Friday, April 4, 2008

An argument for incompatibilism

The following argument is valid:

  1. Normally, if an embodied person freely does A, then x could have done otherwise than she did. (Premise)
  2. It is a common occurrence that an embodied person freely does something. (Premise)
  3. If a general conditional holds normally, and specific cases of the antecedent are common, then it is nomically possible that the antecedent and consequent hold simultaneously. (Premise)
  4. Therefore, it is nomically possible that there be an embodied person x and an action A such that x does A freely and x could have not done A. (By (1)-(3))
  5. Metaphysically necessarily, if an embodied person x does A freely and x could have done otherwise, then determinism is false. (Premise)
  6. Therefore, it is nomically possible that determinism does not hold. (By (4) and (5))
  7. If determinism holds, then it holds of nomic necessity. (Premise)
  8. Therefore, determinism does not hold. (By (6) and (7))

Observe that the Principle of Alternate Possibilities in (1) is not subject to any Frankfurt-style counterexamples. I got this Principle based on an idea of David Alexander, but I don't think he endorses this version.

I think the tough question is whether (3) holds. But I think it at least holds as a probabilistic principle: if normally (if p, then q), and if cases of p are common, then probably a case of both p and q is nomically possible. In fact, a stronger probabilistic claim seems to hold: probably some case of both p and q actually holds. (When I talk about cases, I am assuming that the conditional is a quantified one: for all x, if P(x), then Q(x).) If so, then the conclusion would be that probably determinism does not hold. (Not earthshaking in light of the fact that there is some direct reason from physics to think it does not hold.) But the stronger non-probabilistic claim is also plausible. How could something be normal and yet nomically impossible?

[Edited to fix typo in argument and attribution of PAP.]

14 comments:

Mike Almeida said...

Metaphysically necessarily, if an embodied person x does A freely and x could have done otherwise, then determinism is false.

That just begs the question against compatibilism. Compatibilists who are also determinists do claim that x could do other than A. Only that, had x done other than A, it would have been true that the past was different. That's a familiar story.

Alexander R Pruss said...

Mike:

Some compatibilists accept this story. Some just reject the principle of alternate possibilities and allow that one can act freely without having been able to do otherwise. In any case, I take it that this compatibilistic account of "could have done otherwise" is false. (Quick argument: This kind of argument seems to show that we could have made the past different.)

Alexander R Pruss said...

Mike:

Some compatibilists accept this story. Some just reject the principle of alternate possibilities and allow that one can act freely without having been able to do otherwise. In any case, I take it that this compatibilistic account of "could have done otherwise" is false. (Quick argument: This kind of argument seems to show that we could have made the past different.)

Scott Carson said...

The second premise appears to beg the question, though it may depend upon what your precise definition of "determinism" is.

Alexander R Pruss said...

Well, it's an argument for incompatibilism, so I'm entitled to assume compatibilism. :-)

Mike Almeida said...

Quick argument: This kind of argument seems to show that we could have made the past different

Well, we might find ourselves reheasing a very old story. But, as Lewis notes, you can't change the past, but you can do something such that, had you done it, the past would be different. I, for one, have never found the interminable talk of power entailment principles remotely moving or enlightening on this issue.

Scott Carson said...

It's hard to imagine what the motivation for the argument is, if not to defeat determinism, since the conclusion appears to deny determinism. If one of the premises of the argument assumes that determinism is not true (as the second premise appears to do), it seems that you are taking your entitlements a little too far.

This is why it seems to me to be important to be clear about what you mean by determinism. If, by determinism, you mean an ontological state in which free choices are not possible, then you cannot assert the second premise as part of your argument against determinism. But, of course, this need not be what is meant by "determinism". If something else is meant by it, however, then the argument will no longer support incompatibilism.

Alexander R Pruss said...

My feeling is that most contemporary philosophers accept the claim that determinism is compatible with free will.

Alexander R Pruss said...

Mike:

You can't change the past, but you also can't change the future. :-)

Alexander R Pruss said...

Mike:

The following also seems really plausible to me.

Suppose I know that:
(1) I can do A.
(2) If I were to do A, then p would hold.

Claim 1: If (1) and (2) hold, then I know exactly how to bring about p: viz., to bring about p, I would need to do precisely A.

Claim 2: If I know how to bring about p, and I can do that which I would need to do to bring about p, then I can bring about p.

Given determinism, it is in principle possible that I should know just how the past conditions would be different were I to do A. (Currently, we don't know enough physics to know that, but in principle I could know that.) If that were so, then (1) and (2) would be true for some p about the past. By Claims 1 and 2, it would follow that I could bring p about.

Mike Almeida said...

Suppose I know that:
(1) I can do A.
(2) If I were to do A, then p would hold.


Right, this might move intuitions to the conclusion that you cooud bring about the past. But what you should say in this case, so as not to be too misleading is that,

Suppose I know that:
(1) I can do A.
(2) If I were to do A, then it would have been the case all along that p.

I'm not changing the past at all in performing A. I'm doing something such that, were I to do it, the actual past would always have been different. It is never the case that the actual past is "changed". There is no change occuring in anything.

Alexander R Pruss said...

Yes, I'm not changing the past. But our intuitions are that there is a difference between our ability to bring about a past state of affairs and our ability to bring about a future state of affairs. (Namely, an ability to bring about a past state of affairs is rather limited. Maybe through prayer we can do it, and maybe there is something weird happening near space-time singularities.) But if (a) determinism holds and yet (b) we are able to do otherwise than we did, then we do not have this kind of a disanalogy.

That we cannot change the past does not seem to me to be relevant: that is symmetric between the past and the future, since neither can we change the future. If tomorrow I were to mow the lawn, then it would have been all along true that tomorrow I would mow the lawn. Our ability to affect the future is not, then, an ability to change the future, and so the admitted fact that we cannot change the past does nothing to show that we cannot affect the past.

(Minor qualification: People who accept an open future view think we can literally change the future--we can change it from being open to being closed in some particular way. However, the argument is aimed against determinists, and it is hard to imagine a determinist open futurist.)

Alexander R Pruss said...

Scott:

Sorry for my curt tone in an earlier comment.

I headed the post "An argument for incompatibilism." So the structure of the argument was, basically, to show that if we have free will, then determinism is false. This is what the compatibilist denies.

Now strictly speaking, I didn't show that. What I showed was that determinism is incompatible with free will being common. But most incompatibilists deny that, too.

Justin Capes said...

Alex,

Semi-compatibilists (SC) would almost certainly reject (1), for the following could be true: determinism holds, but since determinism only precludes the ability to do otherwise (the ability to do otherwise is not the same as acting freely, says the SC) we can act freely even if we never can do otherwise. Besides, we don't know whether determinism is true or false, says the SC, and since determinism (probably) precludes the ability to do otherwise, we don't know whether (1) is true.

That's what they'd say, I think.