A Rational Look at Abortion

by Mario Derksen, B.A.
November 2000


One of the most hotly contested issues inside and outside of biomedical ethics today is abortion. Just lately the discussion flared up again due to the release of the controversial abortion drug RU-486, “a pill to increase access to abortions and let women get them privately from their own doctor instead of facing shouting protesters at clinics.”[1] Due to the grave nature of the issue, there are very passionate people on both sides of the fence. Unfortunately, a lot of the debate has been pervaded with rhetoric and polemic, something which only obscures the issue even further rather than helping to clarify it. In this brief essay, I shall attempt to clear away some of the confusion by cooling the rhetoric with reason enlightened by scientific facts. Given the limited scope of this paper, however, I will be forced to restrict myself to picking out two common pro-abortion arguments, taken from two fairly well-known essays by Mary Anne Warren and Judith Jarvis Thomson. I shall examine and reflect on them to see whether they can stand up to rational criticism.

Before even beginning to discuss the issue of abortion, it is imperative to agree upon a starting point from which to reason. Because some people differ even about this very point, the incommensurability of the pro-abortion and the anti-abortion paradigms is enhanced, and this is probably also the reason why people are tempted to arrive at different conclusions about this topic. It seems to me, however, that to start with the definition of abortion and an examination of the beings involved would be a fair move.

Abortion is generally defined as the unnatural termination of a pregnancy, and it necessarily involves the killing of (at least) one human fetus. This definition is not contested, and I think it seems clear that it is correct. Science confirms that life begins at conception, and that this life is human is a—scientific as well as logical—necessity, because it is the product of two humans. Ergo, the fetus involved is human. Secondly, the fetus is, at least scientifically speaking, a singular and individual organism, as evidenced by its own unique genetic make-up, which it shares with no other human being on earth. Finally, that the fetus is alive is confirmed through empirical observation, and hence forcing that life to come to an end involves killing (whether actively or passively). Therefore, the unavoidable conclusion is that abortion forcibly puts to death a unique human being. Again, the explication of this definition is uncontested and thus I shall not dwell on it any further. Rather, I shall now turn to the moral implications necessarily connected with abortion.

The question that arises is as to whether or not abortion is morally justifiable. It cannot be wrong by definition, since sometimes there is moral justification for forcibly putting to death another human being, in such cases as self-defense, just war, or capital punishment.[2] Hence, it is reasonable to raise the question whether abortion might be another such instance where one is justified in taking a human life.

The pro-abortion side submits that it is, and different arguments have been put forward to substantiate that claim. A very popular one is the contention that human fetuses are not persons, and that only persons have a right to life and justice against others, at least when these others’ rights are at stake. Mary Anne Warren’s essay “On the Moral and Legal Status of Abortion”[3] argues just that. Curiously, though, Warren does not indicate in her essay why one should accept the view that one ought to be a person in order to have moral rights in the first place; it seems to me that she considers it self-evident that this is so, mentioning only that she thinks “there are very good reasons for not defining the moral community”[4] in such a way that every human being is ipso facto included. Be that as it may, she asks: “What moral characteristics entitle an entity to be considered a person?”[5] and goes on to list five “traits which are most central to the concept of personhood”[6] in her opinion. They are: (1) consciousness, (2) reasoning, (3) self-stimulated activity, (4) the capability to communicate, and (5) the presence of self-concepts and self-awareness.

It is here that I feel reason ought to make an objection. While it seems true to me that the categories Warren lists are sufficient for personhood, I see no reason to believe that they are necessary. In fact, the author herself admits a few lines later that we do not “need to insist that any one of these criteria is necessary for personhood.”[7] But if not “any one” of them is necessary, then none of them are, and if that’s the case, what does her argument prove? And what if one thinks that to define personhood in terms of functional abilities is not prudent at all? It seems that for Warren’s argument to be forceful, one would have to presuppose her functionalist paradigm as true, i.e. the basic idea that there even are criteria a human being must meet in order to be considered a person. But this would be an instant of question-begging, in my estimation. After all, one could very well adopt the view that all humans are persons in virtue of their humanity, and thus being human necessarily involves being a person.

