In Charles Tandy (ed.), Death And Anti-Death, Volume 19: One Year After Judith Jarvis Thomson (1929-2020), Ria University Press. pp. 139-152. 2021.
Abstract
1. First there is an outline of the libertarian approach taken here. 2. On the assumption of personhood, it is explained how there need be no overall inflicted harm and no proactive killing with abortion and infanticide. This starts with an attached-adult analogy and transitions to dealing directly with the issues. Various well-known criticisms are answered throughout. 3. There is then a more-abstract explanation of how it is paradoxical to assume a duty to do more than avoid inflicting overall harm and, instead, positively benefit. A putative counterexample is explained away. 4. A positive theory of intellectual personhood is defended in principle, but not made precise, which is sufficient to be practical. The greater moral value of intellectual-personhood is defended. An important putative reductio of the potential-personhood argument is refuted. 5. Several further criticisms that apply to both types of defences are answered. 6. Conclusion.
Key words: abortion, infanticide, libertarianism
1. Introduction
This essay takes one radical interpretation of a libertarian approach (it appears to differ significantly from the others[1]). However, non-libertarians could accept the specific arguments used here without thereby committing themselves to that ideology. Libertarianism is about not infringing the liberty of persons. In practice, this means not engaging in any offensive (as opposed to defensive or rectificatory) assault, bodily harm, trespass, damage, theft, fraud, contract-breaking, etc. (collectively, not initiating impositions) with respect to other people’s bodies and possessions (where these are acquired and held without initiating impositions).[2] A libertarian approach appears to explain how abortion and infanticide can be compatible with the liberty of persons and not be immoral. This is primarily because they need involve no overall inflicted harm, and because it is paradoxical to suppose a moral obligation to do more than avoid causing overall inflicted harm. But it is also, separately and sufficiently, because this is not done to persons in the intellectual and the moral sense. In both cases various criticisms can be shown to be mistaken, with property and contracts often being unavoidably and crucially relevant.
2. No overall inflicted harm and no proactive[3] killing
First, consider a position relevantly similar to that taken in the classic article Thomson 1971. But that article assumes rights, which we will see it is not necessary to do (although we will come back to them). Rights often make for a more complicated and disputable approach: they are all-too-promiscuously assertable and often without clear explanations, origins, and implications. Also, that article does not fully and consistently draw out and accept its own relevant logical implications. This will then transition into applying the arguments to unborn and infant humans.
On the assumption that the unborn human is a person, or has equivalent status, pregnancy appears to be analogous with your becoming physically attached to an unconscious adult person—whether by choice, chance, carelessness, or coercion—where no one else could or would have supported that particular adult and he requires your bodily support for nine months. If you decide that you do not wish to continue the benefit of your support, then there is no overall inflicted harm on the unconscious adult by unplugging yourself and stopping: he is no worse off than if you had not started to support him in the first place. The idea of ‘no overall inflicted harm’ is crucial. We might be said to ‘cause harm’ in the sense that the act of unplugging is the proximate cause of the adult’s death, and causing death here is to cause a harm. But clearly there is no overall harm inflicted on the adult, even if we first plugged in by choice. This appears to be not too controversial a claim when it comes to a simple unplugging and walking away. That might not even appear to most people to constitute direct killing: it looks more like letting die.
But what if the only way to stop your support is by directly killing the unconscious adult? For instance, because you are so intricately connected to him that you have to pull him to pieces in order to free yourself. And here we need to see that confusion is often caused by conflating two completely different types of even direct killing. One is proactively, or offensively, killing someone who would not have died at that time but for you (such as is usual with murder): this is clearly inflicting overall harm on the killed person. The other is reactively, or defensively, killing someone as the only way to escape from the undesired burden that they are to you; as we have supposed here. And, as we also have supposed here, the person is still no worse off than if you had never been connected to him (“no one else could or would have supported that particular adult”). It might sound gruesome, but with this reactive killing you have still inflicted no overall harm.
