Formulating PAP

I’ve recently become puzzled about the best way to formulate the Principle of Alternative Possibilities (PAP). Almost everyone follows Frankfurt (1969) who puts it like this: “a person is morally responsible for what he has done only if he could have done otherwise”.  This is often sharpened like this: “for all persons S and actions A, S is morally responsible for doing A only if S could have done something other than A.”

But there seems to be an obvious problem with this formulation. Frankfurt, as everyone knows, claims that PAP is false; that is, he asserts the negation of PAP. But the negation of PAP (as formulated above) comes out as: “there is a person S and an action A, such that S is morally responsible for doing A and it is not the case that S could have done something other than A.” But this clearly isn’t what Frankfurt (or any contemporary ‘Frankfurtian’) has in mind. This is both too weak and too strong. Too strong: because Frankfurt isn’t claiming that any actual person is responsible. (Despite being a compatibilist about responsibility and determinism, he might think there are other actual considerations that universally rule out responsibility.) Too weak: because Frankfurt isn’t trying to say anything about the actual facts of responsibility; rather, he’s saying something about the conditions for responsibility.

Rather, in denying PAP, Frankfurt means to say that moral responsibility doesn’t require the ability to do otherwise; in other words: it is possible to be responsible despite lacking alternative possibilities. This suggests that PAP is best formulated as follows: “necessarily, for all persons S and actions A, S is morally responsible for doing A only if S could have done something other than A.”  The negation of this is: “possibly, there is a person S and an action A, such that S is morally responsible for doing A and it is not the case that S could have done something other than A”, which seems closer to the mark.

Has anyone else noticed this?

 

Comments

[Fritz Warfield]

Philosophical *principles* are the sorts of things that are necessary if true. The default is to read principles like this as candidate necessary truths. One putting forward the weaker principle you mention (for what purpose I can't imagine...) would bear the burden of explicitly noting that one means the weak principle rather than the modally strong principle.

[Chris Franklin]

Daniel,

I noticed this also. In my recent Phil studies article I just added the necessity operator, taking this to be what Frankfurt intended, but also wanting to be more explicit about the purport of PAP. Neal Tognazzini and Felipe Leon have a nice discussion of this, as well as other questions related to how best to formulate PAP, in their forthcoming PPR article (if it ever comes out!)

[Daniel Cohen]

Thanks for these references Chris!

[Christian Lee]

Hi Daniel,

My two cents. You wrote PAP this way:

“necessarily, for all persons S and actions A, S is morally responsible for doing A only if S could have done something other than A.”

This can't be right. Suppose S didn't do A. If not, S is not responsible for doing A. So...

Necessarily, for all persons S and actions A, S is morally responsible for doing A only if S performs A and S could have done something other than perform A.

I just added another necessary condition, but presumably it should be there. It's just a stipulative definition anyway, but I would put it this way instead:

Necessarily, for all persons S and actions A, S is morally responsible for doing A only if S performs A and S could have refrained from performing A.

[Fritz Warfield]

Christian,
You say the initial formulation "can't be right" and then you propose a replacement that implies it -- your proposal identifies a condition that is necessary for moral responsibility only if the original is.
Clearly something went wrong...

[christian]

Fritz,

I see. All that I meant was, first, that it needed supplementation, namely, another necessary condition, and second, that it should be stated in terms of refrainings. I do agree that, as Daniel stated PAP, it does correctly identify one necessary condition. I suppose that, whether PAP should include these other conditions, those I mentioned, is a matter of terminology.

[Fritz Warfield]

What would it need supplementation for? Perhaps to be a sufficient condition? But it's not being put forward as a sufficient condition. It's being put forward as an important necessary condition.

Whether other clauses should or should not be added depends presumably on the question of for what argumentative purpose the addition(s) might be made. Whether such additions are plausible additions is a substantive philosophical matter and not simply a matter of "terminology".

[christian]

Right, it's not being put forward as a sufficient condition. Of course. I wasn't, at least, I didn't mean to imply that I was putting forward my additions (i.e. that the act was actually performed and that one could have refrained from performing it) to be jointly sufficient conditions for responsibility. What we have is a question about a particular necessary condition on moral responsibility. In particular, we have a question about the "ability to do otherwise".

I think this is best read as shorthand. The question should be about when we are responsible for doing something when we have done it (or, perhaps, failed to do it). So, we should look for necessary conditions on when we have done something when we could have done otherwise. I think that a correct answer to this question entails that we have done the act in question. So, a correct answer should entail that we have done the act in question. This is why I added the (albeit trivial) additional necessary condition, namely, that the act was performed.

What I meant by this being a terminological matter was only that PAP is a stipulative definition. I'm thinking that a better way to flesh out the intuitions that motivate PAP should involve the fact that one has performed the act in question, and that one could have refrained. I think that these additions are, like you say, motivated by substantive philosophical matters.