No matter what the answers to these questions might be, the very idea of functionalism carries with it serious problems, I believe. For one thing, even if one agrees with Warren’s position that the functionalist view of human personhood is true, the criteria proposed by her could be rejected by anyone who disagrees with them, since they seem gratuitously imposed, and what is gratuitously asserted can be gratuitously denied.

Secondly, a major problem, in my view, is the fact that Warren’s view seems to disregard that an argument can be made that “personhood is not something that arises when certain functions are in place, but rather is something that grounds these functions, whether or not they are ever actualized in the life of a human being,”[8] for “to claim that a human being can be functional, become non-functional, and then return to a state of function is to assume that there is some underlying personal unity to this individual.”[9] What Dr. Beckwith means here is that, supposing that John Doe has a car accident and becomes comatose for three months, according to the functionalist view proposed by Warren, one would have to conclude that while Doe was in the coma, he ceased to be a person since he didn’t meet any of her proposed five criteria for personhood, and that when he awoke, he became a person again.

But, surely, this is absurd. Therefore, Beckwith concludes, “it is intelligible for us to say that the person who has returned to functional capacity is the same person who was functional prior to being in a non-functional state and yet continued to exist while not functioning.”[10] There must be some underlying personal unity, then, unless we wish to concede that John Doe before the coma was a different person from the John Doe after the coma. Now, I believe that while this objection to Warren’s position may not be fully convincing and can certainly be elaborated upon, it does present an obstacle for her notion that human beings must first meet certain requirements before they can be considered persons, i.e. beings with moral rights.

Earlier I mentioned that Warren does not give us any justification for embracing a functionalist paradigm as far as personhood is concerned. This statement needs qualification, however. Warren does mention that the view that personhood is intrinsic to any human being from the first moment of his or her existence carries with it the problem that it makes the traditional syllogism[11] against abortion question-begging. I concede that this is true. Nevertheless, I believe that her conclusion, namely that therefore we must reject the humanity-equals-personhood view, is incorrect; for what her argument shows, I think, is not that the view is untenable, but rather that no syllogism is necessary to argue for it. She agrees, after all, that the statement “it is wrong to kill innocent human beings” is “a self-evident moral truth.”[12] In other words, it seems to me that Warren rejects the view that humanity necessarily implies personhood precisely and only because it would make abortion impermissible. But surely this is question-begging!

Moreover, leaving individual arguments aside, it appears to me as though something so controversial and disputed as the applicability of the notion of personhood should not be the fundamental basis on which to justify abortion; that would be to walk a moral tightrope, for lack of a better metaphor. Some pro-abortionists, such as Judith Jarvis Thomson, agree with this assessment, and so they argue instead that even if the fetus is a person, abortion can still be morally justified. I must say that, personally, I deem this position much more daring and therefore worthy of more consideration than Warren’s. Hence, I will now proceed to examine Judith Thomson’s main argument in her essay “A Defense of Abortion.”[13]

According to Thomson, we are to imagine a situation in which a violinist with a fatal kidney disease has been artificially hooked up to you in order to use your kidneys for nine months. This has been done by the Society of Music Lovers and without your permission. If you unplugged yourself from this violinist now, he would die, and in that sense you would be responsible for his death. Inevitably, then, the moral question arises whether you are morally obligated not to unplug yourself from the violinist for the time being, even if this causes all sorts of inconveniences for you, such as staying in bed all day; after all, all “persons have a right to life, and violinists are persons . . . [and] a person’s right to life outweighs your right to decide what happens in and to your body.”[14]

Admittedly, this is a very shrewd argument, even if somewhat strange, and if the analogy holds which Thomson is obviously drawing between the violinist situation and the situation where a woman has an unwanted pregnancy, then the anti-abortion position would suffer a severe setback. After doing some research and critical analysis, however, I must say that I believe that there is plenty of evidence which indicates that Thomson’s violinist analogy is in fact a false analogy and therefore does not hold. To strengthen my objection, I wish to offer several reasons for rejecting the analogy as false, and I shall now share them in the remainder of this paper.

First, the most obvious difference is perhaps that by unplugging the violinist, one would not engage in direct killing but in letting die. The violinist would be killed by a disease, whereas the fetus is aborted by killing it actively; but even if not, the passive killing of the fetus by exposing it to the elements would be a case of active neglect, for one would be withholding essential support for the life of the fetus. Granted, I realize that one could consider the unplugging of the violinist to be a case of active neglect as well (though to me that seems counter-intuitive), and if this is done, then this first objection is certainly weak.