There appear to be two necessary and sufficient aspects for moral permissibility here: (1) you have defended your chosen use of your own body; and (2) the unconscious adult is not worse off than if you had never started to support him in the first place. One far-too-hasty criticism might be that such an argument could seem to imply that killing your adult offspring is permissible, as they are thereby not overall worse off than they would have been had you never benefitted them—by conception, etc.—in the first place. But that would be to overlook (1): killing your adult offspring is not defending your chosen use of your own body.
Another objection is that we do not normally allow people to inflict excessive harm on others even in self-defence. Are you in this situation with the intricately-connected adult? No. It is necessary to do this much bodily harm in order to free yourself, and the supported adult would not have lived anyway without you. Therefore, he is not worse off overall as a result of your behaviour. But when we inflict excessive harm on aggressors (say, by shooting someone who is only poking his finger in your chest) then the level of harm done is probably not necessary to defend ourselves and we do leave them overall worse off as a result of our behaviour. (We can leave aside here how we ought to balance potential proactive harm to ourselves against the reactive harm we need to inflict to stop it.)
‘But you put him in that situation!’ is a popular criticism (when applied to abortion after voluntary sex, at least: the responsibility argument). This criticism might be sound if it were to refer to a person that someone else would, or might, otherwise have saved (or brought into existence). For then our behaviour would have worsened the overall condition of the person compared to what he would otherwise have had. As it is, however, that is not the case. To start to support some unconscious (or unborn) person and then stop is not in itself to inflict any overall harm on him even if we ‘put him in that situation’.
Another criticism, at least if it is a person or has equivalent status, is that there is some sort of consent (the tacit-consent argument) or even a quasi-contract between the mother and the unborn human to bring him to term, etc. It is true that a pregnant woman usually did tacitly consent to, at least risk, starting to support the unborn human. However, that is not thereby tacitly to consent to continuing support of it. Or as Mahon 2016 has it—but in terms of rights—“no one has the right to use my body without my permission; it is also true that I may revoke this permission at any time” (p.73). Strictly, however, a valid contract to the contrary with another person would seem to override any absolute ‘right of revocation’ (but we don’t need to discuss contractual penalties or specific-performance contracts here). Between an unborn human and its mother there is no kind of—even implied—offer, or acceptance of that offer, or any quid pro quo; which contracts require. Neither is there any such thing once the infant is born.
When there are objections to the killing of unborn humans, the explanations often fail to distinguish both inflicted harm from overall inflicted harm and proactive killing from reactive killing. For instance, Marquis 2007 concludes that “Deprivation of an FLO [future like ours] explains why killing adults and children is wrong. Abortion deprives fetuses of FLOs. Therefore, abortion is wrong” (p.764). For the sake of argument, let us assume that an “FLO” is what makes life valuable. However, the “deprivation of an FLO” by the killing of adults and children would, we may suppose, be proactive (or offensive) and also inflict an overall harm on them. The latter two reasons are what make the killing wrong; not merely causing a “deprivation of an FLO”. But with an unborn human the “deprivation” is not offensive but defensive (defending the woman’s chosen use of her own body), and it leaves the unborn human with no overall inflicted harm.
What sense can we now make of the ‘right to life’ of an unborn human? Exactly as with adults, a ‘right to life’ cannot imply an absolute right not to be killed under any circumstances, let alone a right to be kept alive. It can only be a right not to be proactively, or offensively, killed. But the unborn human is not proactively killed by an elective abortion. He is killed in defence of the woman’s ‘right to control her own body’, to put it in equivalent rights terms. And no overall harm has been inflicted on him. However, if someone were to kill an unborn human without the consent of the mother (by a compulsory abortion, or stabbing, or poisoning, etc.) then that would be a proactive killing and it would inflict an overall harm on him. Assuming that the unborn human is a person, or has equivalent status, this would flout his ‘right to life’.
