Article Text

Download PDFPDF

Abortion, infanticide and allowing babies to die, 40 years on
  1. Julian Savulescu
  1. Correspondence to Professor Julian  Savulescu, Faculty of Philosophy, The Oxford Uehiro Centre for Practical Ethics, Suite 8, Littlegate House, Oxford OX1 1PT, UK;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

In January 2012, the Journal of Medical Ethics published online Giubilini and Minerva's paper, ‘After-birth abortion. Why should the baby live?’.1 The Journal publishes articles based on the quality of their argument, their contribution to the existing literature, and relevance to current medicine. This article met those criteria. It created unprecedented global outrage for a paper published in an academic medical ethics journal. In this special issue of the Journal, Giubilini and Minerva's paper comes to print along with 31 articles from some of the best scholars in the world, from the broadest range of perspectives on abortion and infanticide, including those strongly critical of Giubilini and Minerva.

The killing of a baby is among the most shocking of human practices. I am strongly opposed to the legalisation of infanticide along the lines discussed by Giubilini and Minerva. But I would like to explain why a journal of medical ethics published an article examining infanticide and now devotes a special issue to bringing a wide range of perspectives for further examination of these issues.

Infanticide has been practised throughout human history for various reasons and continues to be practised today throughout the world. In rare cases, infanticide can be practised openly and legally—for example, as specified under the ‘Groningen Protocol’, in the Netherlands. The Groningen Protocol allows doctors to end the life of neonates at the request of their parents if the infant is experiencing hopeless and unbearable suffering.2

But medical infanticide arguably occurs elsewhere other than in the Netherlands today.

In neonatal intensive care, it is not uncommon for a doctor to withdraw artificial ventilation with the consequence that a baby dies.3 If anyone else performed that act, including the parents of the child, it would be viewed as an act of homicide, in some cases murder. But when a doctor performs that act, it is not murder. The active withdrawal of life-prolonging medical care (an intentional act that kills, even if not necessarily with the intention to kill) is a standard part of medical practice in relation to people who experience severe disability and suffering, including newborns. This is the case in the USA, UK, the rest of Europe and nearly all of the world. It is sometimes called ‘selective non-treatment.’

Robert Young argues that the reason active withdrawal of medical care is not classified as murder in England is that a legal judgement in the case of Tony Bland redefined the act of withdrawing a life-prolonging medical treatment as an ‘omission’. If the act of withdrawing a life-preserving feeding tube (or other medical treatment) had been properly described as an act, then it would constitute intentional killing and would have been murder. Young argues this was a legal redescription to protect doctors from charges of murder.4 But withdrawal of care was practised long before Bland and not prosecuted.

And, as I have said, anyone else who acted to withdraw life-prolonging medical treatment who was not a doctor (or suitably qualified medical professional) would be judged to have acted in such a way as to kill, not to be ‘omitting to save life’ or letting die. The very same act is described in law as an ‘act’ or an ‘omission’, depending on whether or not a doctor performs it. Whether it is right or wrong might depend on whether a doctor performs it, but the nature of some physical event cannot logically depend on the identity of the person involved. A parallel is that ‘running’ is a certain kind of action of the body, whether or not a doctor or someone else is performing it.

Jeff McMahan has argued that withdrawal of medical treatment is ‘letting die’, not ‘killing’. He gives the analogy of a person, call him Jim, swimming out to rescue a drowning person. If Jim is dragging the drowning person to shore but changes his mind and releases him, he lets that person die, but does not kill him.5 However, imagine that Jim sets out in his boat and takes the drowning person into it. While on the way to shore, he changes his mind and throws the person overboard, back into the sea. That would, in my view, be killing. It is like taking a respirator away. (Of course, it might be justified to throw the person overboard—the person might have a lethal contagious disease and be about to die soon of that anyway.)

McMahan and some others (eg, Kamm6) argue that removing aid that one is oneself providing is stopping saving, not killing. But in some cases, such as when Jim throws the drowning person back out of his boat, refusing to continue saving a person is morally just as bad as killing him. The same is true for these philosophers in some comparisons between selective non-treatment of infants and infanticide.

I am not here intending to suggest that withdrawal of life-sustaining medical treatment is unethical or should be illegal. What I am pointing out is that actions to end the life of newborns are widely practised and legally sanctioned—although not called infanticide. Also, there is some reason to think that these decisions are made on a different basis from those made for older children, and that parents are given a broader discretion in such decisions. If that is the case, it appears that the societal antipathy to infanticide is not as strong as it might seem at first glance.

Another example of what is arguably legal infanticide is the lethal separation of conjoined twins. In a widely publicised case, in 2000, the twin daughters of the Maltese couple Michaelangelo and Rina Attard, Jodie (real name Gracie) and Mary (Rose), were joined at the pelvis with a fused spine. The facts of this case are described by Lord Justice Ward in his judgement: Jodie and Mary are conjoined twins. They each have their own brain, heart and lungs and other vital organs and they each have arms and legs. They are joined at the lower abdomen. … (T)hey can be successfully separated. But the operation will kill the weaker twin, Mary. That is because her lungs and heart are too deficient to oxygenate and pump blood through her body. Had she been born a singleton, she would not have been viable and resuscitation would have been abandoned. She would have died shortly after her birth. She is alive only because a common artery enables her sister, who is stronger, to circulate life sustaining oxygenated blood for both of them. Separation would require the clamping and then the severing of that common artery. Within minutes of doing so Mary will die. Yet if the operation does not take place, both will die within three to six months, or perhaps a little longer, because Jodie's heart will eventually fail.7

The High Court authorised surgical separation, knowing that this would kill Mary. The legal justification for this is striking. Mary may have a right to life, but she has little right to be alive. She is alive because and only because, to put it bluntly, but nonetheless accurately, she sucks the lifeblood of Jodie and she sucks the lifeblood out of Jodie. She will survive only so long as Jodie survives. Jodie will not survive long because constitutionally she will not be able to cope. Mary's parasitic living will be the cause of Jodie's ceasing to live. If Jodie could speak, she would surely protest, ‘Stop it, Mary, you're killing me’. Mary would have no answer to that. Into my scales of fairness and justice between the children goes the fact that nobody but the doctors can help Jodie. Mary is beyond help. Hence I am in no doubt at all that the scales come down heavily in Jodie’s favour. The best interests of the twins is to give the chance of life to the child whose actual bodily condition is capable of accepting the chance to her advantage even if that has to be at the cost of the sacrifice of the life which is so unnaturally supported. I am wholly satisfied that the least detrimental choice, balancing the interests of Mary against Jodie and Jodie against Mary, is to permit the operation to be performed.8

This was, in my view, the correct judgement and separation was in the interests of Jodie, much more so than it was against the interests of Mary. But it is also my view that it is a convenient fiction to describe Mary as a parasite using Jodie's body as a ‘life-support machine,’ as neonatal surgeon Adrian Bianchi reportedly said.9

Conjoined twins arise when a single embryo fails to completely divide into identical twins. Both Mary and Jodie arose from the same single embryo. Neither has a greater claim on the shared organs. Thus, it is my own view that it is not ‘accurate’ to describe Mary as killing Jodie. They were both dying because their anatomy and physiology, as it happened to develop, could not support both of their lives. The common artery no more belonged to Jodie than to Mary. There were two brains, and so two minds, sharing vital parts. Mary was killed, subjected to infanticide, to save Jodie. Of course, Lord Justice Ward did not accept this view, seeing one twin, Mary, as an aggressor against whom her sister needed to be protected by the Court.

Another medical practice and example of selective non-treatment that results in the death of patients is the withholding of life-sustaining medical treatment. Recent articles10 ,11 and newspaper coverage12 draw attention to the practice of not feeding severely disabled newborns. Arguably, this is morally equivalent to infanticide which, if it occurred outside of medical practice, might be referred to as ‘starving a baby to death’ for this reason. If a mother neglected to feed her disabled baby, she would usually be guilty of a criminal act. The medical practices of ‘allowing newborns to die’ are the topics of ongoing ethical debate but are a part of common medical practice and have been for several decades.2 ,4 ,13 Discussions of infanticide should be contextualised in those practices that end life, which society already accepts, even if they are euphemistically redescribed.

There is a vigorous body of literature on whether acts resulting in death (active euthanasia) are morally equivalent to omissions resulting in death (passive euthanasia). Some philosophers have argued that there is no moral difference,14 while others continue to dispute this. Philippa Foot famously argued that to fail to send a food parcel to a starving child is not morally equivalent to sending a poisoned food parcel.15 This is a long-standing debate but does not need to be resolved to recognise that human beings’ (including infants’) lives are intentionally shortened in our society. For even if omissions were morally different from actions, and so withholding life-sustaining treatment morally different from killing, the withdrawing of medical treatment is an intentional action,4 ,16 though this is denied by some.6 ,7 Withdrawing medical treatment might be like a person acting to intercept a food parcel sent to a starving child, intentionally causing the death of that child. It is an intentional act that knowingly causes the death of the child. Even if we can ignore omissions resulting in death, it is important to ethically evaluate the practice of the withdrawal of medical treatment that results in death.

