In response to:

Matters of Life and Death from the May 30, 1985 issue

To the Editors:

The central pro-life argument is simpler and more forceful than Jonathan Glover recognizes [NYR, May 30, 1985]. It begins with a proposition accepted on both sides of the abortion debate: healthy infants have a powerful claim to our care and protection, and we wrong them seriously if we kill them. The second premiss is that there are no morally relevant differences between infants and fetuses. A difference is morally relevant when it justifies a difference in treatment. Is being unborn the difference that deprives the fetus of a right to life? Abortionists sometimes dismember a late-term fetus in the womb to preclude live birth. We shall have to say this isn’t seriously wrong—after all, the fetus isn’t born—but it would be wicked to dismember her immediately after birth.

What about viability? Before five and onehalf months the fetus cannot survive outside the woman’s womb. But that greater dependence comes to this: the fetus needs her care and protection more. How can his greater need for care and protection deprive him of a claim to care and protection? But the fetus isn’t sentient before nine weeks. Does temporary unconsciousness deprive a man of human rights? Suppose a baby is born in a coma which began at conception. Doctors determine that the coma will lift in a few weeks and the child will develop normally. Does she have less of a right to life than other babies? What about human form? The fetus doesn’t even look human for the first eight weeks. But surely fundamental human rights do not depend upon the way your body looks: that is what is the matter with racism and sexism. Why reinstitute that sort of thinking here?

The differences between infants before and after they are born turn out to be like kinky hair and brown skin: they cannot justify a difference in treatment. It follows that fetuses have the powerful claim to care and protection enjoyed by infants. To avoid this conclusion one must deny a premiss. Glover denies that healthy infants have a claim to our care and protection and that we wrong them if we kill them. This is the price of a coherent prochoice position: we must roll back the rights of the born in order to justify killing the unborn. Glover acknowledges that the denial of the rights of babies seems “repulsive.” He tries to make it palatable by arguing that killing babies would still be wrong, not because it would be bad for babies but because it would be bad for us. Many people care about babies and would be upset if we killed them; and if we took to using spare babies for medical research, this practice—while innocent in itself—might lead to Nazi-style medical experiments on people who do matter.

Surely this is more repulsive. Better to embrace infanticide outright than to condemn it, not for the sake of infants, but because of its bad effects on us. One is reminded of Kant’s claim that torturing a dog is indeed evil—because it might make me hard in my dealings with men. It seems heartless and paradoxical to denounce a lethal practice because of the pain it causes those who share the nearly universal concern for its victims, while denying the victims matter. And Glover is committed to the view that if our concern for infants slackened—and what, better way to weaken it than to maintain that infanticide is wrong only because it might be bad for us—and we had good reason to believe we wouldn’t act like Nazis, then there would be nothing the matter with experimenting on anaesthetized infants. Certainly we wouldn’t have to worry about wronging them. And if new developments enabled us to meet vital adult needs by, say, killing spare infants and giving their organs to dying adults, then we ought to do it, so long as we produce more good than harm for those who matter.

We would be naive to assume these consequences never will become actual if the view that infants have no right to life is promulgated. Disenfranchise the powerless in theory and soon they will be disenfranchised in practice. Millions have been murdered in this century because they were supposed to lack human rights. Now, when told a segment of humanity doesn’t have rights after all, surely it is our duty to ask “Exactly why not?” And we ought to reject any answer that is less than conclusive. Our feelings for infants involve the recognition that we wrong them if we kill them. This must be dismissed as a moral hallucination. Why exactly? Because, Glover tells us, infants aren’t persons.

What are persons? Glover never tells us what he takes the word “person” to mean. What is he talking about? Are persons souls or Cartesian egos? Shall we deny infants rights because they fail to be we know not what? Does Glover take “person” to mean “entity with a right to life”? Then the claim that infants lack a right to life because they aren’t persons is circular. Does “person” mean “entity that is counted as a full-fledged member of the moral community”? Then infants certainly are persons, because they are so counted.


According to Glover, “being a person is bound up with a minimal level of self-consciousness, an awareness of a frontier between yourself and the rest of the world and awareness of yourself as having continuous existence over time.” This might be informative if we knew what persons were. Perhaps Glover is saying that self-consciousness is necessary for having human rights. This is implausible. Victims of severe strokes sometimes lose the neural capacity for self-awareness for several months; they behave like infants. Do they lose the right to life too? And if a human animal needn’t manifest present self-awareness to have human rights, why isn’t it enough that she will be self-aware, if we don’t kill her?

Perhaps Glover believes an infant doesn’t benefit from continued life unless she actually desires to go on living. This would have the implausible consequence that killing an infant painlessly is no harm. Why isn’t the fact that she is busy becoming self-aware on account of her biological nature sufficient to ground an interest in continued life? She will have the good which it is her nature to make for herself if she isn’t killed. Certainly we believe that infants have a powerful interest in growing up and living long and happy lives, and that we rob them of a considerable benefit if we kill them. Why shouldn’t this ground a claim to our protection?

