In this paper, I argue that Thomson's famous attempt to reconcile the fetus's putative right to life with robust abortion rights is not tenable. Given her view, whether or not an abortion violates the fetus's right to life depends on the abortion procedure utilised. And I argue that Thomson's view implies that any late term abortion that involves feticide is impermissible. In particular, this would rule out the partial birth abortion technique which has been so controversial of late.
- right to life
- bodily rights
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
I have always found Judith Jarvis Thomson's arguments in “A defense of abortion” to be very persuasive.1 I have even been convinced of her most controversial conclusion, that the termination of a pregnancy which resulted from contraceptive failure is morally permissible, even if the fetus is a person with a full right to life. But recently, as I taught Thomson's article to a group of undergraduates for the nth time, it occurred to me that there is a worrisome gap in her argument. In particular, I have come to believe that it follows from Thomson's own principles that employing an abortion technique which does not merely remove the fetus from the womb, but which more directly kills it, is morally impermissible, even if removing it from the womb would have otherwise caused its death. If this is right, then Thomson must conclude that many abortion techniques commonly in use today are morally impermissible, at least when the pregnant woman's life is not at stake.
Thomson's defence of abortion rights proceeds as follows. Suppose someone wakes up to find that, while asleep, she has been attached to a famous violinist whose circulatory system has been plugged into hers. He needs to remain attached to the woman for nine months, and if unplugged in the interim, will die. Thomson argues that despite the fact that continued use of the woman's body is the bare minimum the violinist needs to survive, the violinist lacks the right to such use. After all, the woman did not consent to being used in this way. As a result, she would not violate the violinist's right to life by unplugging him. Thomson points out that this situation is relevantly analogous to one in which a woman has become pregnant, without consenting to being in such a state. And she concludes on this basis that the fetus does not have the right to continued use of the woman's body despite the fact that it is the bare minimum the fetus needs to survive, even if the fetus is a full moral person with a full right to life. The pregnant woman would not violate this right if she had an abortion in such circumstances.
One way to cause a person's death is to deprive her of the bare minimum she needs to survive. There are, of course, two ways to go about doing this: discontinuing a person's receipt of what she needs, and refraining from providing her with the same in the first place. So, for example, if Sarah needed to be hooked up to a respirator in order to survive, one could cause her death either by disconnecting her from the respirator after she had already been hooked up to it, or by refraining from connecting her to it in the first place.
According to Thomson, one violates a person's right to life by causing her death in this way only if she was entitled to the bare minimum she needed to survive. (Thomson argues that even if a person were not entitled to what she needed, it might still be morally reprehensible, albeit not unjust, to cause her death by depriving her of it. In particular, it would be reprehensible if providing for her needs would only require that one behave as a “minimally decent samaritan”. For present purposes, I am going to restrict my attention to cases in which much greater sacrifices are required). We never get from Thomson a precise account of the conditions which have to met in order for someone to be entitled to the bare minimum she needs, but there are certain tolerably clear cases:
Case # 1: Suppose Sarah needs to be hooked up to a respirator in order to survive, and suppose Sarah owns the respirator. In these circumstances, Sarah would be entitled to the bare minimum she needs to survive and to deprive her of it would be to violate her right to life.
Case # 2: Suppose Sarah needs to be hooked up to a respirator in order to survive, and suppose Fred owns the respirator. Suppose that Fred freely enters into an agreement with Sarah to provide her with the use of the respirator for as long as she needs it. In these circumstances, Sarah would be entitled to the use of the respirator, at least as long as she keeps up her side of the bargain.
Case # 3: Suppose Sarah needs to be hooked up to a respirator in order to survive, and suppose Fred owns the respirator. Suppose that Fred has entered into no agreement with Sarah to provide her with the use of the respirator, and has engaged in no behaviour which could constitute tacit consent to her use of it. In such circumstances, Sarah would not be entitled to the use of the respirator, and even if depriving her of it caused her death, doing so would not violate her right to life.
The hard cases are those in which the owner of the needed goods in some sense tacitly consents to their use by those in need. It's not entirely clear whether Sarah would be entitled to the use of Fred's respirator if, for example, Fred carelessly left it on Mary's front lawn for a couple of days and Sarah proceeded to connect herself to it in the interim. But I do think that Thomson has made a good case for thinking that a woman who gets pregnant as a result of contraceptive failure has not given tacit consent to the fetus for use of her body, and hence the fetus is not entitled to such use.
Things are, as always, however, more complicated than they seem at first glance. A person typically (always?) needs many things to survive, and to deprive someone of any one of these things would cause her death. Suppose, for example, that Sarah needed to be hooked up to both a respirator and a dialysis machine in order to survive. In such circumstances, one could cause Sarah's death either by depriving her of the use of the respirator or by depriving her of the use of the dialysis machine.
As above, one violates Sarah's right to life by causing her death by one of these means only if she was entitled to the use of the machine of which she was deprived. Now it might turn out that Sarah is entitled to the use of one of the machines but not the other. After all, she might own the dialysis machine while Fred owns the respirator (and has made no agreement with Sarah allowing her to use it). Or Fred might own both machines, but has agreed to allow Sarah to use only the dialysis machine. And if either scenario were true, causing Sarah's death by depriving her of the dialysis machine would violate her right to life, whereas causing her death by depriving her of the respirator does not violate her rights.
