1. Different questions and a range of views

The fundamental question about abortion that I propose to discuss should come as no surprise. It’s this:

Question 1:

Is it e​ver overall morally wrong to have or perform an abortion? If so, under what circumstances, and why?

Thi​s question has been answered in a wide variety of ways, some of which we’ll be examining. These answers can be placed along a kind of spectrum, from conservative, through moderate, to liberal.

The most conservative answer to Question 1 is this: It is overall morally wrong under all circumstances to have or perform an abortion. I don’t think there are many people who embrace this answer. Almost all people are willing to countenance abortion under at least some circumstances, such as, perhaps, when pregnancy puts the mother’s life at risk, or when pregnancy is due to rape. Still, there are many people who give an answer that is almost as conservative: It is overall morally wrong under all but very exceptional circumstances to have or perform an abortion (the exceptions being of the sort that I have just mentioned). Of course, there are also people who would give an answer that is somewhat less conservative, one according to which the circumstances in which abortion can be justified need not be quite so exceptional.

The most liberal answer to Question 1 is this: It is overall morally wrong under no circumstances to have or perform an abortion. I know of no one who accepts this answer, and for good reason. Just think of performing an abortion on a fully rational adult woman who doesn’t want one! It would seem to be extremely difficult to justify that. Still, there are of course somewhat less liberal answers that have been accepted, according to which having or performing an abortion can be overall morally justified under a wide variety of circumstances.

Somewhere between positions that may appropriately be called conservative and those that may appropriately be called liberal are positions that may appropriately be called moderate, according to which abortion can be overall morally justified under a relatively limited set of circumstances. There is no sharp division between these three kinds of position: they shade into one another. The labels that I have used (“conservative,” “moderate,” “liberal”) are therefore not ideal (although they are surely much better than the common-but-silly, simplistic, politically-charged labels “pro-life” and “pro-choice” − as if conservatives were “anti-choice” and liberals “anti-life”!); they are supposed to be suggestive, but they should not be relied on too heavily and are no substitute for spelling out precisely whatever the particular position is to which they are being applied.

It is important to distinguish between Question 1 and another question that is currently hotly debated, namely:

Notice that answers to this question can also be placed along a spectrum from conservative, through moderate, to liberal, the most conservative being that abortion ought always to be prohibited, the most liberal being that it ought never to be prohibited.

It is common to hear people saying something like this: abortion is wrong, and so it ought to be prohibited by law. This argument can be put as follows:

By now, I hope you know what we should say about an argument of this sort: it’s hopeless! (You might want to compare what I have to say here about what I had to say about Arg. 6 in the section on homosexuality.) First, Arg. 1 is invalid. Second, it fails to specify the circumstances under which it is supposed to be overall morally wrong to have or perform an abortion or overall morally obligatory to prohibit such behavior. Imagine that someone who gives this argument has “all circumstances” in mind in each case. Then, in order to handle the first problem, the argument can be re-cast as follows:

But, for reasons I gave when discussing homosexuality (and which I won’t repeat here − I invite you to take a look back, if need be), the second premise of this argument is highly problematic, regardless of what should be said about its first premise. Indeed, although it is common to find people arguing for the prohibition of abortion on the basis of its (alleged) immorality, it’s also quite common to hear people arguing that abortion ought to be permitted by law, despite its (alleged) immorality. (Why? Perhaps on the basis that the law ought not to intrude into private matters even to prevent wrongdoing.) In any case, the fact is that, although answers to Questions 1 and 2 often do go “hand in hand” (in the sense that people often give a conservative answer to both, or a moderate answer to both, or a liberal answer to both), they don’t have to. That is, there is nothing inconsistent in giving a conservative answer to one but a liberal answer to the other, and so on.

I will not pursue Question 2 further, interesting and important though it is. Again, it is Question 1 that I want to focus on.

***Here is a good place to stop and consider Question 1 on the Practice Questions for Module 4.***

There is, however, a third question in this context that we should acknowledge, one that concerns the so-called right to life. I will be assuming that a right to life is, more particularly, a right not to be killed; it is a claim-right that one person has against other people that they not kill him or her. (Regarding claim-rights, see the earlier section on moral obligation, right, and wrong.) The question is this:

Answer​s to this question, too, may be placed along a spectrum, the most conservative being that the fetus has a right to life at all stages of its development, the most liberal being that it has such a right at no stage of its development. We will be spending some time on this question, since many believe that the answer that should be given to it is highly relevant to the answer that should be given to Question 1.

(Side-note: I have just talked about “the fetus.” I have in mind human fetuses in particular, of course, since − as I assume you were assuming we were assuming − we’re talking about abortion in humans as opposed to abortion in cats, for instance, or dogs, or cows. Even so, I am using the term “fetus” rather liberally to refer to the individual that exists from conception until birth. This liberal use is a common one, although technically the term “fetus” refers to the post-embryonic individual that begins developing at around the ninth week of pregnancy.)

2. Rig​hts

The general concern in this course is with whether it is overall morally right or justifiable, or overall morally wrong or unjustifiable, to act in certain ways under certain circumstances. Sometimes the answer seems to turn on whether anyone has rights that would be respected or infringed if we were to act in the proposed way. (Whenever I mention “rights,” you should understand me as being concerned with moral rights in particular − rather than, say, legal rights.) This is certainly often assumed to be the case with abortion, as Question 3 indicates.

Some people appear to think that, if acting in some way would infringe a right, then so acting would therefore be overall morally wrong. Consider, for example, this simple argument for the extreme conservative position with respect to Question 1 (where, as just noted, we are restricting our attention to abortion in humans and the term “fetus” is being construed liberally):

This argument is of course invalid, but that is a problem that is easily fixed. One way to fix it would be as follows:

What should we say about this argument?

First, unlike Arg. 3, this argument is indeed valid. That is, the first conclusion, (3), follows from premises (1) and (2), and then the second conclusion, (6), follows from premises (3), (4), and (5). (Notice that (3) is doing double duty here, functioning first as a conclusion and then as a premise for another conclusion.) If you don’t immediately see that Arg. 4 is valid, I hope that a little reflection will reveal this fact to you.

Second, I should amend the claim that I have just made. Arg. 4 is valid unless there is equivocation on any of the terms involved. In fact, there is reason to suspect that equivocation is present here. Consider the first part of Arg. 4:

		(1) All humans have a right to life.
		(2) All fetuses are human.
	∴	(3) All fetuses have a right to life.

What should we say of the first premise? Doesn’t it seem true? Most people believe that they have a right to life. I assume that you are such a person. But what makes you so special? Well, perhaps you are special in a way, a way in which, for example, dandelions are not. That’s why you have a right to life and they don’t. But even if so, you don’t seem to be any more special in this way than, say, me. That’s why I have a right to life, too (as I hope you are aware).

(Side-note: Perhaps a right to life, like other rights, is one that someone can either waive or forfeit, in which case that person won’t have such a right after all. (We’ll take a brief look at the possibility of waiving a right to life when we discuss euthanasia later in the course, and we’ll take a brief look at the possibility of forfeiting such a right when we discuss capital punishment.) If so, then premise (1) of Arg. 4 should be amended as follows: All humans have a right to life, unless they have waived or forfeited it. But this is a complication that we needn’t worry about in this context. What we’re concerned with here is whether fetuses qualify as having a right to life. If they do, it’s surely not one that they have either waived or forfeited.)

Now, what about the second premise of Arg. 4? Doesn’t it seem true, too? After all, as I’ve noted, we’re talking about abortion in humans, and of course what that means is that we’re talking about the abortion of human fetuses. So, premise (2) is true, too.

But then it seems we must conclude that all (human) fetuses do indeed have a right to life, since (3) follows from (1) and (2). But can it really be that easy? Question 3 is the subject of intense debate, yet this simple argument seems to resolve that debate in favor of the most conservative answer to that question.

No, it’s not that easy. Each of premises (1) and (2) does indeed seem very plausible, but only, I think, when the term “human” is used in different senses.

Consider premise (1) again. The claim that all humans have a right to life (unless they have either waived or forfeited it) seems plausible, given that the word “human” is being used to refer to persons. You are a person; I am a person; a dandelion is not a person. Is a fetus a person? Well, that’s something we need to discuss.

Now consider premise (2). Of course, the sort of fetus we’re concerned with is a human fetus, in the biological sense of “human” that refers to being a member of the species homo sapiens. Do all humans in this sense have a right to life? That again is something we need to discuss.

The point that I am making here may be put as follows. Let’s use “human_P” to refer to the person-sense of “human” and “human_B” to refer to the biological sense. Then we can distinguish these three versions of the argument we’re considering:

(A)		(1) All humansP have a right to life.
		(2) All fetuses are humanB.
	∴	(3) All fetuses have a right to life.

 

(B)		(1) All humansP have a right to life.
		(2) All fetuses are humanP.
	∴	(3) All fetuses have a right to life.

 

(C)		(1) All humansB have a right to life.
		(2) All fetuses are humanB.
	∴	(3) All fetuses have a right to life.

