Some philosophical claims are, or at least seem to be, self-defeating. For example, an eliminative materialist who asserts that there is no such thing as meaning or semantic content is implying thereby that his own assertion has no meaning or semantic content. But an utterance can be true (or false) only if it has meaning or semantic content. Hence the eliminative materialist’s assertion entails that it is itself not true. (I’ve addressed this problem, and various futile attempts to get around it, many times.) Cognitive relativism is also difficult to formulate in a way that isn’t self-defeating. I argue in Scholastic Metaphysics that scientism, and Hume’s Fork, and attempts to deny the existence of change or to deny the principle of sufficient reason, are also all self-defeating. This style of criticism of a position is sometimes called a retorsion argument.
At a conference I was at not too long ago, one of the participants suggested that an appeal to a retorsion argument amounts to a tu quoque fallacy (or “appeal to hypocrisy”). A tu quoque fallacy is committed when someone rejects a claim merely because the person advancing the claim acts in a way that is inconsistent with it. For example, if I try to convince you that it is not a good idea to drink to excess and you say “But you drink to excess all the time! Therefore I can dismiss what you’re saying,” you would be committing a tu quoquefallacy. That someone is a hypocrite doesn’t show that what he is saying is false. But don’t retorsion arguments amount to such an appeal to hypocrisy?
No, they don’t. As every logic and critical thinking teacher knows, one of the problems one encounters in teaching about the logical fallacies is that students often settle into too crude an understanding of what a fallacy involves, and thus tend to see fallacies where there are none. Not every use of language which has emotional connotations amounts to a fallacy of appeal to emotion. Not every attack on a person amounts to an ad hominem fallacy. Not every appeal to authority is a fallacious appeal to authority. A reductio ad absurdum argument should not be confused with a slippery slope fallacy. And so on.
In the same way, by no means does every reference to an opponent’s inconsistency amount to a tu quoque fallacy. On the contrary, pointing out that a certain view leads to inconsistency is a standard technique of logical criticism. It is, for example, what a reductio ad absurdum objection involves, and no one can deny that reductio is a legitimate mode of argumentation. The problem with tu quoque arguments isn’t an appeal to inconsistency as such. The problem is that the specific kind of inconsistency the arguer appeals to is not relevant to the specific topic at issue.
So, suppose I am a drunkard but I tell you that it is bad to be a drunkard, on the basis of the fact that being a drunkard is undignified, is damaging to one’s health, prevents one from holding a job and providing for one’s family, etc. My hypocrisy is irrelevant to the truth of the claim I am making, because the proposition:
(1) Feser is a drunkard.
is perfectly compatible, logically speaking, with the proposition:
(2) It is bad to be a drunkard.
and perfectly compatible also with the proposition:
(3) Being a drunkard is undignified, is damaging to one’s health, prevents one from holding a job and providing for one’s family, etc.
Hence to reject (2), or to reject the argument from (3) to (2), on the basis of (1), is unreasonable. But that is exactly what the person in my example who commits the tu quoque fallacy does.
A retorsion argument is not like that at all. Consider the objection, raised against Eleatic philosophers like Parmenides or Zeno, that they cannot coherently deny that change occurs. The idea here is that the Eleatic is committed to the proposition:
(4) There is no such thing as change.
but at the same time, carries out an act -- for example, the act of reasoning to that conclusion from such-and-such premises, where this very act itself involves change -- which entails the proposition:
(5) There is such a thing as change.
Now (4) is not logically compatible with (5). What we have here is a “performative self-contradiction” in the sense that the very act of defending the position entails the falsity of the position. So, it is not mere hypocrisy, but rather implicit logical inconsistency, that is at issue.
Here’s another way to think about it. Could being a drunkard still be a bad thing, even if I am in fact a drunkard myself? Of course. That’s why it is a tu quoquefallacy to reject my claim that being a drunkard is bad, merely because I am myself a drunkard. But could change really be an illusion, if Parmenides is in fact reasoning from the premises of his argument to the conclusion? No. That’s why it is not a tu quoque fallacy to reject Parmenides’ denial that change occurs on the basis of the fact that he has to undergo change himself in the very act of denying it.
Of course, Parmenides might respond: “Ah, but that assumes that I really am reasoning from premises to conclusion, and I would deny that I am doing so, precisely because that would be an instance of change! So, you are begging the question against me!”
But, first, even if the critic’s retorsion argument against Parmenides did amount to begging the question, it still would not amount to a tu quoque fallacy.
And second, it does not in fact amount to begging the question. The critic can say to Parmenides: “Parmenides, you were the one who presented me with this argument against the reality of change. I merely pointed out that since the rehearsal of such an argument is itself an instance of change, you are yourself already implicitlycommitted to its reality, despite your explicit denial of it. I am pointing out a contradiction in your own position, not bringing in some question-begging premise from outside it. So, if you want to rebut my criticism, it is no good for you to accuse me of begging the question. Rather, you have to show how you can restate your position in a way that avoids the implicit contradiction.”
Of course, no such restatement is forthcoming, because the very act of trying to formulate it would involve Parmenides in exactly the sort of implicit contradiction he was trying to avoid. But that would be his problem, not his critic’s problem. (Eliminativism is, of course, in exactly the same boat -- as I show in some of the posts linked to above -- as are some of the other claims against which retorsion arguments might be deployed.)