In his paper “Why Abortion is Immoral”, Don Marquis urges a reorientation of the abortion debate. Metaphysical issues concerning the personhood of fetuses, which previously held center stage in philosophical discussions of the morality of abortion, have proven to be intractable, he says. As long as the debate is left to rest on such issues, it will be at an impasse. In his view, we would do better to focus on a different question: does abortion cause harm, in particular, harm to the fetus? Once the debate is framed in these terms, he argues, it is clear that abortion is prima facie morally wrong, since abortion surely does harm the fetus, at least in the “standard” case.
But Marquis commits a crucial error. Although Marquis does not say so explicitly, he must assume that the question of harm is independent of the issue of personhood; otherwise, he could not hold that the latter was any more tractable than the former. As I shall argue here, however, that assumption is mistaken. We cannot get away from metaphysics so easily. My aim here is to show that, given certain widely held metaphysical views, including views about personal identity, his argument for the immorality of abortion rests on highly implausible premises. Marquis could, of course, reject the metaphysics. But that would be to concede that issues of personhood are of central importance, after all.
The chief effect of abortion on the fetus is death. So, we must ask, what is the harm of death for the fetus? Marquis suggests that we get a better handle on that question by first considering the harm of death for “us”, by which I presume he means normal adult humans. In this case, he follows other philosophers (e.g. Nagel) in holding that the harm of death is deprivation. An adult human is harmed by death just to the extent that she is deprived of benefits — happy experiences, completed projects etc — which she would otherwise have enjoyed. As Marquis puts it, the harm of death is the deprivation of a “valuable future”.
But if death harms adults by depriving them of a valuable future, Marquis argues, then surely it must (in the “standard” case) harm fetuses at least as much, and probably even more so, since they have a longer future to be deprived of. Marquis writes:
The future of a standard fetus includes a set of experiences, projects, activities, and such which are identical with the futures of adult human beings and are identical with the futures of young children.
Now, this claim is, I think, false. The future of a fetus, even a non-aborted one, does not extend beyond the time of birth. Thus, if abortion “deprives” the fetus of any future at all, then it is a very short future (no more than nine months or so) and one that contains none of the benefits that Marquis mentions. Perhaps it will be argued that a fetus may continue beyond birth, but in the form of a baby (and then a child, then a teenager, and so on). But that makes no sense. How could a fetus be a baby? These are just different things; one cannot be the other.
So, by Marquis’s own account of the harm of death, abortion does not harm the fetus. But perhaps it harms something else. There is another entity whose future is more significantly curtailed by abortion: the human being of which the fetus is a (temporal) part. In the typical case, a human being begins as an embryo, then later becomes a fetus, and then a baby, and so on until it is an adult human. In “four-dimensionalist” terms, the human being has all these things — the embryo, the fetus, etc — as temporal parts. Thus, abortion might prevent the human being from having quite a lengthy future (typically 80 years or so). But is this future a valuable one? Should we say that abortion deprives the human being of benefits that it would otherwise have received?
The answer, I claim, depends on personal identity. To illustrate, let me first briefly sketch a view of personal identity that I tend to favour. On this view, a person is not identical to a human being (at least not typically). It is often the case that a person will overlap with a human being, in the sense that they share temporal parts. But in such cases, the human being usually has parts — an embryo, a fetus, probably a baby, perhaps more — that are not shared by the person. On this view, a fetus cannot be part of a person, because a person’s parts are unified by psychological continutity, and a fetus lacks a sufficiently rich psychology to be continuous with the other parts. A fetus can, however, be part of a human being, because a human being’s parts are unified by something else (genetic continuity, I guess).
Now, to see how this account of personal identity bears on the question whether a human being can be benefited, consider the following example. Recently I received news that my PhD had been approved. This was good news indeed. I had completed an important project. Surely this is a great benefit to me (the person). But should we say that the human being with which I overlap also benefited? No, that would be double-counting. Given a common metaphysical doctrine, so-called “unrestricted composition”, there are infinitely many things that have me as a part. If we say that the human being benefited, then it seems we should say that all these infinitely many things benefited, too. But that’s absurd. People receive benefits; human beings do not.
It follows that abortion does not harm the human being whose future is thereby cut short. The human being would not have received any benefits anyway. So the abortion does not deprive it of any benefits. On Marquis’s account of the harm of death, then, the human being is not harmed.
On a different account of personal identity, this conclusion need not follow. Suppose we said, for example, that every human being is identical to a person. (I guess this is the view that is sometimes called “animalism”). Then it would not be so implausible that human beings receive benefits. Perhaps some would try to save Marquis’s argument by appealling to such a view as this. But that would be to abandon his goal of moving the abortion debate away from issues of personhood.