 

Notice the following. We have granted that both premises of (A) are plausible. The trouble is that, given the shift in senses of “human,” that argument is not valid. To ensure validity, then, we must use the same sense of “human” throughout. Both (B) and (C) ensure this. But then the trouble is that each of these arguments has a premise that is highly controversial and in need of further support. In the case of (B), it is premise (2) that requires investigation; in the case of (C), it is premise (1) that does so.

We will undertake such investigation shortly. At the moment, though, let me direct your attention to the rest of Arg. 4. Suppose that we have somehow arrived at the conclusion that (3) is indeed true, that is, that it is indeed the case that all fetuses have a right not to be killed. We still cannot get to the final conclusion, (6), without the aid of further premises, such as (4) and (5). Let’s now take a look at them.

Consider (4) first. Is it the case that abortion always constitutes the killing of a fetus? (Perhaps I should say explicitly here that I intend our discussion, as I again assume you were assuming we were assuming, to concern the intentional abortion of a fetus − by a physician, say − and not the sort of spontaneous abortion that is sometimes called a miscarriage.) If a physician (successfully) performs an abortion, does he kill the fetus involved? Some may balk at saying so. After all, you cannot kill something that isn’t alive, and the question when life begins is hotly contested. But here I want to propose a short-cut. The question when life begins is, in the present context, more particularly the question when human life begins. Now, as we have just seen, the word “human” is ambiguous. And so we must distinguish between two interpretations of the question. One is this: When does human_B life begin? The other is this: When does human_P life begin? If we agree that all persons have a right to life, then the question on its second interpretation is pretty much the same question as Question 3. Here I want to postpone our discussion of this question. We’ll address it at some length later. But that of course still leaves the question on its first interpretation: When does human life in the biological sense begin? Well, I am not a biologist and the matter is not straightforward, but there seems to be very good reason to answer: Human life in the biological sense begins at (or very close to) conception. After all, that point does indeed appear to be the point at which an organism begins to develop. But even if there is reason to contest this claim, I want to grant it here. Notice that granting the claim does not serve to prejudge the morality of abortion in any way. Describing abortion as killing, while it may sound nasty, should not be misconstrued. After all, when you pick dandelions, you kill them, but presumably there’s nothing inherently nasty about that. We have not yet been given any reason one way or another to think that killing fetuses is any more problematic, morally, than killing dandelions. On the contrary, that is precisely the question that we’ll be investigating.

So, I propose that we grant that human_B life begins at conception and that abortion therefore kills the fetus. If the fetus has a right to life, as (3) states, then we must grant premise (4). (Again, that is of course a big “if,” one that we’ll be looking into.) Even so, we still cannot get to (6) without also granting (5). Should we accept this premise? Should we agree that infringing a right is overall morally wrong under all circumstances?

It’s important to recognize that the answer to this question is No. Although this appears often to be ignored or overlooked, the fact is that, even though (given the Correlativity Thesis mentioned in the section on moral obligation, right, and wrong), infringing a right is always prima facie morally wrong, it can nonetheless sometimes be overall morally right or justified. Here’s an illustration.

Suppose that you and I are attending a meeting along with several others. Suddenly, Betty has a seizure. We all know what is wrong with her and that, unless she receives a certain medication very soon, the seizure will prove fatal. We all also know that, as it happens, I have just the right medication in my briefcase and that I do not need it for myself. “Quick,” you say, “give Betty one of those pills!” I refuse. You’re astonished. “Why won’t you give her one of your pills?” you ask. “Because they’re my pills,” I say prissily, “and I want to keep them for myself.” You can’t believe what you’re hearing. You suspect I’m joking, although why I would do so at such a time of crisis you cannot imagine. But it quickly becomes apparent to you that I’m dead serious. So you snatch my briefcase from me, grab one of the pills, and give it to Betty. She soon recovers. Everyone cheers − except me. “You stole my pill!” I protest indignantly.

Were you overall justified in acting as you did? Surely you were. My refusal to give Betty one of my pills was unjustified, as was my subsequent complaint. Nonetheless, my complaint was accurate. You did indeed steal one of my pills − you took it without my permission − and you thereby infringed a right of mine, a property right. That fact notwithstanding, you did the overall morally right thing under the circumstances. Hence premise (5) of Arg. 4 is false.

Someone might agree with what I have just said and suggest replacing premise (5) with a more qualified claim, one that would avoid the sort of example I’ve just given but still enable us to get to the final conclusion, (6). The qualification I have in mind is this:

(5*) Under all circumstances, infringing a right to life is overall morally wrong.

This is an interesting move. Certainly, arguing from (3), (4), and (5*) to (6) is valid. Moreover, the pill example involves only the infringement of a property right and not of a right to life. Nonetheless, the example does serve to show that we cannot simply assume that all infringements of a right to life are overall unjustified. Whether such an infringement can be overall justified is a question to which we’ll return.

Still, we can say this much even at this point. Presumably a right to life is one of the strongest rights someone can have. Given the Correlativity Thesis, what that means is that the corresponding prima facie duty or obligation not to infringe such a right is likewise very strong. Thus, if there ever are circumstances under which it nonetheless would be overall morally justifiable to infringe such a right, it seems that these circumstances would have to be truly exceptional, circumstances that involved some competing consideration that was so strong that it overrode the very strong prima facie obligation not to infringe it. Given this, it seems sensible to inquire whether the fetus does have such a right and, if so, at what stage of its development it acquires this right.

So, although the question of ultimate concern is Question 1, let us now consider Question 3.

***Here is a good place to stop and consider Question 2 on the Practice Questions for Module 4.***

3. A challenge to non-conservativ​es on Question 3

My first remarks concerning Question 3 will be brief. They have to do with a challenge to anyone who is inclined to give a non-conservative answer to this question. Once the challenge has been stated, I will put Question 3 aside for a while, but we will certainly return to it later.

Before I state the challenge, let me make note of a very important point. If someone has a right to something (whether a right to life, or a right to some piece of property, or whatever), this will always be because of some other fact about or feature of that person. This fact or feature may be said to be the ground of the right; this ground will serve to explain how and why it is that the person has the right in question. Of course, in any particular case, a key question will be what the ground is. So, in the case of someone’s having a right to life, we must ask what it is that grounds or explains his or her having this right. As we’ll see, this is not at all an easy question to answer. Nonetheless, we can all readily accept that there is in principle some answer to be given. After all, if you and I are not only alive but have a right to life, whereas dandelions, though also alive, have no such right, there must be something that explains this fact, something that we have but they lack.

Now I can state the challenge in question. It is this. Do you agree that not only you and I but also newborn babies have a right to life? Probably. But if so, what explains this fact? Moreover, how can it be that the explanation of this fact doesn’t cover just you and me and newborns but also fetuses? Think of the matter this way. Suppose we had a film of the development of a fetus from conception through birth. Each frame of the film captures the fetus at a certain stage of its development. Let’s suppose that there are a million such frames. Frame 1 captures Stage 1, which is the moment of conception. Frame 1,000,000 captures Stage 1,000,000, which is the moment of birth. Now imagine rewinding this film slowly, stage by stage. At what stage is it that the fetus acquires a right to life? If you are a non-conservative, you deny that it has this right at Stage 1. But (I am assuming) you agree that it has this right at Stage 1,000,000. So there must be a stage somewhere between these two stages at which it acquires a right to life. But which? Which stage could possibly be so significant that it could account for this acquisition? Stage 536,398 perhaps? But why on earth should it be that stage rather than some other one that “does the trick”? It looks like picking any stage as “the” stage that does the trick would be wholly arbitrary, since it seems inconceivable that what happens at that stage is so different from what happens at the stage just prior to it that it is at that stage, and not the one just prior to it, at which the fetus acquires a right to life. But then no stage after Stage 1 can do the trick. If so, the fetus must have a right to life at conception.

The argument contained in the challenge just made can be put more formally as follows:

The rationale underlying premise (2) is this. If the fetus at Stage 1,000,000 has a right to life, then, given premise (1), so too does the fetus at Stage 999,999. But if the fetus at Stage 999,999 has a right to life, then, given premise (1), so too does the fetus at Stage 999,998. But if…well, you get the idea. What we wind up with is the fetus having a right to life at Stage 1, if it has a right to life at Stage 1,000,000. So premise (2) is fine. But does the fetus at Stage 1,000,000 have a right to life? Premise (4) says that it does, and I’ve been assuming that you agree. Since Arg. 5 is valid, the only way to resist its conclusion, as of course any non-conservative on Question 3 wishes to do, is to find fault with its first premise. But where could this premise go wrong? Where, when rewinding the film, could you find any point at which you would be justified in pushing the pause button and saying, “There! That’s the cut-off point. That’s where the right to life begins!”? That’s the challenge.

4. Marquis ​on abortion

Perhaps because he doesn’t want to get embroiled in the complicated question whether and, if so, at what stage a fetus has a right to life, Marquis takes an approach to the morality of abortion that is quite different. In short, he seeks to provide an answer to Question 1 without first finding an answer to Question 3.

