command here. The British officer says the man has to be got rid of. He is duly shot.

Having been that officer, I am still worried by the ‘case’, which seems to me more difficult to deal with than an imaginary one concerning the ‘sacrifice’ of an orphaned infant, discussed by Stephen Mulhall (LRB, 22 August).

Stuart Hood

Vol. 26 No. 16 · 19 August 2004

What makes David Wootton think that the ‘mysterious mathematics’ of conjoined twins raises questions which ‘have been ignored by philosophers’ (LRB, 22 July)? The moral dilemmas presented by such cases have been extensively debated in the field of applied ethics, usually as part of broader discussions about organ transplants, euthanasia, abortion and the treatment of severely disabled people. The implications of these cases for our understanding of personal identity have been at the heart of contemporary debates in the philosophy of mind, usually in conjunction with a variety of real and imagined cases of brain-division, brain transplantation, bodily fusion and so on. Whether any of these discussions have advanced our moral or metaphysical understanding may, of course, be doubted; but that is a rather different charge.

As is often the case, ignorance of the relevant philosophical debates here nourishes fantasy. Take Wootton’s critique of the legal judgments made in the case of the conjoined twins Gracie and Rosie Attard. He rightly rejects the suggestion that Rosie ‘sucks the life-blood out of Gracie’, but then claims that a better image for their position in a shared body is that of passengers in an overcrowded lifeboat: ‘The lifeboat belongs equally to each one of them.’ The modern source of debates about the relation between human identity and embodiment is Descartes, who is often accused of advocating a simple (even simple-minded) dualism, characterising persons as mind conjoined with bodies. In his Meditations, he explicitly considers the suggestion that a person’s mind is lodged in her body as a pilot is in the ship whose course he is guiding, but rejects it on the grounds that it underestimates the intimacy of the union. Wootton’s image is more Cartesian than Descartes’s own. Even if it permits Rosie to be as much a ghost in this physiological machine as Gracie, both appear as lodged in their body, as if it is their (common) possession; so their body appears as possessed or haunted by their minds or psyches or souls – in any case, by them. Is this really the right way to acknowledge that humans are embodied, flesh and blood creatures?

Stephen Mulhall

David Wootton’s statement that obstetricians face more difficult choices in cases of conjoined twins than they do in cases involving the priority as between a foetus and its mother may be simplistic. English law does favour the survival of a mother over an unborn foetus, but Shyam Kothan, in a recent piece in the British Medical Journal, records a case in which a fully competent and informed pregnant woman chose the life of her unborn child over her own: ‘Just as a runner passes on the baton to the next one in a relay race, so she gave birth to her daughter and died.’ Wootton suggests that had Rosie and Gracie been able to speak, if Gracie had protested, ‘Stop it [Rosie], you’re killing me,’ Rosie might well have responded: ‘It isn’t your life-blood, Gracie, it’s ours.’ However, Kothan’s example raises the possibility that either sister may have willingly sacrificed herself for the other.

Wootton’s claim that for ‘two centuries no one has felt the need to ask whether Byrne ought to be on display’ is incorrect: Hilary Mantel did just that in The Giant, O’Brien in 1998.

Debbie Lawlor

Vol. 26 No. 17 · 2 September 2004

I was surprised by the tone of Stephen Mulhall’s letter in response to my essay on conjoined twins, and to be told that I had demonstrated ignorance of the philosophical debates on applied ethics, personal identity and the mind-body problem (Letters, 19 August). Fourteen lawyers were in court for the Gracie and Rosie Attard case in September 2000; all recognised they were engaged in a philosophical as well as a legal debate. Lord Justice Walker’s judgment included a list of academic authorities on the ethics of euthanasia. A section of Lord Justice Brooke’s judgment was headed ‘Necessity: modern academic writers’. But these authors were of limited help because the choice the court faced was different in character from the situations they had discussed. I gather, however, that the philosophical literature has moved on since 2000, particularly with the publication of Jeff McMahan’s The Ethics of Killing, reviewed by Mulhall in the LRB (LRB, 22 August 2002).

As for personal identity, there are plenty of philosophical discussions of ‘brain-division, brain transplantation, bodily fusion and so on’, but none, I think, of conjoined twins. Finally, I said nothing about the mind-body problem. I did say that the position of Gracie and Rosie was comparable to that of passengers in an overcrowded lifeboat, in that one could survive only at the expense of the other, yet both had a right to be in the boat. Arguments about this sort of situation go back to Cicero. Mulhall moves from my word ‘belongs’ to his word ‘possession’, and, via a pun, to ‘possessed or haunted’, conjuring up a Cartesian ‘ghost in this physiological machine’. So he convicts me of holding a sub-Cartesian position on the mind-body problem when I said nothing at all about the relationship between minds and bodies: what I discussed was the situation of two people who were completely dependent on a resource (oxygenated blood) of which the supply was insufficient, and what I argued was that the blood couldn’t properly be said to ‘belong’ to just one of them. Mulhall claims that my ignorance ‘nourishes fantasy’, but here at least he is arguing with a fantasy figure, not with me.

David Wootton
University of York

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences