should have said the pro-death penalty position, of course.

Is the anti-death penalty position premised on a right to life consistent with support for incarceration? After all, I believe most death penalty opponents believe that people normally have the right not to be involuntarily confined, but only the anarchists among them believe that people cannot forfeit this right by their actions. So why is freedom a right subject to forfeit but life not? The anti-abortion position is that the right to life is subject to forfeit on the basis of one’s own actions, and this seems consistent with how rights are treated generally, including by abortion advocates.

Clay Shirky, while the majority favors abortion only in special circumstances, it does not favor mandatory abortion in those situations. It is therefore not asserting the claim of society over the individual. Furthermore, the most common cited special case, and the one relevant to the violinist example, is rape, which does not produce children distinguishable from others. The cases where abortion would be permitted seem to be those: 1) where the mother has no moral responsibility for the pregnancy, i.e., rape 2) where the burden on the mother is greater than usual. In the latter case, she has a choice to abort. And the most commonly cited and widely supported case here is where there is a threat to the life of the mother, not where the child is physically impaired. Abortion in the latter case has bare majority support, as opposed to almost 90% support where the mother’s life is threatened (source).

I think putting the point the way you do is very helpful, but I still don’t see this, even with this qualification; because even if there really is a well-defined meaning it wouldn’t necessarily be inconsistent. For instance, let’s take a different sort of right:

(1) Murderers have a right to liberty.
(2) Despite (1), it is not wrong to lock murderers up in prison.
(3) I have a right to liberty.
(4) Purely because of (3) it is completely wrong to lock me up in your basement.

Whatever the limitations of this argument, it is not logically inconsistent, because the sorts of actions we take to be forbidden by rights do depend on circumstances, and combining the same right with very different circumstances yields very different result. The conclusion that would have to be drawn from these four statements is not that the person saying them is inconsistent but that they are committed to saying that my being locked up in your basement is really a very different case than a murderer being locked up in prison, despite the superficially similar descriptions. Only if the circumstances here are held to be closely analogous do we actually have an inconsistency; if someone denies they are sufficiently analogous, they may be wrong but they are not inconsistent.

So too with the violinist case: it doesn’t show that the position in question is inconsistent in itself, but that (if the violinist case is sufficiently analogous to pregnancy) then it follows that one could kill a fetus even if it is a person with a right to life just like a violinist is a person with a right to life. That is, it’s not an argument that a certain common pro-life position is inconsistent; it is an argument that the position the pro-lifer is criticizing is, contrary to his claims, consistent even if the pro-lifer is right about the right to life. It is not set up to be a very good argument for the inconsistency of the right-to-life argument Thomson explicitly mentions, because the inconsistency arises not directly from the four claims you mention but from those four claims combined with the fifth claim that the two cases are so similar that for moral purposes they are not significantly different. But nothing about the argument Thomson is criticizing commits the pro-lifer to this fifth claim. What Thomson’s argument is better set up to do is to show that the right-to-life argument is not the open-and-shut case he thinks it is, because anyone who regards the cases as sufficiently analogous can concede that the fetus has a right to life and still hold that it can be aborted, without being at all inconsistent. In other words, the violinist case, if successful, gives us no categorical conclusion but instead the disjunctive conclusion that either the two cases are not sufficiently similar to transfer conclusions from one to another or we should not disconnect the violinist or abortion is consistent with the fetus having a right to life. (It does not merely do this; in giving the argument as a thought experiment Thomson gives no conclusive reasons for eliminating the disjuncts but simply tries to persuade us that we all really reject the second disjunct already and by the same token tries to make the cases as similar as possible to rule out any obvious reason to accept the first disjunct.)

Or so I would say. As I said before, it’s curious that there are so many differing interpretations of the violinist case in this one thread.

But if you are looking for a knock-down argument; not from me.

But, to get at least the questions clear, I don’t think of euthanasia as self-harm. Far from it – it is a choice (and just like in abortion one needs some procedures and checks in order to qualify for it). I certainly don’t think in terms of sin.

(I also don’t think the death penalty is necessarily a question of retributive justice but I have no wish to press you into prolonging this debate by continuously responding with something more detailed)

It actually depends on whether one considers “Is a foetus a person?” and “Is retributive justice justified?” (and, to go back to your own interest, “Is self-harm sinful?”) questions that are morally linked so that if one answers one question in a certain way one must also answer the others in corresponding ways, or else one falls into the dreaded self-inconsistency trap.

Not a lot of people will be daft enough to be 100% pro abortion and 100% anti death penalty in any super-strict way.

Certainly not, but how would this change our discussion? We seem to be discussing the correlation between commonly held beliefs on two or three moral questions with an unclear linkage. If you add the possibility that beliefs differ in intensity, how would that affect the (in)consistencies between them?

By the way, my interest was in an inconsistency between anti-euthanasia and pro-death penalty beliefs. If those are inconsistent that would be helpful. Not to convince me of the death penalty, but to convince some others either pro ‘the right to die’ or anti ‘the death penalty’, both of which be helpful turn-arounds.

• and part of it is what you remarked higher: false dichotomysis. Not a lot of people will be daft enough to be 100% pro abortion and 100% anti death penalty in any super-strict way.

I’d forgotten that (or rather, assumed it implicitly without thinking about it), but it does make sense to assert. It’s not like, analogously, the fetus is communicative. As you said, whole different story if it were possible to hold a conversation…

Of course, the resolution on the other side isn’t much harder—if you believe that fetuses and people convicted of crimes both start out with a right to life, but that someone who commits bad acts forfeits that right, you’re done, without any glaring inconsistency.

In that case the only two remaining self-consistent positions are (pro-abortion rights, pro-death penalty) and (anti-abortion rights, anti-death penalty). I’m interested in, but fail to see, a criterion that makes one self-inconsistency (pro-abortion rights, anti-death penalty) an acceptable inconsistency but the other one (anti-abortion rights, pro-death penalty) unacceptable. Nevermind that likely a sizable majority of the population hold one of those two positions.