Marquis begins by asking why it is seriously prima facie morally wrong (and so overall morally wrong in what I will call “unexceptional” or “typical” or “normal” circumstances) to kill other adult human beings like you and me. What best explains this fact? His answer: killing people like you and me is seriously prima facie morally wrong because it deprives them of a certain kind of future, a particularly valuable kind, a future that they would have enjoyed if they had continued to live. Human beings typically lead lives that are rich and varied, and this makes their lives much more valuable than the lives of, say, dogs or cats, although the lives of such animals are of course also valuable to some extent. Killing a dog or cat is not a trivial business, but killing another human being is a much more serious business, precisely because doing so cuts them off from, prevents them from enjoying, the kind of valuable experiences that (as far as we know) only humans can have.

But then Marquis notes the following: the fact just mentioned, the one that explains why it is seriously prima facie morally wrong to kill human beings like you and me, applies with equal force to the killing of human fetuses. After all, killing them likewise cuts them off from, prevents them from enjoying, the very same kind of valuable experiences that they would have enjoyed if they had continued to live.

Conclusion: abortion is seriously prima facie morally wrong. (See p. 68, right column – p. 69, left column; p. 70, right column.)

Let us state the moral principle that Marquis invokes and relies on as follows:

His argument may now be put more formally as follows:

This is an argument of Type III − an argument from inference to the best explanation. Again, the underlying idea is simply that what best explains why it is wrong to kill people like you and me also implies that it is wrong to kill fetuses.

Notice that the conclusion of Arg. 6, while pretty conservative, is not the most conservative answer that could be given to Question 1. As we have seen, the most conservative answer is: It is overall morally wrong under all circumstances to have or perform an abortion. Marquis doesn’t say this. He says that abortion is seriously prima facie morally wrong, so that it is indeed overall morally wrong under what I have called unexceptional circumstances, but that of course leaves open the possibility of its not being overall morally wrong under exceptional circumstances. Marquis doesn’t commit himself here. He doesn’t try to specify just which circumstances might count as exceptional ones under which the serious prima facie wrongness of abortion might be counteracted by some even more serious countervailing consideration in favor of abortion. He simply leaves the matter open.

Of course, a very important premise in Arg. 6 is premise (2). Clearly, if there is a principle other than M1 that better explains why it is wrong to kill you and me, then we will have to see whether it applies with equal force to human fetuses. If it doesn’t, then Marquis’s strategy will have been undermined. Marquis is fully aware of this aspect of his argument and accordingly takes a look at a couple of moral principles that might be thought to be superior to M1. These are what I will call principles M2 and M3 and may be stated as follows:

M2:	It is seriously prima facie morally wrong (and so overall morally wrong under unexceptional circumstances) to fail to satisfy strong desires.

 

M3:	It is seriously prima facie morally wrong (and so overall morally wrong under unexceptional circumstances) to discontinue the experience of living.

 

Let’s consider these in turn.

Someone might think that M2 provides a good explanation of why it is wrong to kill people like you and me. After all, you and I have a strong desire to continue to live, and killing us would frustrate that desire. Notice, moreover, how this principle does not apply to all fetuses. It’s unclear just when it is that fetuses become capable of having desires, but presumably early-stage fetuses have no desires at all, and it’s doubtful whether fetuses at any stage of their development have a desire to continue to live in particular. This being the case, M2 would not cover abortion in the way it covers killing you and me, and so it would not serve Marquis’s strategy well. And so, if M2 is superior to M1, then premise (2) of Arg. 6 is false, the argument is therefore unsound, and, more generally, Marquis’s entire strategy is undermined.

It won’t come as a surprise to you that Marquis denies that M2 provides a better explanation than M1 does of why it is wrong to kill people like you and me. He says that M2 in fact fares poorly in this regard because, even if you and I have a strong desire to continue to live, there are many people (such as those who are unconscious, or sleeping, or depressed, or suicidal) who do not, and yet it is presumably just as wrong to kill them as to kill you and me. M1 accounts for this fact, whereas M2 does not. Hence M1 is superior to M2, rather than the reverse. (See p. 72, right column.) This seems quite right.

As for M3: notice how, once again, it would appear to apply to you and me but not to all fetuses, since fetuses at an early stage of development presumably are incapable of experiencing anything at all. And so, once again, if M3 were superior to M1, Marquis’s project would be doomed.

As before, though, Marquis claims that M3 is inferior rather than superior to M1 as an explanation of why it is wrong to kill people like you and me. When it comes to cutting people off from their futures, it is surely the value of their future experiences that counts, and not just the fact that they have already had some experiences and will have (or would have had) more experiences ahead of them. That’s why killing people is typically a much more serious business than killing dogs or cats. Moreover, it’s the fact that killing people deprives them of their future experiences, rather than the fact that it interrupts their experiences, that matters. Otherwise, killing a currently non-experiencing person (someone who is comatose, perhaps) would not be a big deal; but it is. (See p. 73, both columns.) Again this seems quite right.

Marquis goes on to point out certain other advantages to M1, beyond the mere fact that it is superior to M2 and M3. One is that it helps explain why euthanasia is not seriously prima facie morally wrong. He claims that euthanasia is relatively easy to justify precisely because the people whose lives are terminated are in such an unfortunate position that they do not in fact have a future available to them that is of any real value. (See p. 70, left column. Of course, whether we should agree with this claim is controversial. We will return to the topic of euthanasia later in this course.) Another advantage that Marquis claims for his account is that it helps explain why it is that contraception is not seriously prima facie morally wrong even though abortion is. The reason is that, prior to conception, there is no specific individual upon whom to pin the loss of a valuable future; it is only once conception has occurred that such an individual comes into being. (See p. 75, right column – p. 76, left column. Once again, it is of course controversial whether we should say that contraception is not seriously prima facie morally wrong. Although I will return to this issue soon, it is not one that we will be discussing at length.)

***Here is a good place to stop and consider Questions 3 and 4 on the Practice Questions for Module 4.***

5. Critical evaluation ​of Marquis’s article

As noted above, the success of Marquis’s argument depends crucially on whether principle M1 provides the best explanation of why it is seriously prima facie morally wrong to kill people like you and me. The fact (if indeed it is a fact, as Marquis claims and as I have agreed) that M1 is better in this regard than either M2 or M3 does not of course mean that it is better than all potential rivals. Perhaps there is some other principle out there that Marquis fails to consider and which (a) provides a better explanation while (b) not covering fetuses in the way it covers people like you and me. If so, then, once again, his strategy is doomed.

Paske:

In his article, Paske claims that principle M1 cannot be the best principle available because it has an unacceptable implication, namely, that, all else being equal, it is less seriously prima facie morally wrong to kill someone the older that person is. (See p. 79, right column. I will call this the “Variability Problem.”) To see why this is so, consider a line that represents various stages of our lives. (“C” represents conception, “B” represents birth, and the numbers represent years after birth. Obviously, the line is not drawn to scale!)

C		B		1		5		10		20		40		80

Think of someone who is 80 years old. He (or she) will, all else being equal, have only a relatively short future ahead of him. Assuming that he is leading a life of average quality, then, his future has only a relatively limited value, as depicted in the filled-in area below:

80

Now consider someone who is 40 years old. Even if he is leading a life of only average quality, the future available to him is far more valuable than the future available to the 80-year-old:

40

Now consider someone who is 20 years old. Even if he is leading a life of only average quality, the future available to him is far more valuable than the future available to the 40-year-old:

20

And so on, and so on. The net result is that, the older you are, the less valuable your future (all else being equal), while the younger you are, the more valuable your future. If, as M1 states, what makes killing someone wrong is the fact that doing so deprives him of a valuable future, then the implication would appear to be that, the older you are, the less seriously wrong it is to kill you, while the younger you are, the more seriously wrong it is to kill you. In fact, on this approach abortion immediately upon conception would seem to be the most seriously wrong form of killing humans that there is!

C

Well, as a relatively old person, I find that objectionable − as Paske does, and as I hope you do, too!

In place of M1, Paske proposes the following alternative principle as providing the best explanation of why it is wrong to kill you and me:

M4:	It is seriously prima facie morally wrong (and so overall morally wrong under unexceptional circumstances) to cause the immediate loss of personhood.

He explains at some length what he takes personhood in this context to be. It has especially to do with the ability to experience certain complex and sophisticated “conceptually based” emotions, such as guilt, indignation, hope, and pride (see p. 78, right column – p. 79, left column). And he is at pains to point out that, although M4 applies to you and me, it does not apply with equal force to fetuses; in fact, it doesn’t apply to them at all, since fetuses aren’t persons in the relevant sense. In this way, Paske seeks to undermine Marquis’s strategy.

But there are problems. One concerns Paske’s emphasis on the immediate loss of personhood. (This is reminiscent of the emphasis in M3 on the discontinuation of the experience of living. As Marquis noted with respect to M3, there seems to be no good reason to insist on such a restriction.) What if there were a temporary interruption in what Paske calls personhood (attributable, perhaps, to some reversible form of brain damage)? Surely we wouldn’t find it acceptable to kill the individual who is suffering from the temporary loss just because he is not a person (in Paske’s sense) at this very moment − especially not if he will soon become one again. Well, if it’s not acceptable to kill someone who’s about to become a person again, how is it supposed to be any more acceptable to kill someone who’s about to become a person for the first time?