The ethical discussion of infanticide dates back several thousand years. At least 100 articles have been published on infanticide in the Journal of Medical Ethics over its rather shorter history, with articles both for and many against it. Some of the world's most famous living philosophers have written about its merits and justification over the last 40 years, including Michael Tooley,17 Jonathan Glover,18 Peter Singer,13 ,19 Jeff McMahan5 ,20 and John Harris.21 Four of these five have contributed to this issue. McMahan argues that the permissibility of infanticide in some circumstances is not only implied by certain theories, but by beliefs that are widely held and difficult to reject.22 Michael Tooley's book is entitled Abortion and infanticide. Peter Singer wrote a book in 1985 with Helga Kuhse called Should the baby live?

Over the last 40 years, there has been an active debate on the ethics of killing or allowing severely ill or disabled newborns to die. Jonathan Glover's landmark Causing death and saving lives notes that ‘Dr Francis Crick (the Nobel Laureate who discovered DNA with Jim Watson in 1956) once proposed a two-day period for detecting abnormalities, after which infanticide would not be permissible’.22

Such debates are a continuation of very ancient theological and philosophical debates on abortion, on whether there is a moral difference between the effects of acts and omissions, on ‘ordinary and extraordinary means’ and on ‘double effect’—whether there is a moral difference between the effects of actions that are intentional and those that are merely foreseen, but not intended.

In the case of abortion, termination of pregnancy is now permissible in many countries not only for severe disability but also for reasons of maternal welfare (or other reasons). Giubilini and Minerva extend the long-running debate on infanticide to ask: if abortion is permissible both for social as well as medical reasons, why is infanticide permissible only for medical reasons (assuming that selective non-treatment is a form of infanticide)? They ask: what is the moral difference between a fetus and a neonate? As McMahan points out in this issue, there is at least a 4-month period during which a human being could be either a fetus or an infant, depending on whether delivery has yet taken place.23 Giubilini and Minerva point out that both have similar capacities and if one is permissible, why not the other? The presence of disease or disability should not make a difference to moral status, so if infanticide is permissible for medical reasons, why is it not permissible for social reasons?

Their extension of the existing debate about infanticide from medical indications to social indications is relatively novel. I personally don't agree with the authors’ conclusion. (One obvious difference between early abortion and infanticide is that infanticide involves killing a human being who could be adopted or cared for by people other than the mother. Indeed, I have previously argued on this basis that we should encourage the donation of excess frozen embryos produced by IVF to non-genetically related parents rather than discarding them.24) The important debate is rather about the nature, permissibility and limits of euthanasia in society. But their arguments—based on the similar moral status of the fetus and neonate—call for rebuttal. (Of course their argument can cut both ways—I understand that one relatively common response amongst the online community was to assume the paper was written from a pro-life perspective as a reductio ad absurdum of the permissibility of abortion.)

Giubilini and Minerva's paper was handled by Associate Editor Professor Ken Boyd, a distinguished theologian and medical ethicist with over 10 years editorial experience with the Journal. It was scrutinised by three peer reviewers and revised in light of their comments. The Journal neither publishes nor declines to publish articles because of the controversial nature of their conclusion; rather, editorial decisions are made on the basis of the quality of the argument. If an argument were based on clearly mistaken premises or obvious failings in logic, then we would expect the review process to detect it and require its correction or reject the paper. But in this case, there has been a long and ongoing debate on the moral status of early human life—including neonates—and many protagonists in this debate have taken the same premises seriously, or even endorsed them, and have made similar logical inferences.

The Journal aims in this issue to promote further and more extensive rational debate concerning this controversial and important topic by providing a range of arguments from a variety of perspectives. We have tried to be as inclusive as possible and provided a double issue to include as many as possible of the submissions we received.

Infanticide is an important issue and one worthy of scholarly attention because it touches on an area of concern that few societies have had the courage to tackle honestly and openly: euthanasia. We hope that the papers in this issue will stimulate ethical reflection on practices of euthanasia that are occurring and its proper justification and limits.



  • Contributors Thanks to Ingmar Persson, Jeff McMahan, Peter Singer, Raanan Gillon, Guy Kahane, Imogen Goold, Tom Douglas and Dominic Wilkinson.

  • Competing interests None.

  • Provenance and peer review Commissioned; internally peer reviewed.

Other content recommended for you