Glover admits that killing an infant involves the loss of the years of life she would otherwise have enjoyed. But this argument doesn’t make infanticide objectionable, Glover says, because it makes contraception as bad as infanticide. He writes, “All eliminate a lifetime that would otherwise have been enjoyed.” This is to misconstrue the argument. Infanticide is objectionable because it deprives the infant of the years she would have enjoyed, and this is a harm. We do not harm the infant in merely preventing her from being conceived: she cannot be deprived of anything if she doesn’t exist.

The moral claim of infants is more obvious and forceful than any arguments to the contrary. To discount that claim would involve a terrible failure of sense and humanity. But then, sooner or later, we must oppose abortion, for the differences between infants before and after they are born are like the differences between men and women, whites and blacks, Aryans and Jews. We shall have to come to terms with the fact that we have failed to extend protection to fetuses because they were invisible, because they were powerless, and because it was convenient. If infants have human rights, the 1.5 million abortions each year represent an appalling injustice. Shall we deny the claims of the born in order to kill the unborn? In matters of life and death, if not philosophy, argument must end somewhere.

Jim Stone

University of New Orleans

New Orleans, Louisiana

Jonathan Glover replies:

Professor Stone’s powerful and reasoned comments challenge just those points of my argument that are most open to question. And Professor Stone’s version of the prolife argument is indeed simple and forceful. His first premise is that “healthy infants have a powerful claim to our care and protection, and we wrong them seriously if we kill them.” This has obvious intuitive plausibility. And to deny his second premise, that “there are no morally relevant differences between infants and fetuses,” seems to require good reasons for drawing a line, and Professor Stone is rightly skeptical about the standard attempts to do this. The two premises, if accepted, entail the prolife conclusion. It is a pleasure to have elicited this powerful reply. And yet I am still not convinced.

Consider the second premise, that there are no morally relevant differences between infants and fetuses. This is used to extend the protection we give to infants all the way back to conception. The first thing that should make us uneasy about this strategy is that it works just as well in the other direction. To many people, their belief that taking the morning-after pill is not murdering someone has as much intuitive force as the belief that infants have a right to life. If infants and newly fertilized eggs do not differ in morally relevant ways, accepting either of these beliefs commits us to denying the other. We can slide smoothly from platitude to paradox either in one direction with Professor Stone, or in the other direction with some of his opponents. Perhaps the premise which allows this needs more careful scrutiny.


The apparently simple premise, that there are no morally relevant differences between infants and fetuses, blurs an important distinction between what can be called external and internal differences. Someone’s external characteristics are constituted by their relationships with others, while internal characteristics are independent of those relationships. The distinction is not always sharp, but an example may make it clear. Suppose there is a fire, and I can save some, but not all, of those threatened. If I save my own children in preference to other children, some very severe equal rights theorist might think I have shown unjustifiable discrimination. If only internal features are relevant, other children had an equal claim. But if external features count, the fact that they are my children may be morally relevant. The plausible-seeming claim that infants and fetuses do not differ in morally relevant ways depends on the difficulty of citing relevant internal differences. It assumes that our relationship with an infant is irrelevant. This assumption should be questioned.

We also need not accept that a newly fertilized egg is morally of the same standing as a late fetus. Where differences are of degree, difficulty in justifying sharp lines proves less than is often supposed. There is no sharp line between some leaves and a heap of leaves, but a hundred leaves piled on top of each other are a heap and one leaf is not.

Now consider Professor Stone’s other premise: “Healthy infants have a powerful claim to our care and protection, and we wrong them seriously if we kill them.” This premise appeals to intuitions almost everyone shares, and at the same time implies a theoretical view about their basis.

The element of debatable theory in this apparently innocuous premise can be brought out by mentioning one way in which Professor Stone misunderstands the things said in my review about the moral claims of infants. He says, “Glover denies that healthy infants have a claim to our care and protection and that we wrong them if we kill them.” Of course healthy infants have a claim to our care and protection. As I said in the review, killing a healthy baby is unthinkable, and “we are right to recoil from any position that removes moral protection from the lives of babies.” The difference between our two positions is not about whether this moral protection should be maintained, but over what the reasons are for it. Denying the grounding given by prolifers for this claim about moral protection is not the same as denying the claim itself. (I do not want to slur over, however, the real difference between us about keeping alive some very severely handicapped babies.)

Professor Stone thinks that the objections to killing a healthy baby must depend on some internal feature of the baby, while I think they are bound up with our relationships with the baby. Our two positions agree that an infant is and should be accepted as a “full-fledged member of the moral community.” We differ in that Professor Stone thinks this is based on features which independently give them moral claims, while I think these claims are generated by relationships with other members of the moral community.

Professor Stone thinks that this second view has the consequence that, if our concern for infants slackened, and we also had good reason to think that we would not act like Nazis, “then there would be nothing the matter with experimenting on anaesthetized infants.”