A second and, I dare say, even bigger problem has to do with Paske’s emphasis on the possession of what he calls personhood in the first place. If being able to experience the so-called conceptually based emotions is supposed to be a pre-condition of being a person, then it is not just fetuses that don’t qualify. Neither do newborns! Neither, it seems, do infants and perhaps even toddlers. Surely this is too restrictive. Surely someone who is in favor of abortion will balk at endorsing infanticide or “toddlericide.”

The problem I’ve just pointed out is reminiscent of the challenge to non-conservatives on Question 3 that I mentioned in Section 3 above, and it too can be put in the form of a challenge to non-conservatives on Question 1. The challenge is this: Justify your discrimination. You claim that it’s seriously prima facie morally wrong to kill postnatal human beings but not so seriously prima facie morally wrong to kill prenatal human beings. What is the relevant difference between these two sets of individuals that could account for this difference in your moral verdict?

If you want to defend a non-conservative view, the “trick” is to find a criterion which is not too restrictive (as Paske’s seems to be) but also not so liberal as to include in the sphere of “morally protected” individuals those whom you seek to exclude.

Paske is aware of this problem and seeks to address it by distinguishing between two sorts of rights, “person rights” (had by persons and not by infants) and “social rights” (had by infants), claiming that the latter render infanticide seriously prima facie morally wrong after all. But this move, too, is problematic.

First, Paske’s remarks here are hard to follow. Secondly, the idea of social rights that he has in mind would seem to imply that the possession of these rights is highly precarious, being contingent on how the right-holder is viewed by others. If infants are prized in a certain society, then they will have the social rights Paske has in mind. But if they’re not prized? Then, it seems, Paske would have to deny that they have such rights. But this seems wrong. Just consider, for example, cultures in which young girls are regarded as “dispensable.” Should we say that they therefore have no moral rights? That doesn’t seem right. A third and final problem is that it’s not clear why it is that, if infants do or can have social rights, fetuses don’t or can’t as well. In fact, Paske appears to concede that fetuses can indeed have social rights (see p. 82, right column). So, once again, Paske’s attempt to place moral protection on postnatal humans but not on prenatal humans would appear to have failed.

***Here is a good place to stop and consider Questions 5, 6, and 7 on the Practice Questions for Module 4.***

Norcross:

In his article, Norcross also claims that Marquis has not found the best explanation of why it is seriously prima facie morally wrong to kill people like you and me. Unlike Paske, Norcross doesn’t offer an alternative to principle M1 that he finds preferable, but, like Paske, he does claim that M1 has unacceptable implications and must therefore be rejected.

Norcross’s complaint is that, contrary to what Marquis himself says, M1 does after all imply that contraception is seriously prima facie morally wrong, and that this is a mistake. In brief, he gives the following argument of Type V, an argument by reduction to absurdity:

Arg. 7:		(1)	If M1 were true, then contraception would be seriously prima facie morally wrong.
		(2)	Contraception is not seriously prima facie morally wrong.
	∴	(3)	M1 is not true.

Remember that Marquis claims that, prior to conception, there is no specific individual upon whom to pin the loss of a valuable future. Norcross denies this. Along with Marquis, Norcross notes that there are four possibilities to consider having to do with the sperm (whom I will call Sammy) that would in fact have fertilized the ovum (Olive) had contraception not taken place.

The first possibility is that Sammy is the individual in question. Sammy had a valuable future ahead of “him,” but contraception put an end to that. The second possibility is that Olive is the individual in question. “She” also had a valuable future ahead of “her,” had contraception not intervened. Like Marquis, Norcross rejects these suggestions. He thinks that neither Sammy nor Olive taken individually should in fact be said to be “the” individual who is deprived of a valuable future by contraception. Why? Because there’s no good reason to say either that it’s Sammy, rather than Olive, or that it’s Olive, rather than Sammy, who is so deprived. It is partly because of this fact that Marquis denies that there is a specific individual upon whom to pin the loss. But we have not yet exhausted the possible candidates.

A third candidate is the “mereological sum” of Sammy and Olive, which I will call Sammy+Olive. Mereological sums are supposed to be individuals in their own right that are comprised of other individuals. An example might be a particular cup+saucer, comprised (of course) of some cup and some saucer. You might be scratching your head at this. Why think that, in addition to the cup and the saucer, there is a third thing that is the cup+saucer? Likewise, why think that, in addition to Sammy and Olive, there is also another thing that is Sammy+Olive? Well, for a variety of reasons, philosophers have argued that we should indeed accept the existence of such things. Norcross is one such philosopher; Marquis is not. The arguments are difficult, however, and pursuing them would be a distraction. So I think the best stance for us to take here is to remain non-committal regarding the question whether there really are such things as mereological sums.

There is a fourth possibility: each of Sammy and Olive is deprived of a valuable future when contraception occurs. Marquis rejects this possibility, saying (p. 75, right column): “On this alternative, too many futures are lost. Contraception was supposed to be wrong, because it deprived us of one future of value, not two.” His point might be put graphically as follows. Suppose contraception had not taken place, that Sammy and Olive had thus “joined” one another, and that, as a result, a certain embryo had come into existence. Let us call this embryo Eddie. Suppose now that Eddie is aborted very soon after having been conceived. The picture would look like this:

E:

C

That’s a picture of something very bad, according to Marquis. Look at what’s lost! (I’ve indicated the loss by putting the future that would otherwise have occurred in red.) But now suppose that contraception did take place, so that Sammy and Olive never joined and Eddie never came into existence. Then, it might seem, the relevant picture would look like this:

S:

Є

 

O:

Є

 

Here, “Є” is supposed to represent the fact that conception did not occur. Marquis’s point is that, on this picture, there are two losses, which presumably would be twice as bad as the loss represented in the previous picture. But this, he thinks, is absurd. It’s absurd to think that contraception is twice as seriously wrong as abortion. Hence he thinks that we should not think of contraception as imposing any kind of loss at all.

We might well want to agree that it’s absurd to think that contraception is twice as seriously wrong as abortion, but Norcross points out, in effect, that there’s an alternative picture that Marquis fails to consider, namely this:

S/O:

Є

The idea that this is supposed to represent is that, when contraception occurs, it is indeed the case that only one future is lost, and not two, and that’s because each of Sammy and Olive is deprived of the same future. It’s not as if Sammy would have had his future and Olive hers; rather, they would have had theirs. This strikes me as a very interesting observation. I am inclined to think that Norcross is right, and that therefore the first premise of Arg. 7 is true. Since I am also inclined to accept the second premise, I believe that Norcross’s criticism of Marquis is sound. But I leave it to you to reflect further on this.

I noted above that Norcross doesn’t offer an alternative to Marquis’s principle M1 that he declares preferable. But he does ask us to consider the following:

M5:	It is seriously prima facie morally wrong (and so overall morally wrong under unexceptional circumstances) to deprive an individual of a valuable future to which he or she has a right.

His point is a simple one. If we agree that you and I have a right to life, then M5 would seem to be at least as good as M1 in explaining why it is seriously prima facie morally wrong to kill you and me. But then, in order to figure out whether M5 applies with equal force to fetuses as to us, we would after all have to address the question whether fetuses have a right to life. If they do, then they are covered by M5; but if they don’t, then they are not covered. Thus Marquis has not succeeded in bypassing Question 3 after all, since there’s a rival principle that implicates that question and which is at least as good as his in explaining why it’s wrong to kill postnatal human beings. This again seems to me a very good criticism.

In fact, I would go further than Norcross. It seems that M5 is actually superior to M1, for two reasons.

First, M5 provides a way to avoid the Variability Problem to which, as Paske pointed out, M1 is vulnerable. Remember that this problem is that M1 apparently implies that, the older you are, the less seriously wrong it is to kill you, while the younger you are, the more seriously wrong it is to kill you. The reason why M5 would seem to avoid this unwelcome implication is this. It might be said that, although an 80-year-old’s future is, all else being equal, less valuable than a 40-year-old’s, which in turn is less valuable than a 20-year-old’s, and so on, still the 80-year-old has just as strong a right to his future as the 40-year-old has to his, who in turn has just as strong a right to his future as the 20-year-old has to his, and so on. In this way, M5, unlike M1, gives us a way to account for the equally serious wrongdoing involved in killing each of these individuals.

The second reason why M5 would seem superior to M1 is that, whereas M5 seems pretty plausible, M1 on reflection seems not nearly no plausible. To see why, consider a competition. Suppose that there are 10 people running a race in which “the winner takes all.” Let us suppose that the prize is very valuable. Alf wins the race and gets the prize. Had he not participated, Bert would have won and got the prize. As it is, though, Bert is left empty-handed, just like all the other runners (except Alf). Alf has thereby deprived Bert of something very valuable. But has he done something seriously prima facie morally wrong? It doesn’t seem so, but why not? The most natural answer would appear to be: because Bert had no right to the prize. Contrast this case with the following. Alf takes the prize home, and Bert then steals it. Here it seems that Bert has indeed done something seriously prima facie morally wrong. Why? Because he has deprived Alf of something valuable to which he has a right.

And now notice this. If M5 is indeed superior to M1 for the reasons just given, then Marquis’s attempt to bypass Question 3 and go straight to Question 1 has certainly failed. The question whether the fetus has a right to life would appear to retain its significance regarding the morality of abortion.

