Saturday, April 12, 2008

Quaaalia.... rrrr....

Richard (Chappell) in reply to the below makes three points. First, he offers a case which is supposed to parallel mine. Second, he tries to support the use of intuition-pumps in philosophy. Third, he discusses rational persuasion. Shockingly, I find none of his reply convincing. Isn't that amazing? I know I was surprised: philosophers not agreeing with each other. Next we'll use terms like "rigid designator" and not understand why everyone else is snickering.


The case he gives is this. Suppose we define "bachelor" as "unmarried man". A putative counter-example is the Pope, who is unmarried but not, in some sense, a bachelor. It would be odd to complain, Richard says, that the counter-example just presumes the definition is false. I don't see this at all. It looks perfectly legitimate to me to complain that the counter-example is presumptive. Indeed, if the counter-example weren't followed by an argument as to what it is about the Pope that should lead us to not count him as a bachelor, then the counter-example strikes me as perverse.

The point Richard is trying to illustrate with this case is more interesting than the details of the case. He's claiming that provocative thought-experiments can show us the presumptions that were inherent in the positions they are counter-examples against, and thus they are philosophically useful. Raising the example of the Pope would expose something else about the concept "bachelor" that wasn't captured by the concept "unmarried man"; and similar remarks apply to zombies, amoralists, and the rest. That's fine, as far as it goes. My point was not to deny this; indeed, I accepted it, and went further, arguing that provocative thought-experiments (except for the direction of fit mess) can open up new avenues of useful philosophical investigation. My point instead was that this pointing that the intution-pumps do, all on its own, is not philosophically interesting work. That is, if this is all the philosophical work being done by entertaining zombies and Frankfurtian demons, if all that's being done is provocation -- which, as I understand it, is the very definition of an intuition-pump -- we go nowhere. Provocation is only any good if we do something else with it.

Intuition-pumps don't go very far because, most obviously, it's always legitimate to dig in one's heels in response to the pumping and deny the relevant intuitions. And this is not always disingenuous or ad hoc; if experimental philosophy is doing anything useful, it's at least showing that intuitive mileage will vary. Which means that the attempted pumper is guilty of begging the question. And which means the dialectic has not been advanced one jot. All intuition-pumps need to be attached to arguments that show the intuitions being pumped are ones that should be taken seriously.

On this point, though, Richard calls into question the sense in which something, such as an intuition-pump, begs the question, by appealing first to formal deductive validity. He is correct, of course, that any formally valid argument is such that the premises contain the conclusion. This does not mean, in any sense, that valid arguments beg the question. Strictly, begging the question requires stating or assuming as a premise the exact conclusion -- anything less doesn't turn the argument in a tight enough circle to be vacuous reasoning (which is, after all, the real problem with begging the question, as Richard notes -- indeed, it's the problem with all fallacious arguments). So, implying the conclusion isn't question-begging. Which means the appeal to deductively valid arguments is a bit of a red herring, or at least a bright purple porpoise. (See what I did there?)

Richard then claims, more interestingly, that intuition-pumps are useful because they are rationally persuasive, at least to some. He points out that many, upon learning of things like zombies, do find the cases persuasive and accept the implied intuitions. I don't find that observation terribly persuasive, though; really, all it establishes is shared intuitions. Amazingly enough, people with roughly similar cultural and educational backgrounds share some of the same intuitions. Again, if experimental philosophy is doing anything useful, it's getting us out of the habit of appealing to those intuitions rather than intuitions read more broadly in the folk.

Let's talk about rational persuasion, though. As I understand it, an argument is rationally persuasive iff the premises of the argument give a rational agent sufficient reason to accept the conclusion as true. From this, though, it will follow that deductive validity is rarely particularly persuasive, for at least three reasons. First, the premises in a given valid argument are rarely if ever established themselves by valid arguments. Consider the classic syllogism:
  1. If Socrates is a man, Socrates is mortal.
  2. Socrates is a man.
  3. Therefore, Socrates is mortal.
Clearly valid, but not persuasive, except insofar as (1) and (2) can be established. (1) is possibly a conceptual truth, i.e., necessary; but (2) is an empirical claim which requires testing and thus can only appeal to inductive or abductive reasoning. Which means that this deductive argument is only going to persuade insofar as the non-deductive support for (2) is persuasive.

Second, there is a presumption -- a standard philosophical prejudice, but one I don't find compelling -- that accepting deductively valid arguments is part of rationality. That is, anyone who is rational will be persuaded by a deductively valid argument to accept its conclusion as true. Perhaps I'm too much of a pragmatist these days, but I don't think that any deductively valid argument is necessarily persuasive, in the above sense. The "so what?" response is often a reasonable option. Taking the example above again: so what if Socrates is a man? Unless you're a historian interested in the life of Socrates (or in some other way have a project which connects to this argument's conclusion), it seems perfectly rational to ignore the argument, and neither accept nor reject its conclusion. Suspension of judgement is a rationally permissible attitude and, indeed, it may often be epistemically obligatory.

Third, there is another presumption that deductive validity is somehow more fundamental -- and thus a basic standard of rationality. And this, too, is suspect. Deductive validity strikes me as a bastardized form of explanatory reasoning: that is, deductively valid arguments are explanations which provide purely logical understanding. This is not useless, but it is of limited interest to any but formal philosophers.

If this third point is right, then rationally persuasive arguments are going to be explantorily superior arguments: they will present more interesting hypotheses, unify phenomena under fewer natural laws, offer more understanding, and so on. These will only rarely be deductive; as (for example) Hempel found, to his mounting horror, there's not much a deductive argument will do for you even if you limit yourself to nomological explanations of scientific observations. Explanatory considerations that stretch more broadly, such as whether the qualia hypothesis is part of or apart from materialism/physicalism, are going to be even more poorly served by appeals to deductive reasoning.
Intuition-pumps aren't explanatory. They aren't hypotheses. They aren't laws. They're, at best, metaphors. They, as Richard says, offer conceptual possibilities. Conceptual possibilities can, as I pointed out previously, lead us to better understanding by opening up useful avenues of investigation. But possibilities on their own are completely inert. So, to go back to the original case, zombies tell us nothing about consciousness. What tells us something about consciousness is trying to find an actual zombie. If we can find one, we've learned something. Or, trying to find the physical structure in the brain or CNS responsible for our conscious experiences. If we can find that, we've learned something tremendous. And, if we find that the brain, CNS, and the entire physical world is simply not capable of realizing consciousness, then we've found something good, too.

But that is where the philosophical action really happens. Intuition-pumps like zombie cases, the amoralist, and Frankfurt-style examples point somewhere, but you have to actually go there. I suspect strongly that this is Richard Brown's point against Richard Chappell, i.e., that the zombie case is not an argument, and hence does no philosophical work unless attached to an argument.

(Aside: this is the only thing I miss about Rorty. He had an odd genius for spinning out compelling stories -- and then actually advancing an argument on their back. The example of the Antipodeans in Philosophy and the Mirror of Nature is a good case in point. I don't buy it, but at least he's trying.)

No comments: