« PhilPapers | Main | On Tipping »

Friday, June 05, 2009


Feed You can follow this conversation by subscribing to the comment feed for this post.

"What the Frankfurter needs is a possible set-up in which an agent is morally responsible but cannot act otherwise. Now if compatibilism is true, he can easily get what he wants -- and indeed without these far-fetched stories of psychosurgery, etc. But if incompatibilism is true, then the Frankfurt cases are impossible."

Michael Bergmann thinks otherwise (and, BTW, there are plenty of attempted FSC's that attempt to do what you claim needs to be done).

Here's Bergmann's paper:



No doubt they *attempt* to do it. Thanks very much for the Bergmann link. I'll check it out.

Hi Bill,

Yes, you're right. Whether they are successful can be debated. However, a possible impression that your post gives is that no such attempts have been given. You make mention of what the Frankfurter needs. But apparently he has this (and I'm granting that indeterminist world objections even work).

You claim that "If indeterminism is true, then Frankfurt cases are impossible." But there are many (see Mele, Robb, etc) FSCs that make arguments while assuming indeterminism is true. It certainly doesn't seem impossible. At least it's not clear that it's impossible.

What's more interesting is that there are many libertarians (some who specialize on the metaphysics of free will) who deny that PAP is necessary for responsibility. They usually base this on FSCs. So, you have William Lane Craig, David P. Hunt, Linda Zagzebski, for example. You also have those like Bergmann, though not a specialist on the metaphysics of free will, a very fine philosopher.

So, it's not only determinists who find FSCs persuasive, there's indeterminists too. And, many of those indeterminists are secialists in the metaphysics of free will and aware of indeterminist world objections.

When you have people on both sides denying PAP, that might indicate that matters aren't so simple.

Bill & several other Mavericks have regularly been paying obeisance to the PAP in this discussion of Frankfurt counterexamples. Let me say why I think this is a bad idea. (Yes, Bill, this is going to be the mother of pap smears!)

Many of you have heard of a Trojan Horse strategy in other contexts. One way to wreck a deal or contract is to quietly attach a condition that that cannot be executed or satisfied. So, if we say a agent shall be responsible for doing x only if conditions a and b and c obtain, and it turns out that c cannot obtain, or we can never ascertain or be sure that c obtains, we have rendered it impossible ever to hold someone responsible for doing x. Yes? The PAP is a Trojan Horse if we attach it to moral responsibility. Some defenders of the PAP sees this very clearly and want the conclusion that no one (themselves especially) is ever really responsible for the bad things they do. The PAP opens the barn wide open to the excuse that “I could not have done otherwise.”

“I cannot understand how I did those horrible things to that nice lady. Yes, I knew it was terribly wrong, but it felt so good, and I just could not stop myself. I guess I could not have done otherwise, and the philosophers tell me that means I’m not responsible, and so I'm not going to feel guilty about it or when it happens again.” This is the kind of disgusting and dangerous self-excusing that a PAP principle aloows. It is the kind of excuse psychopaths have been praying for —and if only it could only be converted into a legal defense! Then no truly evil monster could ever be held criminally responsible for his horrible deeds.

The criminal law wisely and steadfastly refuses to allow a Trojan Horse like the PAP into the criminal codes, despite the fact that some benighted philos recommend it. Think about what would happen if a PAP clause—“he could have done/chosen otherwise”-- became a material element of a criminal charge. Then the prosecution would have the burden of proving beyond a reasonable doubt that the perpetrator could have acted/chosen otherwise. First of all, this would be the biggest payday ever for criminal psychiatry, as defense and prosecution experts lined up and debated how deviant the brains of Gacy & Bund had become. There almost certainly is no conclusive answer at this point, and so we could not convict them.

The PAP is a Trojan Horse designed to wreck the concept of moral responsibility by requiring undecidable calls as to what someone could or could not have done, especially when there are genuine issues of psychopathology involved. Some people would very happy with this result, both for reasons of excusing their own misdeeds and for the sake of pressing their idealogical opposition to capital punishment, etc.

