Suffer the Violinist: Why the Pro-abortion Argument from Bodily Autonomy Fails


Richard J. Poupard

Article ID:



Oct 5, 2023


Oct 26, 2018

This article first appeared in the Christian Research Journal, volume 30, number 4 (2007). The full text of this article in PDF format can be obtained by clicking here. For more information about the Christian Research Journal, click here.



The argument from bodily autonomy claims that one human being does not have the right to use the body of another human being for its survival. Abortion advocates have advanced this argument in order to justify elective abortion even if one grants that the fetus is a rights-bearing individual. This allows pro-abortion choice proponents, then, to concede the major premise of the pro-life position and still justify elective abortion. This argument was illustrated by Judith Jarvis Thomson’s famous violinist analogy, and has recently been defended by legal scholar Eileen McDonagh and philosopher David Boonin. The bodily autonomy argument and their defenses of it fail for at least four reasons. First, the argument fails to account for situations in which a mother harms but does not kill her child; given its logic, it would affirm a mother’s decision to intentionally take a medication that will cause birth defects in her child, for example. Second, the argument assumes that prenatal parental responsibilities are largely voluntary. Third, the analogies used to support the argument fail to take into account the difference between diseased and healthy physiological states. Fourth, the argument results in absurdities if taken to its logical conclusion. Taken as a whole, then, the bodily autonomy argument does not give us justification to jettison our deepest moral intuitions that mothers should not intentionally kill their offspring, whom proponents of this argument concede are rights-bearing individuals. Intentionally killing human fetuses in the act of elective abortion thus remains a great moral wrong.


A few years ago, I was preparing to perform third molar surgery under intravenous (I.V.) sedation on an adolescent girl, which is a common procedure in my clinical practice. As I began the I.V. line and started to administer the sedative medication, the patient mentioned something that distressed me greatly. She stated, “I guess I should tell you that I just found out I’m pregnant.”

There was an important reason for my concern. I had already administered midazolam, which is known to cause birth defects to a prenatal child when taken in the first trimester. I quickly gave the patient an agent that reverses the sedative effect of the midazolam, but does not remove the drug from her system. As she became more aware, I prepared to tell her the difficult truth that she received a drug that could cause harm to her unborn child. I expected this to be one of the more difficult discussions I have ever had with a patient, informing her that I may have caused harm to the child she was carrying.

I informed her, and she was quite upset, but for a different reason than I suspected. She was upset that I had not gone ahead and completed the procedure. In fact, she stated that she did not care that I had given her a medication that could have harmed her child. At first, I was quite alarmed by her attitude, but what she then told me helped explain her situation better. She told me that she was scheduled to have an abortion the following week. She returned a few months later (no longer pregnant) and I completed her surgery.

This situation presented a challenging ethical dilemma. One moment, I was very concerned for the child that I inadvertently may have harmed; yet, in the next moment, that concern was simply irrelevant. The child, in all likelihood, was going to be killed intentionally by another physician the following week. What struck me was that we were talking about the same human being. The only change that occurred from one moment to another was the knowledge that this child was unwanted by her mother.

Virtually all medical professionals who treat pregnant women acknowledge that there are two human beings of concern in these situations. For example, when they prescribe a medication, they realize that the drug affects both mother and child. Every drug handbook lists a medication’s FDA pregnancy category, which gives information about the potential harm to a mother’s fetus. A physician frequently needs to balance the best treatment for the mother with the safety of her child in mind. This intuitively is the most ethical course of action. There is an argument, however, that has been presented, and recently defended, in support of abortion rights that disputes this view. I term this the argument from bodily autonomy.


Most arguments concerning the abortion issue hinge on the moral status or standing of the fetus with regard to the rights he or she possesses and the obligations others directly owe him or her. These arguments typically fall along the following two lines. Pro-life advocates argue that all human beings, including those in the fetal stage of development, have intrinsic value that confers to them the right not to be unjustifiably killed. Pro-abortion choice advocates claim that the human fetus lacks some accidental quality (usually termed “personhood”) that affords it any rights or significant moral status.

The bodily autonomy argument, contrary to moral status arguments, does not focus on the “personhood” of the fetus. At least for the sake of argument, it concedes that the fetus is a human person with some degree of moral status. The bodily autonomy proponent argues that no human being, regardless of moral status, has the right to use the body of another human being against his or her will. The human fetus, then, does not have the right to use the body of his or her mother for sustenance or survival against her will. The mother who wishes to support her child by sustaining the pregnancy is performing a virtuous act, but one that she is not obliged to perform.