What of an infant: surely, he is not killed in any kind of defence? True. But given that he is on our property (or property we are contractually using) and we have no wish to support him or to give him away, then he is going to die. Therefore, he may—even must—be euthanised to alleviate any suffering of which we would be the cause. There is no inherent obligation to give the infant to other people to care for him. And other people would be proactively imposing if they trespassed in order to take the infant. But on the assumption of personhood status, at least, this would be a rescue rather than a theft.
3. The paradox of assuming a duty to benefit
Perhaps we can avoid the plethora of contingent complications surrounding abortion (such as are elaborated in Johnson 2019) and the disputable relevance of the analogy of being attached to another adult person (the stranger-versus-offspring argument). It is possible to make a more fundamental and abstract moral argument that should be clearer and more cogent.
To bestow a benefit on others, for their own sakes, is prima facie morally good. To proactively impose—inflict overall harm—on others, for whatever reason, is prima facie morally bad. To do neither is prima facie morally neutral (or innocent). It cannot proactively impose on other people to deny them, or to stop, a bestowal of a benefit (to genuinely contract to help someone in some way and then break one’s contractual obligations is a proactive imposition and not the denial of a bestowable benefit).
Libertarianism, in particular, appears to require and make these three moral distinctions. But, in any case, logical analysis also appears to imply them. For if we want to classify mere failure to benefit people as immoral, then—conversely—we seem bound to classify mere failure to proactively impose on people as in itself positively moral. However, proactive impositions are usually far easier to bring about than equivalent bestowable benefits (e.g., destroying someone’s house by arson versus providing someone with a free house). Consequently, we omit to proactively impose on people to a far greater extent than we omit to bestow benefits on them. This implies a paradox: merely by doing nothing, we are usually both moral and immoral at the same time or—on balance—positively extremely moral. There is no conceptual room for moral neutrality (unless, perhaps, when we are contingently not in a position to do either or they are by sheer chance in perfect balance). To avoid this paradox, it seems only coherent to distinguish good, bad, and neutral (or innocent) moral behaviour.
From this more-fundamental argument we can see that the unborn and infant human is only benefitted by conception and support (or, at least, there is no inherent proactive imposition in that process). Consequently, removal of that continuing bestowed benefit cannot in itself be a proactive imposition—or overall inflicted harm—even if we assume personhood. Therefore, if they do not inflict pain or suffering, abortion and infanticide are in themselves morally neutral.
What about Singer 1971’s well-known example of a drowning child in a shallow pond? Isn’t there a moral obligation to save him? And if there is, isn’t a foetus or infant—at least with the status of a person—in a sufficiently similar moral situation? But if there is a moral obligation to save the drowning child, then this is not because it is immoral not to be positively moral (at least, when this costs us little). As we have seen, that creates a paradox. If it is immoral—and it seems that it usually is—then that is because there are common contracts, implicit or explicit, concerning our behaviour when we choose to enter any area. And one of those contracts is usually that in the event of temporary and extreme emergencies (serious fires, crimes, accidents, etc.) we are obliged to help directly or seek appropriate assistance (in more detail see Lester 2020). Therefore, abortion and infanticide are allowable unless we have a contract to the contrary. And even then, breaking the contract will proactively impose on the person with whom we have the contract and not on the unborn or infant human.
4. Personhood and its moral relevance
It would be an omission to say nothing about personhood. Liberty is for persons and so there ought to be some idea of what persons are. Such a theory can help to alleviate criticisms of the previous arguments about abortion and infanticide if we see that they do not even involve persons. Moreover, it seems that non-personhood is sufficient in itself for a defence.
Unborn or infant humans are potential persons, as Tooley 1972 famously argues. But that article also elects to “treat the concept of a person as a purely moral concept, free of all descriptive content” (p.40). Here, rather, a person as such is assumed first and foremost to be a purely descriptive concept. It is a logically separate matter whether persons have moral status or rights and what they might be. The various theories of positive, rather than normative, personhood need not be compared and contrasted. They tend to agree that some degree of reflective consciousness has to be obtained, and that is—or creates—a sophisticated mind that non-persons lack. And if it is not inherently wrong to kill animals that are not persons, then it is not inherently wrong to kill humans that are not persons (presumably, vegetarians would need a different argument). Consequently, accounts of personhood have been used to argue in defence of abortion, such as, seminally, Warren 1973; but that article does not fully and consistently draw out and accept its relevant logical implications with respect to infanticide.
