More on the Consequence Argument

A howler in Harry Potter is an abusive letter that reads itself aloud, embarrassing the recipient. A howler in philosophy is a really bad mistake. If I'm about to commit a howler, don't send me any howlers.

Why don't I think that the Consequence Argument  (in the kind of context in which Fischer is moved by it, ie, a context in which what is in question is alternative possibilities, and not sourcehood) is a problem for compatibilism? One reason is that it seems to be to be over-general: it doesn't tell us anything about determinism and alternative possibilities. Here's one way of getting at the point. Suppose you think the following argument, or some suitable refined version of it, is sound

P

(P ⊃ Q)

Therefore, Q

Then it seems to be that you should also find the following argument sound:

P

(P ⊃ Q v R)

Therefore, Q v R

But the second argument is just a simple version of the Consequence Argument for indeterministic worlds, and therefore allowing for alternative possibilities. In other words, the 'no choice' operator does not depend upon the assumption of determinism for any force it has.

Comments

[Mark]

I'm not worried about the argument's form being over general. The argument is formally valid for P's that are positive propositions, but invalid for P's that are negative propositions. Since all versions of the consequence argument invoke a type of P that is, by definition, a negative proposition, they are all formally invalid. No amount of revision can get around that obtrusive detail...

1) From the non-existence of A, it is impossible to deduce the non-existence of B unless it is generally accepted or readily demonstrable that B requires A.

2) Any case where we can reasonably posit the existence of B, despite the presumed non-existence of A, serves as a counter-example to any argument to the effect that B-only-if-A.

3) Having posited an example of B without A, defenders of B-only-if-A must provide a substantive argument that discredits the counter-example by producing a coherent, constructive account that formally demonstrates the connection between B and A.

Myself and others have provided numerous counterexamples, on this blog and in print, to the consequence argument and to G. Strawson's Basic Argument that force the issue of formal validity to the forefront. The results is that there are only two potential classifications for these purportedly freedom undermining arguments: they are either superfluous or invalid.

If (3) can be achieved in any debate, then the B-only-if-A axiom, while valid, has no dialectical value since (3) would be self-sufficient. Moreover, lacking (3), the axiom still has no value: it shares no part in discovering whether (3) is achievable, except only insofar as it motivates the restless mind, and worse still it does not confers epistemic warrant on the desire to believe that it is B-only-if-A is true.

That's something I would like to hear more about from the incompatibilists. Perhaps Kip's attempt to "define" free will by survey is an attempt at producing (3). I doubt the effort will yield much fruit, but at least he's trying.

[Randy Clarke]

The Consequence argument doesn't conclude that the No Choice operator rules out ability to do otherwise, or that rule beta rules out this ability. The argument concludes that determinism rules out this ability. With that in mind, Neil, I don't see what worries you about the Consequence argument.

[John Fischer]

Right--I agree with Randy Clarke. Also, btw, he modal version of the Consequence Argument is just one version.

[Neil]

The howler alert was intentional, because I'm really not sure that this is a good argument. So I'm not surprised that my philosophical betters don't find it so. But I'm not seeing why it isn't a good argument. Of course determinism rules out alternatives, in some sense: that's what determinism *means*. I took the CA not to be telling is that we lack alternatives (in some sense) given determinism, but that given determinism we lack choice. I wanted to say that if that is what the CA is doing, then it over-generalises (or at least its modal version does).

[Neil]

I'm not imagining it: here's Van Inwagen on the modal version of the CA (from O'Connor's anthology): 'If this argument is sound, then determinism entails that no one has or ever had any choice about anything'. Now he also says that this amounts to saying that 'determinism entails that no one could ever had done otherwise'. But given that the revised CA yields the result that *indeterminism* entails that that no one has or ever had any choice about anything, then this can't amount to saying what Van Inwagen takes it to amount to; it can't actually be adding anything to the uncontroversial fact that determinism yields a single path into the future in *some* sense.

Take all this with salt to taste. Whenever I see formal logic I spend so much time wondering why the A's have got flipped that my brain turns to mush.

[Alan]

There is a difficulty here Neil--namely a dispute about whether the box distributes over disjunction. Different formal systems yield different results in this case.

[Randy Clarke]

Suppose that the laws and history until now leave open only these alternatives for you (a moment from now): your A-ing, and (instead) your B-ing. Let P say that you A, and Q say that you B. The Consequence argument will then conclude that you have no choice about whether the following disjunction is true: P or Q. But that's not to say that you have no choice about whether P; and it's not to say that you have no choice about whether Q.

There is perhaps a connection between what you're thinking, Neil, and van Inwagen's use of rule beta to state the Mind argument. But note that van Inwagen thinks that the Mind argument has a false premise (though he thinks it's a mystery how that premise can be false). (I think I'm remembering this correctly.)

[John Fischer]

Neil,

I think that you are correct that there's an intersting question about the dialectical role of the Consequence Argument, given that it is so obvious that causal determinism rules out one salient sort of freedom--the freedom to do otherwise, holding fixed the laws and past. One might wonder: what's the big deal? If I'm right about what you're worrying about, I am actually working on a paper on just this point, which I'm contributing to a symposium on moral responsibilitiy edited by our very own Joseph Campbell (in Journal of Ethics). It is not done yet, but soon...

[Neil]

Randy, thanks. I think that I was right (about the howler, that is).

[Joe Campbell]

I was going to contribute but Randy said everything I was going to say -- only better.

Thanks, John! Should be a great volume!