Bodily autonomy proponents thus view a pregnant mother who allows her child the use of her body as a “Good Samaritan,” particularly because performing such an act places a burden on her own body. A mother who seeks an abortion is doing so merely to retain autonomy over what occurs in her own body, and in this view is therefore justified. To accomplish this, the child, unfortunately, must be forcibly removed, thus resulting in its death.

This argument is well illustrated by philosopher Judith Jarvis Thomson in her famous violinist analogy. Thomson asks us to imagine a scenario in which a woman is involuntarily attached or connected to a “famous violinist” for nine months in order to save him from a fatal disease.[1] No one would argue that the violinist is not a valuable human being with a right to life, but it seems intuitive to most that the woman is not under a moral obligation to use her body to support him for those nine months. Proponents of the bodily autonomy argument believe that just as that woman is under no moral obligation to use her body to support the violinist, she is under no obligation to use her body to support a child she does not want.

If the violinist analogy holds, the pro-abortion choice advocate can concede what pro-life advocates have been trying to prove, and abortion still would be morally permissible. Pro-life apologist Greg Koukl declared when he first heard this argument, “It shook me up so much I almost had to pull over.”[2] Thomson’s analogy, as it originally was offered, was criticized roundly. Other scholars, however, recently have refined and defended Thomson’s idea that a mother’s right to bodily autonomy allows her to kill her offspring in order to remove the unwanted person and keep him or her from using her body against her will.

Legal scholar Eileen McDonagh defends Thomson’s view based on the legal concept of consent.[3] According to her view, a mother who does not consent to pregnancy has no obligation to continue to provide support for her offspring while it is using her body. The human person developing inside her is an unwanted threat to her, and it is permissible to defend herself against this threat by using deadly force.

McDonagh uses a variety of polarizing terms to explain the relationship between mother and fetal offspring. The fetus “intrude[s] massively on the body of another,”[4] “imposes wrongful pregnancy,”[5] and makes a woman “a captive samaritan by taking her body and liberty against her will to serve its own needs” (emphasis in original).[6] She describes a normal pregnancy in an interesting way when she states “the fetus does to a woman when it coerces her to be pregnant: namely, the fetus seriously injures her, even in a medically normal pregnancy, by forcing pregnancy on her against her will” (emphasis added).[7] Following in her footsteps, Philosopher Margaret Olivia Little describes a nonconsensual pregnancy as “the evil of unwanted occupation.”[8]

McDonagh argues that if a fetus is a rights-bearing individual, it strengthens her argument that abortion should be legal. She states, “The pro-life premise that the fetus is a person strengthens rather than diminishes a women’s right to an abortion and also to abortion funding” (emphasis in original)[9] In other words, if the unborn is a human person with intrinsic rights, as pro-lifers argue, then the state not only must allow abortions, but must also use public funds to pay for them.

Philosopher David Boonin has also made a significant contribution to this argument. Prominent pro-life philosopher Francis Beckwith called Boonin’s book A Defense of Abortion “arguably the most important monograph on abortion to be published in the last twenty years.”[10] Boonin defends the bodily autonomy argument (which he terms the Good Samaritan argument) by arguing that although the mother is responsible for the creation of her offspring, she is not responsible for the neediness of her offspring. In other words, since a mother bears no responsibility for the fact that she created a human person who is dependent on her for his or her life, she has no moral obligation to continue her support. He states, “The violinist’s right to life does not include or entail the right to be provided with the use or the continued use of whatever is needed in order for him to go on living.”[11]

Do mothers have the right to intentionally kill their offspring even though they are human beings with moral status and the right to life? I believe that the violinist analogy and the bodily rights argument that it illustrates fail for a number of reasons, which I will discuss in the following sections. I also believe that we can demonstrate that a mother’s right to control her own body does not override her obligation to sustain her unborn child’s body.


The bodily rights argument is compelling if and only if we grant that a woman’s right to control her own body is so sacrosanct that carrying another human being inside of her has no bearing on that right. In other words, for this view to prevail, we must concede that because of the autonomy she has over her body, a pregnant mother has the absolute right to do whatever she wants with it in order to retain that autonomy, regardless of what it does to the child she is carrying. This includes killing the child, in the case of elective abortion.