To what extent is it a problem if we cannot agree, even in principle, exactly when a person exists? We can try to determine more precisely the theory and criteria that fit our intuitions as to when a human has sufficiently become a person. If we find that the age is higher than we feel morally comfortable with as regards infanticide, then there is no reason that we are obliged to practice it at that age or contract into living in a private-property area that draws the line higher than we can accept. Just as with abortion and age-of-consent laws, there is likely to be variation from place to place. However, exactly what constitutes a person is a separate matter from the thesis that sufficient intellectual sophistication does seem to be what gives persons their greater moral value and makes it immoral to inflict overall harm on them.
A critic might even ask: ‘Why is actual possession of a sophisticated mind necessary for greater moral value?’[4] This seems somewhat like asking, ‘Why are unhappiness or pain bad and happiness or pleasure good?’ We might be inclined to wonder whether the questioner understood the meanings of the words or the experiences they represent. However, some argument is possible. For instance, suppose three apparently identical human adults: the first is brain dead; the second has the permanent mental capacity of an infant; the third is normal. If you could only save the life of one of them, then which one is it the most moral for you to save? Surely, the normal third. (And if you could save two, then you would also choose—if, indeed, you would choose either—the permanent-infant second over the brain-dead first.) If we have to explain this choice, then we might say that a greater consciousness is more valuable for anyone to possess than a lesser consciousness and that we simply feel that to make it of more inherent moral value (this is not to imply that normal people can be morally ranked by intelligence or that being a person is the only morally valuable thing). If someone were to insist that all three humans—or, at least, the second and third—are of equal moral value, then we might suspect that this person is morally confused.
One important attempt at a reductio ad absurdum of this position is that by this theory of personhood a temporarily unconscious adult human might seem to be only a potential person, and therefore morally on a par with an unborn or infant human as regards the alleged permissibility of killing him. But then it seems that one would also have to say that someone in a dreamless phase of sleep is not currently a person either. Or maybe even someone who is not currently thinking at a sufficiently sophisticated level. And one might use the same type of argument to claim that a surgeon is only a surgeon when he is engaging in surgery; otherwise, he is only a potential surgeon. Such considerations appear to violate our understandings as to what being a person, or surgeon, amounts to. It looks far more persuasive to see this as an existing person: personhood is not merely potential but has already been achieved; a life as a conscious person has been experienced before and will be experienced again. It is simply that this person’s consciousness is temporarily interrupted, and that is why the moral standing relating to personhood remains.
5. Some further criticisms and responses
What of “the concrete ethical fact that giving a fetus fetal alcohol syndrome is immoral” and therefore abortion, because it is worse, is also immoral (Hendricks 2019, p.2)? Assume that any ensuing child would still have a life worth living. Contrary to the popular intuition, a woman carrying a foetus does not overall inflict harm (proactively impose) by drinking alcohol that impairs its development. This is really only failing to give the foetus the benefit of optimal developmental circumstances. Possibly even the majority of pregnant women do this: how many women choose the best-balanced diet, most pollution-free area, stress-free circumstances, healthiest bodies, etc.? Hence the benefit given, or withheld, is usually a matter of degree; and it can never amount to an overall harm inflicted because it is all a gift. As we have seen, it is paradoxical to assert that a failure to benefit—enough—can be immoral. Hence, “giving a fetus fetal alcohol syndrome” is not in itself immoral.[5] But this is so more obviously when the foetus is to be aborted or the infant is to be euthanised, which appear to be separate sufficient reasons that causing foetal alcohol syndrome need not be immoral. Consequently, this argument against abortion (or against “pro-choice”) fails. And as a consideration of “The Axiology of Abortion” rightly argues, “it would be better if the pro-choice position is right” (but we will not evaluate those arguments here) despite the fact that it wrongly argues, as we have now seen, that “there is no good reason to think the pro-choice position is correct” (Hendricks forthcoming, p.1).
