Philosophy 257 Behavior and Other Minds

Asst. Prof James Pryor Dept. of Philosophy

Malcolm

Malcolm on Hampshire and Price

Hampshire says:

I can observe others' successes and mistakes in inferring what feelings and experiences I have; and from this I can confirm which methods of inference are reliable.

Malcolm responds:

Until I've established that others have minds, I'm not allowed to assume that the sounds they emit are "inferences." So Hampshire's suggestion is no advance on the argument from analogy.

Price says:

If X makes noises which seem intelligible, then it's reasonable to believe X has a mind.

Malcolm responds:

Nothing could establish that a tree has a mind, since a tree has nothing like a human face and body, and a tree can't pass or fail the tests that a child is put through to determine whether he understands the words he utters. (These tests require looking, pointing, reaching, and so on.)

Knowing What Pain Is From One's Own Case

• Does the claim that one learns what pain is from one's own case lead to solipsism?
  1. Malcolm says that if you learn what pain is from your own case, then you have to think of behavior as merely an accompaniment or effect of pain, rather than a criterion for pain.

  2. But if you have no criterion for pain in others, then you have no idea what would count as evidence for or against someone else's being in pain.

     Why should the fact that one has no criterion for pain in others show that one can have no conceivable evidence for pain in others? This just assumes that non-criterial approaches to the epistemology of other minds will fail. That's quite a hefty assumption to make in an argument against the argument from analogy! (See Notes on Criteria.)

  3. If you have no idea what would count as evidence for someone else's being in pain, then the possibility of someone else being in pain is not intelligible for you.

     Why should the fact that one can have no conceivable evidence for p show that one can't understand what it is for p to be true? Perhaps this is defensible; but it sounds pretty verificationist.

• Does solipsism lead to nonsense?

  Malcolm reconstructs this argument as follows: Solipsism commits one to the possibility of a private language. However, a private language is not possible. A speaker of a private language would have no way to tell whether he's applying the terms of that language truly, or whether he's merely under the impression that he is. So, Malcolm concludes, he would have no idea of what a mistake would be. (Again, this smacks of verificationism. There are other interpretations of the private language argument which don't rest on verificationist assumptions.) So the notions of correct and incorrect applications of a term don't make sense for a private language. Hence, a private language is not possible.

  For more discussion of whether a private language is possible, see Edward Craig's "Meaning and Privacy" (on reserve in Robbins).