Now let's suppose that Sarah is currently hooked up to both the respirator and the dialysis machine, and if either machine is disconnected, she will die. As before, let's suppose Sarah is entitled to the use of the dialysis machine but not entitled to the use of the respirator. And let's suppose that Fred decides to retrieve the respirator, denying Sarah continued use of it, as he is well within his rights to do. (As before, we'll assume that allowing Sarah continued use of the respirator would require more of Fred than minimally decent samaritanhood.) Broadly speaking, there are four procedures (of interest, for present purposes) by means of which he (or his agent) might go about doing this:
Procedure # 1: First, Fred detaches Sarah from the respirator. (Fred leaves her connected to the dialysis machine.) Second, Sarah dies.
Procedure # 2: First, Fred detaches Sarah from the respirator. Second, he detaches her from the dialysis machine. Third, Sarah dies.
Procedure # 3: First, Fred detaches Sarah from the respirator. Second, Sarah dies. Third, Fred detaches Sarah from the dialysis machine.
Procedure # 4: First, Fred detaches Sarah from the dialysis machine. Second, he detaches her from the respirator. Third, Sarah dies. (Note: it does not matter if the order of steps two and three is reversed.)
What I want to do now is to consider a number of cases. These cases differ in the procedures available to Fred. I am, for now, going to assume that none of these procedures will be more or less beneficial or detrimental to Fred's health (or general wellbeing) than any other, and that his life and/or health will not be at risk if he fails to retrieve the respirator.
Case # 1: All four procedures are available to Fred. Nothing is preventing Fred retrieving his respirator by means of any procedure he chooses.
In these circumstances, it is fairly clear that Fred ought to use procedure # 1 to retrieve his respirator. The use of procedures # 2 and # 4 would violate Sarah's right to life and so ought to be avoided. Procedure # 3 would not involve a violation of Sarah's right to life because she would be detached from the dialysis machine only after she was dead. But Fred does not need to disconnect Sarah from the dialysis machine in order to retrieve the respirator (procedure # 1 is available after all). And there may be other moral grounds for refraining from doing so, such as its being disrespectful of the dead.
Case # 2: Procedure # 1 is unavailable to Fred. Procedures # 2–4 are available. He can detach Sarah from the respirator without detaching her from the dialysis machine, but he cannot retrieve the respirator without detaching her from the dialysis machine as well. (Let's suppose that the tubes and wires are inextricably tangled).
In these circumstances, it is fairly clear that it is permissible for Fred to use only procedure # 3. Once again, procedures # 2 and # 4 would violate Sarah's right to life and should be avoided. One might argue that since procedure # 3 is permissible, procedure # 2 is permissible as well. After all, the only difference is that you kill someone who is going to die very soon in any event. But this is akin to claiming that the deathbed homicide of a rich heiress would be permissible because she is, well, on her deathbed.
Case # 3: Procedures # 1–3 are unavailable. Only procedure # 4 is available.
In these circumstances, it would be impermissible for Fred to retrieve his respirator from Sarah. To do so would violate her right to life and, other things being equal, a right to life outweighs any property right. Now it could be argued that the claim that the right to life outweighs property rights entails that property could be taken away from people to save lives. One might resist by claiming that it is only the negative right to life that outweighs property rights. But even if this manoeuvre is ultimately unsatisfactory, this implication should not worry us too much. After all, it would come into play only if the only way to save lives involved seizing private property. Moreover, considerations of justice would presumably prevent too great a burden in this regard from falling on any single property owner.
It is worth noting how this relates back to the issue of abortion. Fred is intended to be analogous to the pregnant woman and the respirator is analogous to the pregnant woman's womb. Sarah is meant to be analogous to the fetus and the dialysis machine is analogous to the internal organs of the fetus. The fetus needs both continued use of the pregnant woman's womb and continued proper functioning of its organs in order to survive. I in no way mean to suggest that the pregnant woman's womb is in any sense property of the fetus.
But now let's suppose that these procedures are not equal with respect to Fred's wellbeing: some of them are more detrimental to Fred's health than others. To make this concrete, let's suppose that Fred's respirator is a double and that both he and Sarah are hooked up to it. Moreover, let's suppose all four procedures are available to Fred, but (a) procedure # 1 is complicated and likely to damage the respirator and (b) the only way to implement procedure # 3 would involve shutting down the respirator, which would likely result in Fred's death as well as Sarah's.
The first thing to note is that unless Fred's life or health is at serious risk, none of this makes any difference. Remember, Fred has a fifth choice: he can always leave Sarah attached to the respirator. And unless leaving Sarah attached poses a serious risk to Fred's life, the fact that procedures # 1 and # 3 would pose a serious risk does not give good ground for using procedure # 2 or # 4, and thereby violating Sarah's right to life.