It is easy to demonstrate that the last paragraph is clearly false. Isotretinoin (Accutane) is a drug that is used to treat acne, but that causes severe fetal injury and birth defects.[12] The FDA restrictions for isotretinoin are so tight that before the medication can be dispensed, a woman of childbearing age must pledge to use two forms of contraception[13] if she is sexually active. Prior to filling the prescription, she also must verify the types of contraception she is on via the Internet or telephone[14] and take two pregnancy tests (one administered by her doctor, and one by a certified laboratory), both with negative results. She must use the most accurate tests available (never home pregnancy tests) to confirm that she is not pregnant.[15] We accept these as reasonable restrictions on a woman’s right to bodily autonomy in order to optimize the safety to her child. How, then, would we react to a pregnant patient who wishes to continue isotretinoin (Accutane) therapy for her acne despite her awareness that it causes severe fetal injury and birth defects?

Similarly, what about a pregnant mother who insists on taking thalidomide to treat her symptoms during the first trimester of pregnancy, despite her awareness of the harm it would do to her child? Thalidomide is a drug that was given to treat nausea and insomnia in pregnant women in the late 1950s and early ‘60s. It was never officially available in the United States, but it was taken by thousands of women in Canada and in countries in Europe and South America.[16] Soon after thalidomide was available, physicians began to notice an increase in severe birth defects, ranging from malformations of the ears to absence of the arms to phocomelia (hands [or sometimes feet] attached to abbreviated arms [or legs]).[17] Researchers have yet to discover a medication that they deem safe to treat nausea and insomnia for pregnant women that is as effective as thalidomide.[18]

It is likely that no one reacted negatively to the women who took this medication 45 years ago, since they had no idea that their children would be harmed. How would we react today, however, to a pregnant mother who acquired thalidomide even after her physician refused to prescribe it, and took it anyway, which resulted in her child being born without arms? Would we applaud her actions based on her right to bodily autonomy? According to the bodily autonomy argument, the fetus, after all, is an uninvited guest who has no right to use her body, let alone a right to a healthy or pathogen-free environment.

If the right of bodily autonomy is absolute, as it needs to be to defend the ultimate act of intentionally killing a human person, how could we fault the mother in this case? Which is worse: causing harm to a child or intentionally killing that same child? If a mother can kill a child because it is intruding on her bodily autonomy, then it is unreasonable to disallow her to harm the same child using the same reasoning.

The symptoms that thalidomide was meant to treat are a direct result of pregnancy. In fact, nausea and insomnia are symptoms that proponents of the bodily rights argument use to justify a woman’s decision to procure an abortion. Boonin lists “nausea” and “difficulty sleeping” in a list labeled “Physical Costs” in his response to the “different burdens” objection.[19] McDonagh lists nausea among the symptoms that she describes as a “serious injury” to the mother.[20] If it is permissible, however, for a mother to kill her unborn child in order to stop experiencing these symptoms, it ought to be permissible for her to take a medication such as thalidomide that would cause sub-lethal harm to her child in order to treat her symptoms, since, although the fetus would be harmed, he or she would not be harmed as much as in elective abortion. [

One may respond that a mother who agrees to allow a pregnancy to continue has an obligation to make the environment of the fetus as safe as possible. This is consistent with Boonin’s brief discussion of a parent as guardian.[21] Boonin argues that since a mother has elected to allow the use of her body by the child, she has an obligation as its guardian to not harm her child.

Let us examine this argument by assuming that the woman in Thomson’s analogy agrees to let the violinist use her body (a great kindness, according to Thomson). Two months into the treatment, the woman suffers horrific nausea that can be treated with a certain medication. Unfortunately, this medication carries the risk of harming the violinist, however, by making his hands unusable. Would it be ethical for the woman to take the medication? Does it make a moral difference that she volunteered to be the violinist’s life support system?

It seems there would be three options for the woman. She can suffer through the nausea, she can take the medication and possibly harm the violinist, or she can choose to detach or unplug the violinist, resulting in his death. According to bodily autonomy proponents, she has an absolute right to bodily autonomy; therefore she has no obligation to suffer through the vomiting. Of the remaining two options, then, who (other than bodily autonomy proponents) would state that it would be better for her to kill the violinist than to take a course of action that could cause possible harm to the violinist? Death, in fact, is the ultimate harm.  If the violinist could voice his opinion to the woman, he would surely choose the option that would preserve his life.



The bodily autonomy proponent assumes that prenatal parental responsibilities are largely voluntary. Proponents argue that because the mother is the only one who can provide a safe environment for the child, it is morally permissible for her to deny the use of her body even if such denial results in her child’s death.

Let me offer a thought experiment to challenge this argument. Suppose that a woman who faces an unplanned pregnancy decides to gift her child for adoption to another couple. In other words, she agrees to allow the child use of her body during the period of gestation but explicitly states that she is unwilling to care for the child after the birth event.

This mother takes a vacation in a cabin in the mountains when a freak snowstorm strikes and closes down all the roads in and out of the area for at least two weeks. The cabin has adequate food and water stores for the mother, but there is no baby formula, and there are no baby bottles or supplements available for a newborn child. As the storm strikes, the mother goes into labor and delivers a healthy baby girl.

The only way the newborn can survive is to feed on the milk that her mother’s breasts naturally provide. There is no formula to feed her, and no means to give the child hydration except for breastfeeding. Does the mother have any moral obligation to use her body (against her stated desire) to feed this child?[22] Per Boonin, although the mother is responsible for the existence of the child, she is not responsible for the child’s neediness or the circumstance that has placed that child in need,[23] despite the fact that the mother can easily fulfill that need in a natural, healthy way. According to Boonin, therefore, the mother appears to have no obligation to share her body with her own child, even if the baby girl dies from dehydration.

Suppose the mother also brought a young kitten with her to the cabin. The kitten would be in the same position as the baby girl. What if, instead of allowing her own child to drink her milk, the mother elects instead to give it to her young kitten? After all, she wants  the kitten, and she has already stated that she did not wish to care for the child after the birth. She reminds herself of the slogan “My Body, My Choice” as she watches her child die.[24]

If the authorities find her child dead from dehydration two weeks later, how would we judge her actions? What if we found the child dead, but the kitten alive, even well? Would we consider her actions powerful assertions of her right to autonomy, or see them as a morally unconscionable acts of selfishness? It would be very difficult for the mother to justify allowing her own child to die based on her desire to keep her body to herself. Further, granting that the mother does have an obligation to feed her child in this scenario  would indicate a weakness of her bodily autonomy rights in other situations. The right to bodily autonomy is not strong enough to override the moral obligation we have to our children.



Thomson’s analogy, in all of its forms, presents someone in a seriously diseased state. Her violinist develops a kidney ailment that threatens his life if not for the aid of another. This is the reason why he needed to be “hooked up” to the unsuspecting patient. Virtually all of Boonin’s analogies that directly address the bodily autonomy argument involve a serious pathology of some sort. Are these situations truly analogous to most pregnancies?

The difference between how we view physiologically healthy states and physiologically diseased ones is profound. For example, a mother who intravenously injects medications that cause profound effects such as nausea and vomiting into her healthy child is committing an unspeakably immoral act. However, the same mother performing the same action on her child who is suffering from leukemia is showing courageous virtue. The difference between these two cases is the presence of a diseased state, and that difference is essential.

The vast majority of pregnancies involve physiologically healthy situations. The woman’s body is functioning as designed. In fact, one may say that every unwanted pregnancy in some way occurs because a woman’s reproductive system worked too well. That is because physiological health functions independent of one’s desires. A person’s situation may not coincide with his or her wishes, but it still can be one of health. For example, someone who injects himself with narcotics for its euphoric effects may desire that the drug stay in his system as long as possible, yet his liver and kidneys remove the drug from his body. No one would claim that the person’s liver and kidneys are unhealthy for functioning independently of his desires.

John Wilcox has challenged the violinist analogy on the basis that pregnancies are natural.[25] Not only are pregnancies natural, they usually are healthy. We may not have an absolute obligation to use our bodies to support another human being who is in a pathological situation, but this does not compel us to deny a mother’s obligation to her offspring in an otherwise healthy situation. Pregnancies are usually completely healthy for both mother and child, so analogies that confuse pregnancy with disease states are not appropriate.



Given that bodily autonomy argument proponents concede at least for the sake of argument that the fetus is a rights-bearing, fully human individual, it is easy to demonstrate some odd consequences if we take the argument to its logical conclusion. For example, although we frequently speak of a woman’s right to terminate her pregnancy, all pregnancies “terminate” naturally at some point in time. In some sense, imposing our will via voluntary abortion changes the natural end to pregnancy. In other words, abortion dictates that the pregnancy will end when the woman desires it to.

Suppose, however, that the ending of pregnancy is a completely voluntary process and that births occur only via a voluntary action of the woman. If the woman does not agree to have the birth, the child continues to develop all of its normal capacities (awareness, etc.) but stays small in stature. The only way for the child to be delivered, detached, or “unplugged” from the woman’s body is through the woman’s body (thus violating her bodily autonomy) in a painful process similar to childbirth.

What if the mother does not give consent in this scenario? The child essentially, then, would continue to be bound by his or her mother for as long as the mother desires, even to his or her natural death.