Just as the law refuses to allow any PAP condition to undermine criminal responsibilty, we must not allow a PAP condition to undermine our own personal sense of moral responsibility.

On the technical issue of whether Frankfurt-style counterexamples like Bad Bob require/presuppose causal determinism, I don’t see the beginning of an argument that only in a causally deterministic world could Evil Genie monitor and influence Bad Bob’s actions. Keep in mind how strong a claim causal determinism makes about the universe. According to causal determinism, EVERY causal sequence is fully deterministic and governed by non-probabilistic laws. There is no way to deduce that from the fact that Evil Genie can reliably anticipate some last minute aversive reactions Bad Bob might have and response to them. (If you can produce this argument, I will drive to Gold canyon with a CASE of Fat Tire as your condign tribute.)

Conscious human decisions, even very simple ones like “soup or salad”, arise from complex neural processes that take time. No neural processes, no decisions or even consciousness. If the libertarian wants to pretend that “decisions” can spring into existence instantaneously without neural processes, he is riding lunicorns again and he is not talking about decisions that human beings make. Now we have no idea how deterministic all this neural processing is, and there are certainly no deterministic laws that anyone has a clue about. What we can be sure of is that high amygdala activation is NOT good for someone continuing to do what he doing. So if we are watching and can react quickly, in one or another way we can overwrite this aversive reaction and keep Bob feeling confident and happy about the embezzling he is doing. Show me how you can deduce causal determinism from these very modest predictive and interactive claims, and I will be very surprised.


You are right: it is not simple at all, and what I say above may contain mistakes. These posts, like most of my posts, are first-draftish and exploratory: I am working through the recent literature and trying to understand the free will problem in all its complexity. There is a tangle of issues: modal issues, questions about causation and determination, metaethical questions, the mind-body problem . . .

You say that there are libertarians who deny that PAP is necessary for MR. Would these be the 'source' as opposed to 'leeway' incompatibilists?


Son of a gun, that is indeed the mother of all PAP smears! Your comment reminds of something Sartre says deep in the bowels of Being and Nothingness (if memory serves): determinism is a bottomless well of excuses.

But surely that a principle like PAP can be misused by some turkey doesn't mean much.


As for you second point I refer you to Kane, Widerker, and Goetz.


The libertarians who deny PAP are diverse. 'Source' incompatibilists are broken down further into 'wide' and 'narrow' source incompatibilists. Wide ones, like Timpe (for instance), claim PAP is necessary, the narrow ones, like Zagzebski (for instance) claim that it is not. But some libertarians seem to deny PAP for reasons other than 'source incompatibilism', issues like God's foreknowledge work for some, like Hunt (for instance). Others just find FSCs too doggone persuasive :-)


Thanks. You say that Timpe claims that PAP is necessary while Zagzebski claims that it is not. Do you mean that Timpe claims that PAP is a necessary truth while Zag claims that it is not a necessary truth, but a contingent truth?


Sorry for the confusion. Narrow source incompatibilists (like Zagzebski) do not think PAP necessary to ground ascriptions of moral responsibility. An agent can be responsible without having any alternatives. Wide source incompatibilists (Kane, Pereboom, etc) claim that sourcehood is ultimate but that a weak form of PAP is *entailed* by a right view of sourcehood. Timpe thinks narrow collapses into wide--i.e., we can only be the ultimate source onlyif we have alternatives. So, in that sense, for wide incompatiblists, since ultimate sourcehood is what is most relevant for ascriptions of responsibility, and since we can only be the ultimate source only if we have alternatives, PAP is necessary that way.

The comments to this entry are closed.

My Photo
Blog powered by Typepad
Member since 10/2008



June 2024

Sun Mon Tue Wed Thu Fri Sat
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
Blog powered by Typepad