There are also the following relevant claims: “our overall reductio succeeds. That is, psychological accounts of personhood do at present suggest the permissibility of infanticide, even for healthy infants. These accounts also imply the permissibility of pre-personal acts such as forced organ donation, use of infants for medical research, use of infants for sexual gratification, and discrimination against infants” (Rodger, Blackshaw, & Miller 2018, p.119). As we see, this is intended to be aimed at “psychological accounts of personhood” in particular. But the same criticisms can also be regarded as applying to the earlier sections of this essay. Therefore, let us very briefly consider all of these in turn leaving the sexual one for last. (Brevity is all that is possible here, and it will be too brief for some readers.)
If parents or guardians do not wish to support a healthy infant—to assume the harder case—or give it away, and there is no contract to the contrary, then they cannot be proactively coerced into doing so without infringing their liberty. If they opt for painless infanticide, for whatever reason, then no overall harm is thereby inflicted even if it were done to a person.
Why “forced” organ donation? That might seem to be implied only if consequentialism is assumed. However, long-term good consequences usually seem to fit liberty better (with its diverse voluntary alternatives, including commercial ones) than proactively imposed behaviour (with its immediate victims plus all the procrustean authoritarianism and moral hazards). If interpersonal liberty is to be respected, then no persons can be “forced” to donate their infant’s organs.
Medical research on non-suffering, non-persons seems moral and compatible with both interpersonal liberty and consequentialism. Research on humans is often more useful than research on other animals. However, any private company is likely to be influenced by consumer opinion on these matters.
What is wrong with “discrimination against infants”? By this the article means to refer to race, sex, sexual orientation, disability, etc. This appears to be based on the politically-correct authoritarianism that flouts freedom of association and choice. Consequently, it is possibly not even a popular opinion. Parents or guardians should not be obliged to keep alive, or hand over to other people, an infant when they do not wish to do so for whatever reasons. There may also be the epistemological problem of determining whether they hold the ‘wrong’ reasons for their behaviour rather than ‘acceptable’ ones.
“Sexual gratification” is substantially different. The article might also have added cannibalism, as that is in principle possible too (at least for non-vegetarians), and so it is added here. Both of these appear to be cases where any hypothetically minuscule ‘gains’ from allowing them appear to be completely overwhelmed by the horror and revulsion that virtually everyone would feel. There is also the possibility of undermining the general moral taboos against paedophilia and cannibalism. Moreover, no one is advocating that either of these things should be allowed and arguing that there is a significant welfare loss if they are not. Therefore, this example is in principle valid but in practice vacuous.
Consequently, far from being logical reductiones ad absurdum each of these objections amounts to versions of the logical fallacy of argumentum ad populum. It may be true that “most people will feel the force of these intuitions and not give them up lightly” (p.118). Therefore, it is perhaps a political reductio (outside the current Overton window of political acceptability). However, intellectually, popular intuitions cannot refute philosophical arguments.
6. Conclusion
This essay has attempted to offer clear and cogent arguments that abortion and infanticide need not be immoral. In short, because they need not inflict any overall harm—even with direct killing—and because it is paradoxical to assume a moral duty to do more than that. But also, separately and sufficiently, because they involve non-persons. And it is only persons that have a particularly high moral value such that we ought never to inflict overall harm on them. This has been a different and more radical approach than is normal in the literature. Consequently, there is undoubtedly plenty of scope for criticism. But, as critical-rationalist epistemology explains, all arguments are only ever conjectural explanations (rather than ‘supporting justifications’).[6] Hence criticisms are essential to test arguments and are always to be welcomed.[7]
(June 2021; revised 4th April 2022.)
References
Block, Walter. 2011. “Response to Wisniewski on Abortion, Round Two.” Libertarian Papers, 3 (4): 1-13.