But what if leaving Sarah attached to the respirator did pose an imminent risk to Fred's life? In my view, Thomson has successfully defended the thesis that one can in self defence violate the right to life of an innocent person. (If one finds the notion of a morally permissible violation of rights unpalatable, we can speak instead of the performance of an act that in ordinary circumstances would involve a violation of rights.) And if this is right, then it would clearly be permissible for Fred to use procedure # 2 or # 4 to disconnect Sarah, given the risks involved in using either of the other procedures, or in leaving Sarah attached to the respirator.
Consider now the procedure used in intrauterine cranial decompression abortions (or ICD abortions), popularly, and notoriously, referred to as “partial birth” abortions.
1) the cervix is dilated;
2) the fetus is partially removed from the womb, feet first;
3) a sharp objected is inserted into the back of the fetus' head and then removed;
4) a vacuum tube is inserted through the resulting hole and the brain is removed;
5) the head of the fetus contracts, allowing the fetus to be more easily removed from the womb, and the fetus then is removed.
Clearly, this procedure is a variation of procedure #4 discussed above. The fetus needs both the use of the pregnant woman's body and a functioning brain in order to survive. And while depriving the fetus of continued use of the woman's body would not violate its right to life (unless the woman has in some way consented to this use), depriving the fetus of a functioning brain would constitute a violation of its right to life. After all, if anything is the property of a person, his or her brain is. (If one objects to taking a person's bodily organs to be her property, the argument can be recast in terms of a right to bodily integrity. While this would make the analogy with the case of Fred and Sarah weaker, the argument would still retain its plausibility.) And so the ICD procedure deprives the fetus of something which the fetus needs and to which it is entitled, before depriving it of use of the woman's body to which it lacks entitlement. The upshot is that, if we accept Thomson's defence of abortion, the ICD abortion technique is morally impermissible unless the life or health of the pregnant woman is a serious risk and any other alternatives pose a serious danger as well.
But suppose that the fetus is going to die anyhow and available alternative procedures subject the pregnant woman to greater risks. In such circumstances utilitarian considerations would incline in favour of the ICD procedure. There are two comments I wish to make regarding this point. First, at best it would permit only limited use of the ICD procedure. And second, it is far from clear that utilitarian considerations justify violating the right to life of even a dying person.
Now one might argue that this is all moot: given that ICD abortions are performed late in pregnancy (during the fifth month of gestation or later), Thomson's own commitments imply that it is impermissible. There are (at least) three ways one might argue from Thomson's commitments to the conclusion that ICD abortions are impermissible. First, one might argue that by refraining from having an abortion for five months, the pregnant woman has tacitly consented to the fetus's continued use of her body. And hence, an abortion by any means would be impermissible. But even if you think a woman could give tacit consent in this way, there are obvious cases in which a pregnant woman who waits at least five months before having an abortion clearly has not done so. Consider, for example, a poor woman in a state in which there is no public funding for abortions, there are no nearby abortion clinics, and there is a 24 hour waiting period for having an abortion. Refraining from getting an abortion because your circumstances make it very difficult for you to do so could not plausibly be considered to suffice for tacit consent of the relevant sort.
Second, one might argue that even though ICD abortions do not involve a violation of the fetus's right to life, once one is in the late stages of pregnancy, carrying a fetus to term requires only minimally decent samaritanism of the pregnant woman. Thomson says: “[it] would be indecent in the woman to request an abortion, and indecent in the doctor to perform it, if she is in her seventh month, and wants the abortion just to avoid the nuisance of postponing a trip abroad”.2 And, hence, an abortion by any means at this late date would be impermissible (or at least indecent). The trouble here, of course, is that it is far from clear that carrying a fetus to term for as long as four months requires only minimally decent samaritanism of the pregnant woman. And, of course, it is far from clear that it requires only minimally decent samaritanism of women who have more serious reasons for wanting an abortion, even if continuing the pregnancy does not pose a serious risk to life or health.
And third, Thomson argues only for a right to have an abortion and not for a right to secure the death of a fetus.2 And so as long as there is an alternative procedure which would allow the fetus to survive outside the womb, an ICD abortion is impermissible. And given that ICD abortions occur late in pregnancy, there is an alternative procedure: remove the fetus by caesarean section and place it in an incubator. But what Thomson fails to note is that even if there is no procedure which would allow the fetus to survive outside the womb (and, arguably, at five months there is no procedure which would reliably do so), or if the only such procedure subjected the pregnant woman to serious risks, an ICD abortion would still be impermissible (unless, of course, continuing the pregnancy was, itself, to risk serious harm).
The upshot of all of this seems to be that, if one accepts Thomson's defence of abortion, how an abortion is performed makes a moral difference. If the pregnant woman's life is not at stake, then, no procedure which causes the fetus's death by any means except its removal from the woman's body is morally permissible. And, as we have seen, this implies that the ICD abortion procedure is morally impermissible. Moreover, similar considerations rule out all late abortions involving feticide. Now, of course, such restrictions are simply unacceptable to people who want to defend robust abortion rights. But I think it is clear that Thomson's strategy will not yield a ground for such a robust right. Unless some other approach can be found for reconciling the fetus's putative right to life with robust abortion rights, the personhood issue must be faced head on.