***Here is a good place to stop and consider Question 8 on the Practice Questions for Module 4.***

6. Tooley o​n abortion

Let me remind you again that an argument’s being unsound does not entail that its conclusion is false. Thus nothing I have said has given us any reason to believe that Marquis’s conservative view on abortion is mistaken. What we would need for that is a sound argument for some non-conservative view. So let’s now take a look at one. The argument in question − Tooley’s − is an argument for a very liberal answer to Question 3 and, by implication, also to Question 1.

Earlier, I made note of a challenge (captured in Arg. 5 above) to those who are inclined to give a non-conservative answer to Question 3: How can you justify ascribing a right to life to newborns but not to fetuses at conception? Tooley rejects the presupposition underlying this challenge. He denies that newborns have a right to life.

Tooley asks the key question: What is it that makes a human_B a human_P? That is, what features must a member of the species homo sapiens have in order to be a person with a right to life? His answer (see p. 122, right column – p. 123, left column) is this: “An organism possesses a serious right to life only if it possesses the concept of a self as a continuing subject of experiences and other mental states, and believes that it is itself such a continuing entity.” In his article, Tooley provides considerations in support of this answer and then, on the basis of this answer, argues for a liberal view on abortion. His article is rich and complex. In what follows I will address only some of the highlights.

Tooley gives the following general account of what it is to have a right to something. Someone (or something), A, has a right to X just in case A satisfies the following four conditions:

(a)	A is the sort of thing that is a subject of experiences and other mental states,
(b)	A is capable of desiring X,
(c)	if A desires X, then others are under a prima facie moral obligation to refrain from actions that would deprive A of X, and
(d)	either	A does desire X
	or	A would desire X were it not for the fact that
		(i)	A is in an emotionally unbalanced state, or
		(ii)	A is temporarily unconscious, or
		(iii)	A has been conditioned to desire the absence of X.

This is very complicated, but we needn’t worry about all the details. We should, however, note the following considerations in favor of each of the four conditions.

First, condition (a) is supported by the commonly observed fact that inanimate objects like tables and chairs don’t have rights. (When I say “supported,” I don’t mean that the fact that inanimate objects lack rights shows conclusively that (a) is true. I mean only that this fact provides some reason to think that (a) is true.)

Second, condition (b) is supported by the idea that rights serve people’s interests and that interests are rooted in desires. Here “interest” is to be understood, not in the sense that people sometimes take an interest in something, but rather in the sense that certain things (such as being healthy or free of pain) are in their interest. Tables and chairs have no interests, in this sense. They have nothing “at stake.” You can damage a table or chair, but you can’t harm it in the sense of doing anything that runs counter to its interests. Again, tables and chairs have no interests. One plausible explanation of this is that such objects are not capable of having any desires.

Third, condition (c) is in keeping with the following thesis that was introduced in the section on moral obligation, right, and wrong:

The Correlativity Thesis:
X has a moral right against Y that Y do (or not do) A just in case Y has a prima facie moral duty to X to do (or not to do) A.

Finally, condition (d) accommodates “abnormal” cases in which rights are plausibly thought to exist even though the relevant desire does not.

For our purposes, it is condition (b) that is crucial, since it is this that lies at the foundation of Tooley’s argument. Roughly, his argument is that, because newborns are incapable of having the desire that is relevant to the possession of a right to life, they have no such right. This argument may be put more fully and more formally as follows (notice that the first premise just is condition (b) in Tooley’s general account of rights):

It is important not to misconstrue the conclusion of this argument. To say that killing newborns does not wrong them is not to say that it is never wrong to kill them. To see why, consider this example.

You have a neighbor, Bob, who is an avid gardener. He is especially fond of his daffodils, since to him they are a sign that winter is over and he can now look forward to many months of bloom and blossom. You are fond of his daffodils, too, and so one day, when Bob is not around, you sneak into his garden and cut down a whole bunch of daffodils, put them in a vase, put the vase on your living room table, and gaze at them with wonder and joy. You have done this without Bob’s permission because you know that he wouldn’t have given it to you. He is not only fond of his flowers but jealous of them, too. He likes to keep them to himself, dazzling the local community with his botanical skills.

I think that it is safe to say that you have done wrong in taking Bob’s daffodils without his permission. Indeed, you have wronged him. But you have not wronged them. Why? Because to wrong someone is to violate a right that that person has. It was Bob who had the right that you not take his flowers without permission. The flowers themselves had no such right.

This is what Tooley has in mind regarding newborns. Not having a right to life, they cannot be wronged if you deprive them of life. Nonetheless, you can certainly do wrong if you kill a newborn. In particular, you might well wrong its parents.

Despite the impression that the title of his article might give, then, Tooley is not advocating the wanton killing of newborns. He recognizes that, even if they cannot be wronged by being killed, there can be good reasons not to kill them. Nonetheless, he is of course saying that one very good reason not to kill you or me does not apply to newborns, let alone fetuses, and that is that you and I do have a right to life, whereas they do not.

The question of course arises: If newborns don’t have a right to life but we do, when precisely is this right acquired? Tooley believes that, though interesting, this is not a question of any great practical urgency, because, in almost all cases, whether infanticide is desirable will become apparent within a short time after birth, well before there is any chance that the infant will come to possess the concept of being a self-conscious subject of experiences, and hence well before it acquires a right to life. He suggests that a timeline of one week after birth be adopted as marking the period during which infanticide should be permitted. (See p. 130, right column.)

7. Critical evaluatio​n of Tooley’s article

Premises (2), (5), and (7) of Arg. 8 appear very plausible, and so I won’t question them. Premises (1) and (3) are more problematic, though.

Consider premise (3) first. Tooley states that a right to life is a right that one’s life as a self-conscious subject of experiences not be discontinued. This is pretty complicated. Why not just say, as I did earlier, that a right to life is simply a right not to be killed? Why bring in this whole business of being a self-conscious subject of experiences? Actually, though, this question is less important than it might first seem. This is because, if we concede (as is implied by premises (1) and (2) taken together) that A has a right to X only if A possesses the concept of X, then we must acknowledge that self-consciousness is in a sense “built into” the possession of a right to life, since one cannot possess the concept of one’s (not) being killed unless one is self-conscious.

So, if we grant premise (2), the really crucial premise in Arg. 5 is premise (1), i.e., condition (b) in Tooley’s general account of rights. I said earlier that this condition is supported by the idea that rights serve people’s interests and that interests are rooted in desires. And, although I find this a plausible idea, I think there is nonetheless good reason to reject the particular way in which premise (1) gives expression to it. The reason is that, even if interests are rooted in desires, it doesn’t follow that having an interest in something X requires that one be capable of desiring X itself. So, in particular, some person’s (or, indeed, some animal’s) having an interest in continuing to live may not require that he (or she, or it) be capable of desiring to continue to live. In his article, Stevens criticizes Tooley on this score. He gives the example of a patient, Smith, who has a right to immunosuppressive therapy, even though he has no grasp of, no understanding of, the concept of such therapy and hence is incapable of desiring such therapy. On Tooley’s behalf, Stevens commendably entertains, but in the end rejects, various responses to this proposed counterexample to premise (1). The details are interesting but difficult, and I won’t pursue them here, but I strongly urge you to reflect carefully on them.

In any case, the conclusion of Tooley’s article is surely very troublesome. In a manner reminiscent of Paske, he sets the bar for the possession of moral rights very high indeed. He claims not to be worried by the practical implications of his position, but I confess that I find his attitude here very puzzling. In particular, I am baffled by his view that the question just when a human being acquires a right to life has no great practical urgency. He says that in almost all cases it will become apparent shortly after birth whether infanticide is “desirable,” but he neglects to ask the question: desirable for whom? I suspect that he means “desirable for the infant,” and that the sort of case he has in mind is one in which an infant is born with such profound physical and/or mental problems that euthanasia would be the proper course of action to take. Even if we accept that euthanasia in such a case would be morally justified, and even if we accept that whether such action is called for would almost always become apparent within a week of birth, this entirely overlooks the question whether infanticide might later become desirable for someone else − for example, for the parents.

Consider this case. The Browns got married last year and immediately started trying to have a child. Their sustained and concerted efforts were successful! Mrs. Brown duly gave birth to little Billy. The Browns were thrilled, delighted, over the moon with joy. They took Billy home and fawned over him. After a while, though, their joy started to fade. The novelty of having a child began to wear off. Billy didn’t look quite as cute as he did at the beginning. Not only that, he never slept through the night, and his parents became more and more exhausted. Not only that, constantly having to change his diaper had become a real turn-off. The Browns started to miss their previous, carefree, childfree life. They wanted to go out on the town, the way they used to. They wanted to stay up all night and party! But they were too tired. Besides, they couldn’t afford a baby-sitter, not with all the new expenses they had incurred because of that damn kid. This was not the life they had envisaged! One day, the Browns reached a decision. They would smother little Billy and drop him in the nearest dumpster.

