1. Introduction
In his ‘The Counterfactual Argument Against Abortion’ (Reference Kulesa2023) Ryan Kulesa argues that it is prima facie wrong to kill a ‘counterfactual person’. Some early foetuses, though still lacking consciousness, are counterfactual persons. Hence, it is prima facie wrong to kill (abort) these foetuses. It does not follow that it is prima facie wrong to allow some embryos to die; hence there is no conflict with the widely shared intuition about embryo rescue cases (see e.g., Giubilini & Minerva Reference Giubilini and Minerva2013, Hassoun & Kriegel, Reference Hassoun and Kriegel2008, Liao Reference Liao2006, Stretton Reference Stretton2008, Hendricks Reference Hendricks2019, Räsänen Reference Räsänen2016) that it is permissible, indeed obligatory, in such cases to leave the many frozen embryos to die to save the one endangered child. On the other hand, the prima facie impermissibility of infanticide is secured. Kulesa thus claims that by appeal to the thesis that it is a sufficient condition of the prima facie wrongness of an action that it is the killing of a counterfactual person he can accommodate intuitions that philosophers writing about abortion find it otherwise hard to reconcile.
I shall argue that Kulesa fails to establish his goal because it is not a sufficient condition of an act being even prima facie wrong that is the killing of a counterfactual person. It is not, even prima facie, a sufficient condition of wrongdoing that one’s action kills a counterfactual person, as Kulesa defines this notion. Thus, while it may be that it is prima facie wrong to abort an early foetus which is a counterfactual person, more has to be said, other than the mere classification of it as a counterfactual person, to explain this.Footnote 1
2. Counterfactual persons and moral status
Kulesa’s crucial notion is that of a ‘counterfactual person’. As Kulesa explains, it is open to the reader how exactly ‘person’ is understood here; it is sufficient for his purpose that its definition excludes some early-stage foetuses and frozen embryos and also excludes familiar non-human animals such as cats and poultry. Imagine personhood so defined. Then x is a counterfactual person at a time if and only if: if x were not killed at that time x would have been a person later. A counterfactual person, Kulesa emphasises, is not merely a possible person. Something could be such that, if it were not killed at a time, it could have become a person thereafter, without it being true that, if it were not killed at that time, it would have become a person thereafter. The latter is Kulesa’s notion. This notion of a counterfactual person is to be understood in the familiar Lewisean way (Lewis Reference Lewis1973), in terms of what happens in the closest to the actual world in which the antecedent of the counterfactual is true, i.e., the entity is not killed.
Suppose then that I am contemplating killing a young child, who is in a state of unconsciousness. I may know, however, that, if I were not to kill her, she would gain consciousness later and would qualify as a person (even if she does not yet). That is, I may know that it is not the case that if I were to refrain from killing her, she still might not gain consciousness and later qualify as a person (because, say, there are other potential killers about). Then, if I go ahead and kill the child, she was, on Kulesa’s account, a counterfactual person, whether or not she was already a person. On the other hand, suppose I am contemplating killing a chicken. It will not be true that if I were to refrain the chicken would develop personhood, though I may not be able to exclude the possibility – the metaphysical possibility – that it might. This accords with our intuition that, even if neither the non-conscious child nor the chicken is a person, still it is massively more wrong to kill the child than the chicken (even if it is to some extent always wrong to kill any sentient being).
Kulesa is able to explain this intuition and is also to explain our intuition that it is also massively wrong to kill an adult human who is temporarily comatose, who would, however, subsequently regain consciousness if I were not to kill him, just as it is massively wrong to kill an adult human who is not massively comatose. Importantly for Kulesa, he is also able to explain, using his notion of a counterfactual person, why it can be permissible to leave to die the many frozen embryos in an embryo rescue case in order to save the one child. Kulesa describes the following sort of case. I have the choice of rescuing a young, unconscious, child or ten frozen embryos from a burning building. Intuitively it is no contest. I must rescue the child. Yet how can this be? None of the eleven possesses current consciousness. All eleven are at least potential persons. How can they differ relevantly in moral status? Kulesa’s answer is that the child is not merely a potential person, but (at least) a counterfactual person. If I were not to leave her to die, she would subsequently achieve personhood (even if she does not already possess it). But this is not true of the embryos. If they were rescued, they might develop personhood (they are potential persons) but given the actual realities about the storage and use of frozen embryos, it cannot be said that they would develop personhood if they were rescued. Hence, if they are not rescued, they are not now counterfactual persons, but the child is now a counterfactual person if it is not rescued.
But, Kulesa now explains, an early-stage foetus is typically different from the frozen embryos. If I am contemplating aborting an early foetus, I can know that, if I were to refrain, it would survive and develop personhood. So, if I do carry out the abortion it was a counterfactual person. This is what distinguishes it from the frozen embryos. Of course, this might not be the case. It might be the case that if I were not to perform the abortion someone else would have, or the foetus might have died from some other cause. In this case, according to Kulesa’s reasoning, his argument for the impermissibility of my aborting the foetus is inapplicable, since it is not a counterfactual person. But this is not the typical case when abortion is being considered.