This argument makes legitimate use of the philosophical technique of thought experiments: casting doubt on a principle by showing that, under different circumstances from those that do prevail, it would have unacceptable consequences. But it is worth noting just how different things are in this case. We have to imagine people utterly callous toward babies, but whose emotions are so compartmentalized that this does not in any way affect their feelings toward older children or adults. And presumably this coldness toward babies is supposed to leave their emotional development unharmed. These people seem more like Martians than like us. It is far from obvious that a principle should be thrown out because, when combined with a repellent Martian psychology, it generates conclusions we find repellent.

I said that our developing relationship with a baby starts at birth: “The emotions of childbirth and of the early days and weeks of parenthood are utterly incompatible with regarding babies as disposable living material. Most of us are incapable of adopting such an attitude. And even if we could adopt it, the world would be a worse place if we did so.” If we were Martians, we might have to choose between Professor Stone’s approach and infant experimentation. On the alternative approach, the important thing is not to become like Martians.

Professor Stone is wrong in thinking that I attach importance to the claim that infants are not persons. I put no weight on this, saying in my review that “perhaps there is not enough agreement on the use of the word ‘person’ for it to be clear that any of the rival views on this issue of definition are simply mistaken.”

Professor Stone and I agree that the moral protection a healthy baby now has should be preserved. To some, our disagreement over whether this is to be explained by citing internal or external characteristics will seem to be metaphysical hairsplitting. But two of the distinguishing features of Professor Stone’s view are of real importance. Making relationships irrelevant gives plausibility to the claim that infants and fetuses have no morally significant differences. And his view has the problem of citing the relevant internal feature. What is the internal characteristic that gives a newly fertilized human egg a stronger claim to life than that of an oyster? (It is one of the problems of using the “no relevant difference” argument that the relevant feature of infants has to be present as far back as the argument is pushed.)

The difficulty of finding a relevant internal characteristic already possessed by fetuses is often seen by supporters of the prolife view. They sometimes move away from basing fetal rights on what a fetus is now like. They cite its potential to become a developed person. I argued that this too has problems. If what is wrong with abortion is the loss of the developed person there would otherwise have been, the argument proves too much. That argument rules out contraception, and chastity too.

Professor Stone offers a different version of the argument. He thinks that abortion and infanticide are objectionable because they deprive fetuses of the years they would have enjoyed. But, “we do not harm the infant in merely preventing her from being conceived: she cannot be deprived of anything if she does not exist.” The underlying principle here may be that you only do wrong if your act harms someone who exists. This sounds plausible, but on closer examination things again turn out to be more complicated.

To illustrate this, I will adapt a thought experiment devised by Derek Parfit. Suppose, as may be true, that there are biological mechanisms favoring the conception of normal babies over those that will be abnormal. Imagine a factory polluting the atmosphere with a chemical which, in the case of some kinds of handicap, reverses such a mechanism so that it favors the conception of handicapped children. And suppose that the handicap, although severe, does not make life not worth living. (It might be congenital blindness.) The pollution causes the birth of blind children rather than normal ones. It has not made those children worse off than they would have been, since without it they would not have existed. The blindness may distress other members of their family, but it is perverse to object to the pollution solely on these grounds. (That would be like Kant’s repellent views on cruelty to animals.) It may seem obvious that starting a process resulting in more handicapped children does some harm. But the question arises: to whom is this harm done? The handicapped children are not worse off than they would have been, so it is hard to see them as harmed. Although the world is a worse place because of the pollution, the case that any particular person has been harmed by it is hard to sustain.

This thought experiment might seem loaded because it involves an act (introducing the pollutant) which interferes with normal biological mechanisms. But contraception also involves this. And any objection to the thought experiment on those grounds would need a defense of the presumption against interfering with natural processes, which is not easy to provide.

The plausible-sounding principle, that to do wrong you have to harm someone who exists, separates abortion from contraception, but at the cost of making it hard to object to the pollution except by citing distress to the children’s families. The principle deals less well than at first appears with these complex issues about the ethics of conception.

A possible view of the pollution case is that creating these handicapped children is wrong although no person is harmed by this policy. But if wrongs need not involve harm to individuals, it is unclear how abortion is shown to differ morally from contraception by saying that someone is harmed by it.

There is an underlying difference between Professor Stone and me about how to think about moral issues. I have stressed the complexity of the abortion issue, and the need to submit our intuitive responses to analysis before accepting them as final. Professor Stone resists going too far from our first responses: “In matters of life and death, if not philosophy, argument must end somewhere.” His preference is rooted in something important. In the arguments I have given, the thought experiment may seem artificial and the distinctions scholastic. There is a danger of abstract theorizing weakening the human responses that are the source of any living morality. And I share, and hope others do, his protective responses to babies. But I also share, and hope others do, some of the prochoice responses to the idea of forcing women to continue with unwanted pregancies. All argument must end somewhere. But when we, as a society or as individuals, are torn between deep but conflicting responses, perhaps we should not stop exploring the arguments too soon.

This Issue

February 13, 1986