Why the deprivation argument fails

In his 1989 paper, “Why Abortion is immoral”, philosopher Don Marquis argues that abortion is generally wrong because killing fetuses deprives them of a future with enjoyments, relationships, and projects. As Marquis puts it, abortion deprives fetuses of a “future-like-ours.” I will argue that Marquis's argument is unsound, in that having a future-like-ours is not sufficient for having full moral worth (i.e. the moral worth of a person).  

Marquis’s explanation for the wrongness of abortion is initially quite plausible and potentially has a wide scope. It can also be used to explain common moral intuitions about the permissibility of euthanasia (under certain circumstances) and the wrongness of killing. When an innocent person is killed, there are typically several sources of harm. The harm done to the person’s family and friends, psychological harm done to the killer (e.g. making them even more vicious), and the physical harm done to the person killed. But, as Marquis argues, the worst part of the wrongdoing is not from the psychological and physical harms. Killing ends personal and romantic relationships, terminates any long-term projects or life goals, and denies the person of ever experiencing pleasure again. It is the deprivation of a future that makes killing seriously wrong.

Marquis’s account of the wrongness of killing explains not only why killing innocent human adults is wrong, but also why infanticide is seriously wrong. And since there aren’t any morally relevant differences between babies and (late-term) fetuses, Marquis concludes that abortion is typically “in the same moral category as killing an innocent adult” (Marquis 183). After all, if you kill the fetus, you deprive it of a future-like-ours, just as in the case of infanticide or the murder of an adult.  

I accept that depriving persons of the goods that life brings is seriously immoral. I also accept that late-term abortion deprives a potential person of a future-like-ours. But I reject the conclusion that abortion is equivalent to murdering a person. I hold this position because there is a morally relevant difference between the killing of persons and potential persons that Marquis does not consider. Actual persons also have a past-like-ours, whereas potential persons do not.

By past-like-ours, I mean a series of connected psychological states that involve episodic memories, experiences of happiness and pleasure, and the actual formation of long-term life-goals and close relationships. To understand why having a past-like-ours is morally relevant, consider the following case.

Imagine that in the near future, scientists are able to create sentient AI. Once they are fully developed, the robots are able to have conscious experiences just like humans. They can experience the same range of complex emotions, fluently speak and understand human languages, enjoy fine-dining and music, and can even contemplate their own existence. These robots are full-blown persons with a future-like-ours.*

In order for the robots to have these psychological capacities, there is a developmental period, much like fetuses go through. You can think of this as a “buffering period” for personhood. In a world like this, it seems reasonable to suppose that humans who decide to turn on one of these robots might end up changing their mind. So, on occasion, humans decide to press a “cancellation button” that is found inside the robot’s computerized brain before it reaches full-blown personhood. If pressed, everything in the robot’s computerized brain is wiped clean, which results in the death of the potential person. If humans decide to restart the process, an entirely new potential person would be generated (it wouldn’t be a clone or a recreation of the one originally terminated). Would it be seriously wrong for someone press the cancellation button during the buffering period? 

According to Marquis's view, it would be seriously wrong to do so, given that it would deprive a potential person of a future. Moreover, it would be just as wrong to push the button as it would be to kill an innocent adult human. But according to my own intuitions--which may be widely shared--this result is extremely counterintuitive. Thus, there is a conflict between Marquis's account and common-sense morality. I think a plausible explanation for why it’s morally permissible is because the robot did not reach the point where it was fully up-and-running. The robot did not have a past-like-ours. If this is right, Marquis's account is, at best, incomplete. 

The personhood robot thought experiment is analogous to late-term abortion in that you are depriving potential persons of having a future-like-ours while they are still in the "buffering" stage of development. What is the morally relevant difference between pressing the cancellation button and abortion? One difference is that one is a manmade machine, the other is a biological organism. But that doesn’t seem to be difference relevant to morality. The following claim seems plausible: all potential persons intrinsically have equal moral status, regardless of their species or physical constitution. Not only does it seem plausible, I am aware of no reasonble grounds for doubting it is true. 

One might think that the consequences of pushing the button and abortion are very different, in that abortion may bring about certain harms to others, that will not result from pushing the button. But any appeals to harms done to family members or to human society at large will no longer be talking about intrinsic moral worth. We want to know, what is seriously wrong about pushing the button, in itself. 

If you think that abortion is seriously wrong, just because it deprives someone of a future-like-ours, then pressing the cancellation is also wrong, equally so. Marquis seems to accept the antecedent. He states that, “having that future by itself is sufficient to create the strong presumption that the killing is seriously wrong [emphasis, mine]” (Marquis, pg. 195).

Logical consistency requires us to either think that pressing the cancellation button is as seriously wrong as killing an innocent adult human, or that we deny that both abortion and pressing the button are seriously wrong. I think the latter option is much more reasonable. However, it is argued that the implications of accepting that abortion is not seriously wrong has disturbing consequences. It implies that infanticide is not seriously immoral either. Here, the moral intuitions for infanticide are much stronger, and it may just seem obviously wrong to kill newborns. Although these strong moral intuitions may be widely shared, it matters whether they have a rational basis.

I suspect that resistance to this conclusion—that abortion and infanticide are not seriously wrong—is largely due to emotional reasons and cultural biases. For example, late-term fetuses and newborn babies are cute. We experience negative emotions when we think of cute babies dying, but don’t feel much of anything when a robot or something ugly dies. There are evolutionary reasons that explain why we feel this way, and why it leads us to take good care of cute creatures. But this evolutionary explanation, by itself, does not provide a sound moral basis for our intuition.

Why else might explain our moral intuitions about infanticide? A plausible answer can be found by looking at historical and anthropological work on infanticide. As philosopher Michael Tooley observes, attitudes about infanticide have not always been widely shared. In fact, these attitudes about have only been around for a few hundred years (Tooley 1984). Throughout most of human history, infanticide was considered morally permissible. Why might this be? It is unlikely that the shift in attitudes was informed by science or the discovery of some new facts about infants.  Tooley argues that a much more plausible answer is that our attitudes about infanticide were inherited from a religious tradition (e.g. Christianity) that values all human life, or posits the existence of an immortal soul to all humans. Christians will likely argue that this is a point in favor of their worldview, and that on atheism, you are left with the disturbing conclusion that babies have no moral worth. But if we reject the legitimacy of the religious tradition, or the existence of the soul, there are seemingly no other legitimate grounds to appeal to. In that case, we should not be misled by our intuitions. The presence of cultural biases and negative emotions are not good reasons for holding moral beliefs.

In conclusion, having a future-like-ours is not sufficient for full moral worth. Persons not only have a future-like-ours, they also have a past-like-ours. It is a combination of these two features that make it seriously wrong to kill someone. Potential persons, such as fetuses and infants, do not have a past-like-ours, in that have not experienced pleasure, established any life projects, goals, or relationships. Thus, the killing of a potential person is not morally equivalent to killing of an actual person.

* My thought experiment is essentially a variation of Michael Tooley's thought experiment involving cat people (Tooley 1972), an argument I have written about elsewhere. 

Works cited

Marquis, D. (1989). Why abortion is immoral. The Journal of Philosophy, 86(4), 183-202.

Tooley, M. (1972). Abortion and infanticide. Philosophy & Public Affairs, 37-65.

Tooley, M. (1984). Abortion and Infanticide. Blackwell Publishing Ltd.