Hendricks, Perry. 2019. “Even if the fetus is not a person, abortion is immoral: The impairment argument.” Bioethics, 33 (2): 245-253.
Hendricks, Perry. Forthcoming. “The Axiology of Abortion: Should We Hope Pro-Choicers or Pro-Lifers are Right?” Ergo: An Open Access Journal of Philosophy. PhilPapers https://philpapers.org/rec/HENTAO-15 (accessed 29/9/20).
Johnson, David Kyle. 2019. “The Relevance (and Irrelevance) of Questions of Personhood (and Mindedness) to the Abortion Debate.” Socio-Historical Examination of Religion and Ministry, 1 (2): 121‒53.
Lester, J. C. 2019. “The Heterodox ‘Fourth Paradigm’ of Libertarianism: an Abstract Eleutherology plus Critical Rationalism.” Journal of Libertarian Studies, 23: 91-116.
Lester, J. C. 2020. “Peter Singer’s ‘Famine, Affluence, and Morality’: Three Libertarian Refutations.” Studia Humana, 9 (2): 135-141.
Lester, J. C. 2022. “Eleutherological-Conjecturalist Libertarianism: a Concise Philosophical Explanation.” MEST Journal (forthcoming).
Mahon, James. 2016. “Abortion and the Right to not be Pregnant.” In Allyn Fives & Keith Breen (eds.), Philosophy and Political Engagement: Reflection in the Public Sphere. New York, NY: Palgrave Macmillan, pp. 57-77.
Marquis, Don. 2007. “An Argument That Abortion is Wrong.” In Russ Shafer-Landau (ed.), Ethical Theory: An Anthology. Wiley-Blackwell, pp. 439-450.
Miller, D. W. 1994. Critical Rationalism: A Restatement and Defence. Chicago and La Salle, Ill.: Open Court.
Rodger, Daniel, Blackshaw, Bruce P., & Miller, Calum. 2018. “Beyond Infanticide: How Psychological Accounts of Persons Can Justify Harming Infants.” The New Bioethics, 24 (2): 106-121.
Rothbard, Murray. 1982. “Children and Rights.” The Ethics of Liberty. Atlantic Highlands, N.J.: Humanities Press, Chap. 14.
Singer, Peter. 1972. “Famine, Affluence, and Morality.” Philosophy and Public Affairs, 1 (3): 229-243.
Thomson, J. J. 1971. “A Defense of Abortion.” Philosophy and Public Affairs, 1 (1): 47-66.
Tooley, Michael. 1972. “Abortion and Infanticide.” Philosophy & Public Affairs, 2 (1): 37-65.
Warren, M. A. 1973. “On the Moral and Legal Status of Abortion.” Monist 57 (1): 43-61.
[1] Notably, for instance, Rothbard 1982 and Block 2011.
[2] The philosophical arguments concerning what abstract theory of liberty this entails (the eleutherology) and how this relates to property need not detain us here; a common-sense understanding is sufficient. But for a brief explanation see Lester 2022 and in more detail Lester 2019.
[3] In this context, ‘proactive versus reactive’ appears to be slightly clearer that the ‘initiated versus responsive’ distinction used in explaining the abstract theory of liberty (e.g., as found in Lester 2022). It is used from here onwards to avoid any confusion that might be caused by changing to the latter distinction.
[4] One anonymous reviewer actually did so.
[5] That said, if a foetus is going to become a child then foetal alcohol syndrome is certainly not desirable. There must be a variety of ways of discouraging it that need not interfere with anyone’s liberty, and which are thereby probably more efficient as a result.
[6] See, for instance, Miller 1994.
[7] An early version of this essay was first submitted for publication in 1980. Since then there have been several other attempts. Whilst the reviews were sometimes useful, none of them appeared to offer anything that looked like a refutation of the central arguments or even a serious problem for them. This may suggest that the real problem is the radical nature of the conclusions: they are interpreted as reductiones ad absurdum, even though the arguments could not clearly be faulted.