Hence Kulesa says, his account explains the impermissibility of aborting the typical early foetus, but at the same time does not exclude the permissibility (or the obligatoriness) of leaving the many frozen embryos to die in the abortion rescue case. It also does not exclude the permissibility of In Vitro Fertilisation (IVF), in which many embryos are created, some unused. ‘So, these embryos are created with the knowledge that many of them will either die or remain frozen indefinitely… It is not true of them that, were they not killed, they would have been persons. As a result, it is not impermissible to create them knowing that many of them will either die or remain frozen indefinitely’ (Reference Kulesa2023: 7).
3. The challenge to Kulesa’s argument
I have skipped over details and only outlined Kulesa’ position. But I hope I have done enough to make it clear. I now want to argue that he is wrong. It is not the case that it is a sufficient condition of an act being wrong, even prima facie wrong, that it prevents something which would have become a person, if that act had not been performed, from becoming a person.
Imagine a situation of the magical or science-fictional type Kulesa borrows from Tooley (Reference Tooley1972). There is a magical pill which if given to a chicken will transform it so that over time it develops personhood (like a naturally developing foetus).
Now consider the following three scenarios.
Scenario One: The pill is about to be given to a chicken, but I can prevent this. If I do so the chicken will not develop personhood. It is true that, if I were not to prevent the pill being given, the chicken would develop personhood. So, if do prevent the pill being given it was a counterfactual person.
Scenario Two: I can prevent the pill being given but only by killing the chicken (thereby, let us suppose, ensuring the pill is given to another chicken). If I were not to kill the chicken, it would be given the pill and thereby develop personhood. If I do kill the chicken, it was a counterfactual person.
Scenario Three. The pill has been given. But there is the opportunity either to administer a second magical pill in time to block the action of the first, or to kill the chicken. If I do neither the chicken will be transformed and over time will develop personhood. So, if I do either, the chicken was a counterfactual person; it would have become a person if I did neither.
The following objection to Kulesa now appears to be valid.
I take it that I am not doing anything massively wrong in ensuring that the pill is not given to the chicken in the first scenario (I might ensure that the pill is not given to the chicken simply by giving the pill to another chicken instead). But in doing so I am preventing what is a counterfactual person from developing personhood. Nor am I doing anything massively wrong in Scenario Two in killing the chicken, at least, nothing worse than killing any chicken.
And, if so, it appears I am doing nothing massively wrong in killing the chicken, or giving it the second, blocking, magical pill, in Scenario Three, after the first magical pill has been given.
So, interfering to prevent a counterfactual person becoming a person is not a sufficient condition of doing something massively wrong (worse than killing a non-conscious, non-human animal). Hence, employing Kulesa’s notion of a counterfactual person does not explain the intuition to which he appeals, that killing an early-stage stage foetus is prima facie seriously wrong in a way that killing a non-human animal is not.
The obvious response to this objection is to say that it is only the killing of a subclass of counterfactual persons which is massively wrong: those which are such that the counterfactual in question is true in virtue of intrinsic, non-relational features of the individual (as is not the case in Scenario One and Scenario Two, which are disanalogous to the third scenario and in which, as a referee helpfully notes, the chicken is comparable to an unfertilised ovum rather than to an implanted and developing foetus). Or perhaps, those which are such that the counterfactual in question is true in virtue of the features of the individual which are both intrinsic and belong to it as a member of the natural kind of animal it is (as in the typical abortion scenario). Either modification of Kulesa’s proposal will distinguish the first two scenarios from the early-stage abortion case, and the second modification will distinguish all three scenarios from the early-stage abortion.
But why should it matter morally, if the relevant counterfactual is true, whether it is true in virtue of intrinsic (or intrinsic and species specific) features of the individual who will be deprived of future personhood if I act? If some non-person who is a counterfactual person can be deprived of future personhood without a great wrong being committed, why is this not true of all of them? Why does what grounds the counterfactual matter morally? This question, I think, is what the defender of Kulesa’s explanation of the prima facie impermissibility of early-stage abortion has to answer. But if some answer can be given there is a further challenge. Why does this not simply establish that it is the present, counterfactual-grounding, categorical properties, rather than the truth of the counterfactual to which moral significance attaches? Why is it necessary to bring in the concept of counterfactual personhood at all?
I add, parenthetically, that this challenge is not just a challenge to Kulesa, but a challenge also to a defender of the ‘future like ours’ anti-abortion argument of Marquis (Reference Marquis1989), to which Kulesa refers. For he too must explain the significance of a counterfactual truth, since, in the relevant sense, if an abortion is performed, the future like ours is not what will happen, i.e., an actual future, but a counterfactual future of the type Kulesa is discussing. Thus, if Kulesa’s explanation is inadequate to explain the impermissibility of early abortion, this has wider implications for the anti-abortion debate.
4. Conclusion
Kulesa’s paper attempts to reconcile conflicting intuitions about abortion and embryo rescue cases by introducing the concept of a counterfactual person. However, his argument does not succeed in establishing that the mere fact that an entity can be classified as a counterfactual person is a sufficient condition for the prima facie wrongness of killing it. Appeal to the concept of counterfactual personhood does not adequately explain the wrongness of killing an early foetus, whilst also providing an explanation of the permissibility of the destruction of frozen embryos. Unless Kulesa can demonstrate why appeal to the concept of a counterfactual person is needed to explain these intuitions his argument fails.