You would surely be appalled if the Browns were to do such a thing. But why? True, Billy’s death might not be desirable from his perspective, but it has certainly become desirable from his parents’ perspective. So why not simply get rid of him? He has no right to life (says Tooley), and hence his parents wouldn’t be wronging him by killing him, no more so than your neighbor Bob would be wronging his flowers if he suddenly grew tired of them and decided to get rid of them by chucking them in the dumpster. You might say that, because Tooley allows for the possibility of doing wrong without wronging anyone or anything, there is always room for him to say that the Browns’ getting rid of Billy would be wrong even if their doing so wouldn’t wrong him. But why would it be wrong for the Browns to get rid of Billy, if Billy has no right to life? It was wrong for you to take Bob’s flowers because they were his. It’s not wrong for him to take his own flowers and throw them in the dumpster. Likewise, it would be wrong for you to bump off Billy without the Browns’ permission, but why would it be wrong for them to do so, since they would obviously grant themselves the requisite permission?

Something seems to be seriously amiss here. The most obvious diagnosis would seem to be that Tooley’s denial that newborns have a right to life is mistaken and that his argument for this denial goes wrong right at the start, in its first premise, just as Stevens claims.

***Here is a good place to stop and consider Questions 9 and 10 on the Practice Questions for Module 4.***

8. A more m​oderate position

If what Tooley says with respect to the morality of infanticide seems unacceptable, then a very liberal position with respect to either Question 1 or Question 3 on abortion would seem also to be in trouble. For what relevant nonmoral difference could there be between a late-term fetus and a newborn that could justify discriminating between them? This isn’t, of course, to say that there isn’t or couldn’t be a relevant difference, but the burden would seem to be on those who claim that there is such a difference to specify just what it is and why it is relevant. Because of this, those initially inclined toward a liberal view might start to rethink their position and move toward a more moderate one.

But the challenge still remains: if it is hard to specify a relevant difference between a late-term fetus and a newborn, how is it supposed to be any easier to specify such a difference between a mid-term fetus and a late-term fetus, or between an early-term fetus and a mid-term fetus? What could possibly function as the relevant basis for the cut-off point between not having a right to life and having a right to life, or between its not being seriously prima facie morally wrong to kill a biological human and its being seriously prima facie morally wrong to do so?

Some people have of course proposed answers to this question. Among the more common proposals are that (a) having lungs, or (b) having a heart, or (c) being sentient (that is, being capable of experiencing sensations), or (d) being viable (that is, being capable of existing outside the womb) is the relevant cut-off point. Let me say something briefly about each of these suggestions.

Consider (a) first. Suppose that a 20-year-old woman, Jane, has lungs that have ceased to function and that she must therefore stay attached to a ventilator to survive. Does the fact that her lungs are incapacitated mean that she has no right to life or that it wouldn’t be seriously prima facie morally wrong to kill her? Surely not. Still, you might say, she does have lungs, whether they’re working or not. But so what? What if her lungs were removed? Would that mean that her moral status has changed? Again, surely not. Still, you might say, even if she doesn’t have lungs now, she once did. But again, so what? What if Jane developed in such an abnormal way that she never did have lungs in the first place and had to be put on a ventilator immediately at birth? (We can imagine that this takes place at some time in the future, when ventilators are small and easily transportable, so that Jane’s quality of life is barely affected.) Would that make a difference, morally? It seems that the answer must be No.

Similar remarks pertain to (b). It’s very hard to see how having or not having a (functioning) heart can itself constitute a morally relevant feature. (If you’re having trouble with this, try imagining a series of cases. In the first, Joe’s heart has ceased functioning and he has been put on a heart-lung machine. In the second, his heart has been replaced by an artificial heart. In the third, he never had a heart to begin with.)

Now consider (c). First, an interesting and perhaps surprising observation: there appears to be no firm consensus in the medical community regarding just when it is that sentiency is attained. It used frequently to be said that sentiency is attained at around 8 to 10 weeks after conception, but nowadays it is more common to hear an estimate of 25 to 30 weeks. That’s quite a difference! Many people are opposed to abortion at least in part because they fear that it causes the fetus pain. But if sentiency isn’t attained until week 25 or later, just about any abortion that is actually performed will occur before the fetus is capable of experiencing pain. (Estimates vary, but it seems that almost all abortions that actually occur take place well before week 25.) But, in any case, the pertinent question again is: so what? Why should we think that the capacity to feel pain has any relevance to the moral justifiability of killing something? Of course, it might well have relevance to which method of killing is used. If a fetus can feel pain, then it might have a right not to be caused pain, but why think it would therefore have a right not to be killed, as long it is killed in a painless manner? (In this regard, compare our treatment of other animals − an issue to which we’ll return later in the course − and how many people, while opposed to factory farming, are prepared to eat meat that is the product of other farming methods.) Moreover, why should it be thought that the absence of sentiency has any relevance to the possession of a right to life? In this connection, it’s worth considering the fact that some humans are born with a very rare disorder, called congenital insensitivity to pain with anhidrosis (CIPA), which renders them incapable of sensing pain. (You might think, “Wow! How lucky! If only my life were free of pain!” The trouble is, though, that these people get themselves into all sorts of trouble − broken bones, infections, and the like − precisely because they are not prompted by pain to refrain from certain behaviors. Not only that, they cannot feel any other nerve-related sensations, either. Think what that would be like!) Surely we wouldn’t want to say that someone with CIPA has a moral status inferior to ours.

Consider, finally, (d). One problem with this suggestion is that viability is a “moving target.” A fetus that is viable under one set of conditions may not be viable under another. Think of Alice. She was born in 2010, very prematurely, when her mother was only in week 27 of her pregnancy. But she made it, thanks to the technological marvels of the modern age, and to the fact that she was born in the United States. Now think of Betty, who was identical to Alice in every way you can imagine (without actually being Alice, of course). She, too, was a “preemie.” Unfortunately, though, Betty didn’t make it, because she was born in 1910, far removed in time from the necessary medical help. And now think of Carol, also identical to Alice, and also a preemie. Unfortunately, she didn’t make it either, because, although it was 2010 when she emerged from her mother’s womb, she was born in the Mozambican jungle, far removed in space from the necessary medical help. If viability were the cut-off point we are looking for, it would turn out that Alice’s moral status is superior to that of Betty or Carol. But that doesn’t seem right, precisely because all three are qualitatively identical to one another.

I don’t pretend that what I have just said is decisive in putting suggestions (a)-(d) to rest. Still, I hope that you can see how difficult it is to defend any of these suggestions (and any similar suggestion). It’s a juggling game. As noted in the commentary on Paske above, the trick is to find a criterion for the possession of a right to life which, when applied to individuals other than fetuses, is not so restrictive that it leaves out individuals who, intuitively, should be included among those who, like us, have such a right, but also not so liberal that it includes individuals who, intuitively, have a moral status inferior to ours. It’s not easy!

It may have occurred to you that the reason why it’s not easy to perform this trick is that it is, in fact, impossible to do so, and that the reason why it’s impossible to do so is that there is, in fact, no sharp cut-off point of the sort that we’ve been seeking. On the contrary, you may say, the acquisition of a right to life is a gradual process, not a one-time all-or-nothing event.

Well, that may be right, and it would seem to fit well with the common view that abortion is harder to justify the later in pregnancy it occurs. But still that doesn’t get you off the hook, in that, to defend this view, you must still find some underlying feature that grounds the acquisition of a right to life. In this case, the relevant feature, whatever it is, must itself be gradually acquired, if it is to provide reason to think that the acquisition of a right to life is itself gradual. What could this feature be?

9. Thom​son on abortion

In Section 2 above, I said that a right to life is presumably one of the strongest rights someone can have, so that, given the Correlativity Thesis, the prima facie duty or obligation not to infringe such a right is likewise very strong. From this it would seem to follow that, if there ever are circumstances under which it nonetheless would be overall morally justifiable to infringe such a right, these circumstances would have to be truly exceptional. That’s why we’ve been trying to find a well-founded answer to Question 3. But finding such an answer has proven very difficult. What are we to do?

Thomson believes that our search for an answer to Question 3 has been based on a false presupposition. She believes that the sort of circumstances in which it would be overall morally justifiable to infringe a fetus’s right to life (if it has one) need not be exceptional at all. In order to show this, she declares in the third paragraph of her article that she is willing to grant, for the sake of argument, that a fetus is a person and so does have a full-fledged right to life from the moment of conception (that is, she is willing to grant that the correct answer to Question 3 is an extremely conservative one); nonetheless, she says, it is possible to argue soundly for the overall moral justifiability of abortion under a wide variety of circumstances (that is, for a very liberal answer to Question 1).

(Side-note: Although Thomson is willing to grant for the sake of argument that a fetus is a person, she states in both the second and the final paragraphs of her article that she in fact rejects this claim.)