Be that as it may, there are much stronger objections to Thomson’s analogy. One of them is that her analogy, if valid, only holds true for rape cases, that is, in cases when a woman has been forcibly impregnated against her will, since in the analogy you were kidnapped by the Society of Music Lovers, and it was involuntary therefore. However, as John T. Wilcox points out, “the title of [Thomson’s] essay is ‘A Defense of Abortion,’ not ‘A Defense of Abortion Rape Cases.’”[15] So it seems misleading to use an analogy that could only hold for rape cases to argue for a position that does not restrict itself to allowing abortion in rape cases only.

Another point is that the violinist incident is extremely bizarre and unlikely, whereas pregnancy is the exact opposite—it is most natural and occurs all the time. Says Wilcox: “In Thomson’s essay we have something as universal and necessary [for reproduction] as pregnancy compared to something so rare it has never happened and perhaps could never happen.”[16] He then goes on to make the point that it is “at least arguable . . . that the moralities we have represent some ways of dealing with the realities and regularities of human life; and they may not fit well the irregularities and impossibilities.” Given that, he concludes that it is “plausible to regard [the two cases] differently from an ethical point of view . . . [since] what is appropriate for kidnapped kidney bearers and their violinist parasites might not be appropriate for mothers and the babes in their wombs.”[17] I think that this is a very good point and, though not necessarily conclusive by itself, one which should be seriously taken into consideration when evaluating the strength of Thomson’s violinist analogy.

A fourth objection that may be raised to Thomson’s analogy is the fact that the two cases are disanalogous inasmuch as unplugging the violinist is in no way comparable to the methods used for abortions. While the different methods vary, they all involve a very inhumane killing of the fetus, whether it be through cutting, suctioning, or intoxication. Yet, at no abortions during any point of gestation is the fetus given pain relief, even though the “centers necessary for pain perception develop early in the second trimester,”[18] and so justice would seem to require at least an attempt to relieve fetal suffering, especially in late-term abortions, where “[f]orcibly incising the cranium with a [sic] scissors and then suctioning out the intracranial contents is certainly excruciatingly painful.”[19] The point here is not to be graphic, but to point out that there is a fundamental difference between such a procedure, which is artificial, deliberate, and unnecessarily painful, and unplugging the violinist, merely letting him die.

 These are just some of the objections one can raise against Thomson’s analogy of the violinist, though multiplying them would extend the scope of this paper. I think, however, that the above objections are illustrative of some serious insufficiencies in Thomson’s argument. Of course, I am not claiming that the above-mentioned criticisms are themselves irrefutable or that they disprove all arguments for abortion, but rather that they are well-argued and seem rational, thus shedding light on this very complex issue, demonstrating that the violinist analogy is not that well-founded after all, though it may appear otherwise at first.

As I mentioned in the beginning, there are, without a doubt, many more arguments both in favor of and against abortion that deserve careful analysis and critical examination. But whatever these may be, I wish to offer one last objection to the pro-abortion position in general, which, if valid, would make any future arguments for abortion nearly indefensible, I believe. It is an argument from common sense,[20] and it can be used very effectively, I think, even if one were to leave all previous argumentation aside.

The argument is simply this. In the abortion debate, there are only four possible scenarios which could obtain, namely that (1) abortion is right and we know it; (2) abortion is wrong and we know it; (3) abortion is right and we don’t know it; and (4) abortion is wrong and we don’t know it. No other option is possible. So let us suppose that (1) obtains. In this instance, abortion would be morally justified, and no problem could arise. However, we know that (1) is false, for, obviously, there is serious controversy about abortion, and hence we do not “know” that abortion is morally right. If one of the other three scenarios should obtain, however, then abortion would be morally wrong, for if (2) obtains, then abortion is murder; if (3) obtains, then abortion is criminal negligence; and if (4) obtains, abortion is manslaughter.

I should perhaps elaborate on (3). Suppose you are a truck driver, and while you are driving at night, you suddenly see in front of you what looks like a man lying on the road, though you are not sure that it’s a man, for it might actually be a dummy. From your view, you simply cannot tell. Would it be morally justifiable to run over this “person”? I think everyone would agree that the answer would be no. The very fact that you don’t know whether it’s a human or a dummy obliges you not to run over it in order to be on the safe side, and to do otherwise would be morally objectionable. This is how we are to understand (3), that not knowing whether abortion is right or wrong when in fact it is right is still morally inadmissible.