If the mother’s right to bodily autonomy is absolute to the point that she can intentionally kill the developing child inside her, what moral principle would be available to compel this mother to consent to the birth of her child? It seems that if bodily autonomy proponents can use bodily autonomy to justify killing a child, even a child who is a full human being with a right to life, they would also use it to deny a child his or her liberty in this situation.  If the bodily autonomy proponent carries their view to its logical conclusion, then her child would always be a bound slave to their mother.



It is basic moral intuition, as well as a bedrock foundation of society, that parents have certain moral obligations to their children, especially when those children are vulnerable. Stories of parental abuse and neglect, as well as reports of intentional killing of children at the hands of a parent bring out some of the deepest moral disgust imaginable.

The proponents of the bodily rights argument for abortion have a gargantuan task in overcoming these deep moral intuitions. This is the reason they use such odd and distorted analogies. They compare innocent human beings who are at their most vulnerable to rapists who impose their will and force pregnancy on unsuspecting mothers. They compare the relationship of mothers and children together in one of the most natural and healthy human states with that of those who are in some of the most horrific pathological disease states.

I respect the work of these thinkers; I can’t help but conclude, however, that the extent that they need to stretch reality to justify their support of abortion rights indicates a clear weakness in their position. In the end, their arguments, though thoughtful, fail to overturn the truth that it is wrong to intentionally kill innocent human offspring.

Richard J. Poupard is a board-certified oral and maxillofacial surgeon in private practice in Midland, Michigan. He is a speaker for Life Training Institute (LTI) and a frequent contributor to the LTI blog.



  1. Judith Jarvis Thomson, “A Defense of Abortion.” Reprinted in The Abortion Controversy: A Reader, eds. Louis Pojman and Francis Beckwith (Boston: Jones and Bartlett, 1994), 113–27.
  2. Greg Koukl, “Unstringing the Violinist,” Stand to Reason,
  3. Eileen McDonagh, Breaking the Abortion Deadlock: From Choice to Consent (Oxford, England: Oxford University Press, 1996).
  4. Ibid, 9.
  5. Ibid, 10.
  6. Ibid, 11.
  7. Ibid, 89.
  8. Margaret Olivia Little, “Abortion, Intimacy, and the Duty to Gestate,” Ethical Theory and Moral Practice 2 (1999): 295–312.
  9. McDonagh, 13.
  10. Francis Beckwith, “Defending Abortion Philosophically: A Review of David Boonin’s A Defense of Abortion,” Journal of Medicine and Philosophy 31:2 (2006), 200.
  11. David Boonin, A Defense of Abortion (Cambridge, England: Cambridge University Press, 2003), 137.
  12. S. Food and Drug Administration, “Accutane (isotretinoin) Questions and Answers,” Department of Health and Human Services, FDA Center for Drug Evaluation and Research,
  13. Ibid, under “What must a patient do to get isotretinoin under iPLEDGE?”
  14. Ibid, under “How should female patients who can become pregnant who do not have access to the internet or a telephone access the iPLEDGE program monthly?” See also “The iPledge Patient Information Introductory Brochure,” The iPledge Program,
  15. S. Food and Drug Administration, “Accutane (isotretinoin) Questions and Answers,” Department of Health and Human Services, FDA Center for Drug Evaluation and Research, under “Can pregnancy testing be done using a home pregnancy test?” (Although beyond the scope of this article, this fact challenges the idea that women’s reproductive health decisions are only between her physician and herself.)
  16. Linda Bren, “Francis Oldham Kelsey: FDA Medical Reviewer Leaves Her Mark on History,” FDA Consumer Magazine (March–April 2001), available at
  17. Widukind Lenz, “The History of Thalidomide,” Extract from a Lecture Given at the 1992 UNITH (Union Nationale pour l’Insertion du Travailleur Handicapé) Congress, Thalidomide Victims Association of Canada, available at
  18. Thalidomide presently, however, is given to women who are not pregnant and is used to treat conditions such as multiple myeloma and erythema nodosum leprosum.
  19. Boonin, 239.
  20. McDonagh, 84–91.
  21. Boonin, 232–33.
  22. This is a modification of an example originally offered by Scott Klusendorf in “The Great Abortion Debate: Scott Klusendorf vs. Amber Dolman and Rob Silver,” audiotape available at Stand to Reason (
  23. Boonin, 168–88.
  24. I am indebted to Steve Wagner for many of the ideas presented in this paragraph gleaned in a personal communication with him.
  25. John Wilcox, “Nature as Demonic in Thomson’s Defense of Abortion,” The New Scholasticism 63 (Autumn 1989), 463–84.
Share This