Thomson begins by asking us how conservatives typically argue for the unjustifiability of abortion on the basis of the claim that the fetus is a person. Here is a reconstruction of the argument that she attributes to them (see p. 95, right column − I’ve added a couple of steps for a reason that will become clear in just a moment):

Thomson agrees that, once the first premise has been granted, this argument “sounds plausible,” but she nonetheless thinks it’s flawed. To bring out what’s wrong with it, she asks us to consider what I will call the Violinist Analogy (see p. 95, right column – p. 96, left column). In this analogy, you, the reader, wake up one morning in a hospital bed and discover that you have been hooked up against your will to an unconscious violinist who is suffering from a rare disease and needs the use of your kidneys to survive. In the first version of this analogy, you’re told that this situation will only last nine months, at which point it will be possible to detach the violinist safely from you. Thomson imagines that you would be outraged if you were told that you must stay hooked up to the violinist for such a long time, even though he is a person who has a right to life. She imagines, that is, that you would not be convinced by the following argument:

But notice that Arg. 10 has exactly the same form as Arg. 9. (It is because I wanted to make sure that this fact is crystal clear that I added the steps I did.) Thus, if Arg. 9 is sound, then so, apparently, must Arg. 10 be. Conversely, if you reject Arg. 10 (as Thomson bets you do), then so, apparently, must you reject Arg. 9. And if Arg. 9 is to be rejected, then something has gone wrong with it, no matter how plausible it might have seemed at first.

The Violinist Analogy is of course supposed to be an analogy with pregnancy that is due to rape. After all, you were connected to Vic against your will. Nonetheless, Thomson’s point is that, if Vic’s right to life is not sufficient to protect him, morally, in your case, then so too a fetus’s right to life is not sufficient to protect him, morally, in any relevantly similar case. Thus a right to life does not trump all countervailing considerations.

In my formulation of Arg. 9, I have followed pretty closely the argument that Thomson attributes to her conservative opponents. It should be clear that this argument is indeed unacceptable, because it’s invalid. In particular, the move from (9) to (10) is fallacious, since (10) mentions issues (having to do with killing and overall moral wrongness) that have not been mentioned in any previous premises. (The same of course holds of the move from (9) to (10) in Arg. 10.) But this problem is easily fixed. In Arg. 9, all we need to do is to substitute (9) with the following:

Then (10) would indeed follow. (The same kind of fix could of course be given for Arg. 10.) But now Thomson’s complaint is, in effect, simply this: clause (a) of (9*) is false, since there are circumstances, such as those depicted in the Violinist Analogy, in which R1 is indeed overridden by contrary considerations.

To drive home this point, Thomson switches our attention from cases of pregnancy due to rape to cases in which pregnancy poses a risk to the mother’s life. She thinks it clear that it would be overall morally justifiable for a woman to have an abortion under such circumstances, and that it would be overall morally justifiable for someone to assist her in doing so, regardless of any right to life that the fetus may possess. In support of this point, she presents another analogy, the Coat Analogy (see p. 98, left column), in which she asks us to consider a scenario in which two people, Smith and Jones, each need a coat to avoid freezing to death. If Smith owns the coat and Jones doesn’t, then that breaks any “tie” that might otherwise obtain. It is the owner who should have the coat. Likewise, even if the fetus needs the mother’s body to survive, if the mother also needs her body to survive (that is, she needs to keep it to herself, since continued pregnancy threatens her life), then the fact that it is her body breaks the tie in her favor.

Thomson goes on to explain that a right to life is the sort of right that we have called a claim-right (the sort of right that one person holds against another), and she raises the question precisely what it entitles us to. In yet another analogy, the Sickness-unto-Death Analogy (see p. 98 right column – p. 99, left column), she imagines that, in order to survive, she needs the touch of “Henry Fonda’s cool hand” on her fevered brow. (Henry Fonda was a Hollywood leading man in the middle of the last century − somewhat beyond his prime when Thomson published her article in 1971, but apparently still able to make her heart flutter.) She notes that she has no right to his touch, even though she will die without it, and that therefore a right “to life” is not a right to be given whatever is required to keep one alive. Her point, of course, is that, even if a fetus has a right to life, it doesn’t follow that it has a right to the use of its mother’s body, even if that is what it requires to remain alive.

The cases that we have so far been concentrating on are extreme and, thankfully, rare. The vast majority of cases of pregnancy are not ones that are due to rape or pose a risk to the mother’s life. So, even if abortion can be justified in those extreme cases for the sorts of reasons that Thomson gives, it of course does not follow that it can be justified in what might be called “regular” cases in which pregnancy is due to voluntary intercourse and poses no serious risk to the mother. And it is precisely in such cases that many conservatives, appealing to the fetus’s right to life, claim that abortion cannot be justified. Thomson is well aware of this and so provides fresh reasons for thinking that abortion can be justified even in such cases, and even if the fetus has a right to life. She does so by means of two further analogies.

First, Thomson gives the Burglar Analogy (see p. 100, right column). Here she asks us to imagine a burglar climbing in through a window that one has voluntarily opened. Surely one doesn’t have to let the burglar (= fetus) stay just because one has voluntarily done what enabled him to get in, especially if he got in only because some bars (= contraception) that one had put in place proved defective.

Then Thomson gives the People-Seed Analogy (see p. 100, right column – p. 101, left column). Here she asks us to imagine that people-seeds drift about in the air like pollen; one such seed passes through a screen window and takes root in your upholstery. Surely, she says, you are under no obligation to let it remain in your house.

On the basis of such considerations, Thomson seeks to show us that, even if the fetus is a person, there are many kinds of circumstance in which it would nonetheless be overall morally justifiable to have or perform an abortion. (Note, though, that she does not say that abortion is overall morally justifiable under all circumstances. On the contrary, she says on p. 104, for example, that it would be “indecent” to have or perform a late-term abortion just to avoid the nuisance of postponing a trip abroad, and it’s clear that she takes “indecent” to imply “overall morally unjustifiable” in this context.)

***Here is a good place to stop and consider Questions 11, 12, and 13 on the Practice Questions for Module 4.***

10. Critical evaluation of Thomson’s article

There is no doubt that Thomson’s article is imaginative and provocative, but has she succeeded in showing that a liberal position on abortion is acceptable? Let’s take a careful look.

Pregnancy​ due to rape:

Consider, first, the Violinist Analogy. By means of this analogy, Thomson seeks to have us accept the claim that, even if the fetus is a person with a right to life as strong as yours or mine, nonetheless abortion can be overall morally justifiable if the mother has become pregnant due to rape.

As with any such analogy, one must first ask whether it is apt, that is, whether the two cases are the same in all relevant respects. If so, then one must reach the same verdict in each case. In the present instance, that leaves us with two possibilities: either (a) agree with Thomson that it is overall morally justifiable for you to disconnect Vic, and so conclude that it is also overall morally justifiable for Mary to abort Fred; or (b) insist that it is overall morally unjustifiable for Mary to abort Fred, and so deny that it is overall morally justifiable for you to disconnect Vic. Thomson is banking on your not being willing to accept option (b), and I suspect she is right, at least initially. However, it may be that, after further reflection, some readers will accept option (b) after all. For consider: what could justify killing an innocent person? The burden that has been imposed on you, that of having to deal with Vic, is of course huge. But huge though it is, how could it match the enormity of the burden you would be imposing on Vic if you were to take his life?

Still, even if you do reject option (b), it’s important to note that doing so does not require that you accept option (a), not if (c) the analogy fails because there is some relevant dissimilarity between the cases. So one question to ask is whether any such dissimilarity can be found.

In his commentary on Thomson’s article, Beckwith claims that there is a relevant difference between the cases: the fact that Vic is artificially dependent on your body, whereas Fred’s dependence on Mary’s body is natural. (See p. 109, right column – p. 110, left column.) His point may be put in the form of the following two arguments:

What are we to make of the second premises? I suspect we can agree that there’s a sense in which Fred’s dependency on Mary’s body is “natural” whereas Vic’s dependency on yours is not (although, as we saw in our discussion of homosexuality, it might not be easy to pinpoint this sense precisely), but, even if so, that still leaves the question why we should think this difference between the cases matters, morally. It’s not clear to me what the answer to this question is.

Another possibility is this. As Thomson notes in the Sickness-unto-Death Analogy, a right to life wouldn’t appear to be a right to have whatever is needed to stay alive. On the contrary, it seems to be a right not to be killed in particular. Now, it may be that, although we should agree that, like all abortion, Mary’s aborting Fred would constitute her killing Fred (or someone else’s doing so on her behalf, if a third party − a physician, say − performs the abortion), nonetheless your disconnecting Vic should not be classified as an instance of killing. Why not? Well, as we’ll see when we discuss euthanasia later, there are many people who claim that “pulling the plug” on someone who is on life support counts not as killing that person but rather as letting him die. (This is controversial, as we’ll see when we discuss euthanasia, but let’s just grant it for now.) And then, it might be said, the relevant difference between Vic and Fred is that Vic is sick, whereas Fred is not, and so Vic is using your body as a form of life support, whereas Fred is not using Mary’s body in the same way. For this reason, disconnecting Vic would not count as killing him but rather as merely letting him die, and so would not constitute an infringement of his right to life.

I leave it to you to reflect on these issues further.

***Here is a good place to stop and consider Question 14 on the Practice Questions for Module 4.***

Pregnancy that po​ses a risk to the mother’s life:

There is no doubt that many would sympathize with the view that, if continued pregnancy threatens the mother’s life, then abortion would be justified, even if the fetus has a right to life. But even here the matter is not straightforward.