Hence, I think it is safe to say that this argument from common sense establishes that even if we reject all other rational arguments, in three out of four possible cases, abortion would be wrong. It seems to me that that number—namely 75%—is sufficient to warrant the claim that abortions ought not to be performed.

Given all of the above, I feel compelled to conclude that abortion, at least as far as the cited arguments from Warren’s and Thomson’s essays are concerned, has not been sufficiently justified. Since it involves human life (on both sides), it is a very serious issue and must be very forcefully argued for—by both sides. The principle of non-contradiction requires that only one of the two sides of the issue can be right, and, necessarily, the other must be wrong, very wrong. Again, then, in order to ever come to a resolution of this issue, the primary requirement is that we let reason be our guide to show us what is morally right and what is morally wrong. We ought to avoid polemic, convenience, and rhetoric, which can slip into the dispute—from both sides—so easily if we are not careful.


Works Cited

“Abortion Pill Heads for Clinics.” USA Today, 20 November 2000. Internet edition.

Beauchamp, Tom L., and LeRoy Walters, eds. Contemporary Issues in Bioethics. 5th ed. Belmont, CA: Wadsworth Publishing Company, 1999.

Beckwith, Francis J. “Abortion, Bioethics, and Personhood: A Philosophical Reflection.” Paper from the Center of Bioethics and Human Dignity at www.bioethix.org, n.d.

Kreeft, Peter. Making Choices: Practical Wisdom for Everyday Moral Decisions. Ann Arbor, MI: Servant Books, 1990.

Sprang, M. LeRoy, and Mark G. Neerhof. “Rationale for Banning Abortions Late in Pregnancy.” Journal of the American Medical Association 280, no. 8 (1998): 744-747.

Wilcox, John T. “Nature as Demonic in Thomson’s Defense of Abortion.” New Scholasticism 63, no. 4 (1989): 463-484.


[1] “Abortion Pill Heads for Clinics,” USA Today, 20 November 2000, Internet edition.

[2] I realize that even this is contested, but I think it is safe to say that most people agree that sometimes killing another human being is morally justifiable.

[3] In Tom L. Beauchamp and LeRoy Walters, eds., Contemporary Issues in Bioethics, 5th ed. (Belmont, CA: Wadsworth Publishing Company, 1999), 222-31.

[4] Beauchamp and Walters, Bioethics, 227; italics added.

[5] Beauchamp and Walters, Bioethics, 227.

[6] Beauchamp and Walters, Bioethics, 227.

[7] Beauchamp and Walters, Bioethics, 227f.; italics given.

[8] Francis J. Beckwith, “Abortion, Bioethics, and Personhood: A Philosophical Reflection” (paper from the Center of Bioethics and Human Dignity at www.bioethix.org, n.d.), 2.

[9] Beckwith, “Abortion,” 2.

[10] Beckwith, “Abortion,” 2.

[11] I.e. killing innocent humans is wrong; a fetus is an innocent human; hence killing fetuses is wrong.

[12] Beauchamp and Walters, Bioethics, 226.

[13] In Beauchamp and Walters, Bioethics, 202-11.

[14] Beauchamp and Walters, Bioethics, 203.

[15] John T. Wilcox, “Nature as Demonic in Thomson’s Defense of Abortion,” New Scholasticism 63, no. 4 (1989): 472.

[16] Wilcox, “Nature as Demonic,” 468.

[17] Wilcox, “Nature as Demonic,” 468f.

[18] M. LeRoy Sprang and Mark G. Neerhof, “Rationale for Banning Abortions Late in Pregnancy,” Journal of the American Medical Association 280, no. 8 (1998): 745.

[19] Sprang and Neerhof, “Banning Abortions,” 745.

[20] I am indebted to Dr. Peter Kreeft for this argument. Cf. Peter Kreeft, Making Choices: Practical Wisdom for Everyday Moral Decisions (Ann Arbor, MI: Servant Publications, 1990), 119-21.


You may contact Mario Derksen by emailing him at mario@cathinsight.com