Thomson’s remarks may seem to suggest the following idea. (I’m not claiming that Thomson does in fact accept this idea. I find the text unclear on this point.) We have granted that Fred is a person and so has a right to life (R1). But Mary is a person, too. So she also has R1. But Mary also has a right to decide what happens to her body (R2). Now, it may seem that, in a “regular” case of pregnancy, Fred’s R1 outweighs Mary’s R2. But in a case where Fred poses a risk to Mary’s life, her own right to life comes into play, and that tips the balance. In brief: (R1 + R2) > R1. (And that, it might be added, is why clause (a) of (9*) above is false.)

This may seem plausible, but we must tread carefully here. First, if we grant that, as a person, Mary has both R1 and R2, must we not then also grant that, as a person, Fred also has both R1 and R2? If so, there would seem to be a stand-off: (R1 + R2) = (R1 + R2). If so, Mary’s rights don’t outweigh Fred’s after all.

You might reply, “But how could Fred have R2? R2 is a right to decide what happens to one’s body. Fred, being a fetus, can’t make any decisions. And if he can’t make any decisions, he doesn’t have a right to make any decisions.”

But this is too quick. Suppose that Dave has just been admitted into the emergency room of a hospital. He’s unconscious, and so he cannot make any decisions about his treatment. Does he therefore have no right regarding how he is to be treated? That would surely be too hasty a conclusion to draw. You might say that we should try to find out whether he has made any prior decision about how he should be treated in such a situation. But what if he has made no such prior decision? In such a case, a proxy will have to be found, someone who will make the decision on Dave’s behalf, on the basis of a judgment of what would be in Dave’s best interests. In this way, Dave’s R2 will have been put into effect indirectly, as it were. But if this is true of Dave, why not also of Fred? As a person, Fred would seem to be just as deserving as Dave of having his R2 put indirectly into effect.

Another question that might be asked in this context is whether Mary’s R1 really is implicated in the present situation. Clearly, her life is at risk, but is her right to life at risk? We have been assuming that a right to life is a right not to be killed, but, although we might well say that continued pregnancy would kill Mary, it’s not clear that we should say that, if the pregnancy is not terminated, Fred would kill Mary. Certainly, her having an abortion doesn’t fit what might be called the “standard” case of self-defense, one in which (as Brody notes in his article) there is a “pursuer” who is responsible for wrongly putting one’s life (or well-being) at risk. It seems rather awkward to characterize Fred as a pursuer, and it is clear that he is not responsible for wrongly putting Mary’s life at risk.

Still, Brody does go on to note that killing in self-defense can be justified even in certain non-standard cases in which the person killed is innocent. (See p. 105, both columns.) But, as he also goes on to note, the mere fact that one person’s continued existence poses a threat to the life of another does not appear to justify the latter’s killing the former. It is thus quite unclear whether Mary’s aborting Fred can be justified by appealing to “self-defense” after all.

Now, nothing I have said shows that Mary’s killing Fred in the name of self-defense cannot be justified, and you might well think that it can be. But notice how limited a concession this still is. To say that, in order to save her life from the threat posed by Fred’s continued existence, Mary would be justified in killing Fred is not to say that some third party (such as a physician) would be justified in killing Fred to save Mary. Thomson is of course aware of this, and it is for this reason that she presents her Coat Analogy. But doubts can be raised about the aptness of this analogy.

First, Thomson assumes that Fred has no right to the use Mary’s body, just as Jones has no right to the use of Smith’s coat. But is that so? Might it not be that, temporarily at least, Fred does have a right to the use of Mary’s body. (Consider what Beckwith has to say about this.) Why assume that Fred is like a stranger rather than, say, like a temporarily conjoined twin?

Secondly, Thomson seems to be ignoring the fact that, if continuing the pregnancy means that Fred would be using Mary’s body without her consent, so, by the same token, terminating the pregnancy means that Mary would be using Fred’s body without his consent.

Finally, when it comes to third-party intervention, there is an asymmetry that Thomson ignores (although she briefly alludes to it on pp. 96-97; it is an issue that Brody discusses on p. 107). Suppose that Phil, the physician who is attending to Mary and Fred, is confronted with the choice between saving Mary and saving Fred. If the pregnancy is terminated, Mary will survive but Fred will of course die. If the pregnancy isn’t terminated, Mary will die but Fred will survive. (How could that be? Consider preeclampsia, a condition in which the pregnant mother’s blood pressure rises rapidly. There can be cases of preeclampsia in which the mother’s eventual survival requires termination of the pregnancy at a stage at which the fetus cannot survive, yet the fetus can survive if the pregnancy is not terminated at that stage.) It might seem that Phil is faced with a very difficult but symmetrical choice: save Mary, or save Fred. But this description of the choice masks an underlying asymmetry. To save Mary, Phil must kill Fred; but to save Fred, Phil need “only” let Mary die. Now, if a right to life is a right not to be killed but (for reasons that Thomson herself alludes to in the Sickness-unto-Death Analogy) such a right is not a right not to be allowed to die, then Phil’s choice is between killing Fred and thus infringing Fred’s right to life, on the one hand, and letting Mary die and not thereby infringing her right to life, on the other. This would seem to imply that, all else being equal, Phil ought to favor Fred over Mary, even in this very extreme case in which Mary’s life is at risk.

There can of course be cases in which Phil’s choice is different: save Mary, or save no one. These are cases in which Fred is not going to survive, no matter what. If he is aborted, he’ll die (but Mary will survive); if the pregnancy goes to term, he will still die (and Mary will too). It may well seem that in such cases an abortion would be overall morally justified, even if the fetus is a person with a full-fledged right to life (a point that Brody himself makes on p. 108 and that Beckwith echoes on p. 110). But notice just how extreme such cases are.

***Here is a good place to stop and consider Question 15 on the Practice Questions for Module 4.***

“Regular” unwanted pregnancies:

Let’s now turn to a consideration of cases in which pregnancy is due to voluntary intercourse and poses no serious risk to the mother’s health. As noted above, Thomson claims that abortion can frequently be justified even in such cases, even when (as has been the presupposition throughout our discussion) it is granted that the fetus is a person with a full-fledged right to life. But the means by which she seeks to support this claim are surely questionable.

Consider, first, the Burglar Analogy. There is good reason to believe that this analogy fails. First, the burglar is not an innocent person in the way in which the fetus is. The burglar is responsible for wrongly climbing through your window. Secondly, and perhaps more importantly, throwing the burglar out of your house or apartment won’t kill him, whereas abortion will kill the fetus.

Thomson seems to be aware of the problems with the Burglar Analogy, and this is why, I think, she turns immediately to the People-Seed Analogy. But it’s hard to know what to make of this analogy, since the sort of scenario envisaged in it is so bizarre, so removed from reality, that our normal moral intuitions seem to have no firm foothold. Still, it’s worth noting that Thomson may be “fixing the game” somewhat by describing the individuals in question as people-seeds rather than as people. Remember that we are presupposing that fetuses are full-fledged persons. It’s one thing to talk about “vacuuming your upholstery” to get rid of seeds; it’s quite another to talk about doing so to get rid of people.

In Thomson’s defense, it might be pointed out that a woman’s voluntarily having sex doesn’t mean that, if she then becomes pregnant, her pregnancy is voluntary. That is of course right. In the extreme case, a woman might not know that she could become pregnant by voluntarily engaging in sex, in which case the ensuing pregnancy would clearly not be voluntary. And even in those normal cases in which a woman is perfectly aware of this possibility, it can seem misleading to call the ensuing pregnancy voluntary, especially if she took certain precautions to avoid it. Even so, the question remains how to weigh the burden of pregnancy against the burden of loss of life, when, remember, it is being assumed that both burdens are being borne by persons − especially when, in the case of pregnancy (but not in the case of loss of life), though the burden itself may be involuntary, the risk of incurring the burden was indeed taken voluntarily.

Concl​usion:

All in all, Thomson’s article strikes me as very interesting, very provocative, and very subtle, yet her case for a liberal position on abortion even when granting that the fetus has a right to life strikes me as inadequate, for the reasons given above. Just as Marquis’s attempt to bypass Question 3 and go straight to Question 1 in order to argue for a conservative position on abortion comes up short, so too does Thomson’s attempt to bypass Question 3 and go straight to Question 1 in order to argue for a liberal position come up short. Frankly, I find it doubtful that an acceptable resolution of Question 1 can be achieved without first resolving Question 3, but that of course remains an open question.

11. Therap​eutic abortion

So far, we have been concerned with abortion for the mother’s sake, but the question sometimes arises whether “therapeutic” abortion, that is, abortion for the fetus’s sake can be morally justified. (This seems to be the sort of thing Tooley had in mind when he talked about the “desirability” of infanticide being apparent within a short time after birth.) This is really just a question about prenatal euthanasia, and, since we will be discussing euthanasia later, I won’t address it. Nonetheless, I do want very briefly to note one possible difficulty.

Many people (including you, perhaps?) who advocate abortion (in certain circumstances) for the mother’s sake also advocate abortion (in certain other circumstances) for the fetus’s sake. The first position would appear to treat the fetus as someone or something that is not deserving of any special consideration, whereas the second position seems to treat the fetus as deserving of special consideration. It is not easy to see just how these positions are to be reconciled with one another (which of course is not to say that they cannot be). Yet another puzzle to ponder…