Philosophy 257 Behavior and Other Minds

Asst. Prof James Pryor Dept. of Philosophy

Putnam

Inference to the Best Explanation

• These hypotheses (or theories which presuppose them) provide plausible explanations of many phenomena.

• No worked-out plausible alternative theories are currently in the field. For example, the demon hypothesis has "never been elaborated in any detail at all, and it obviously is too silly to consider" (pp. 359-60).

  If a theory has genuine explanatory power, we ought not to give it up unless an alternative is in the field. The fact that we know of no plausible alternative theories is "itself our deepest justification for staying with the accepted explanations" (p. 345).

• The assumption that other people have minds is not itself an explanatory theory, and it has never been tested against explanatory theories which deny the existence of other minds. Compare:

  What has been tested is not thing-theory as opposed to "no-thing" theory, but thing-theory as opposed to alternative thing-theory. We have inductively established not that material objects exist, but that this account of how material objects behave is more probable than some other account of how material objects behave. (p. 344)

• It is "built into" the language we use to make observations reports that other people have mental states; this language is not "neutral" with respect to the existence of other minds (p. 346). In a general explanatory induction, on the other hand, the data should be characterized in terms that are neutral between the competing explanations.

  It's not clear what Putnam is saying about the language we use to make observation reports:

  • He could mean that the vocabulary we use to make observation reports is only comprehensible to someone who accepts the hypothesis that there are other minds.
  • He could mean that we describe the behavior we observe in terms of which mental states that behavior is allegedly evidence for.
  • He could mean that we describe the behavior we observe in terms that presuppose the existence of other minds.

  Would each of those three claims, if true, represent a disanalogy between the hypothesis of other minds and scientific hypotheses? Would any of the claims, if true, make it illegitimate to appeal to the behavior we've described in an explanatory argument for the existence of other minds?

Against Criteria Theory

Putnam distinguishes several different questions:

1. Could someone understand the meaning of our word "anger" without having learnt that certain forms of behavior are thought by us to justify the ascription of anger to others?

   Putnam asks:

   1. Must one learn the meaning of terms like "anger," or could one be born with an ability to speak and understand that fragment of language?

   2. If one does learn the meaning of our word "anger," and in the process learns how to apply this word to others, must one apply it to others on the basis of the behavior we take to be evidence of anger? Or would other sorts of behavior also be OK? (Putnam also asks whether it would be OK to apply it to others on the basis of neurophysiological data, rather than behavior.)

      We all know the story about people who are supposed to grimace when happy and smile when sad. If the people in those cultures spoke English, I think it would not be correct or customary to say that the words "happy" and "sad" have a different meaning when used among those English speakers. If one of those speakers sees Jones grimace and says "I see that Jones is happy today," it's not that he uses the word "happy" with a noncustomary meaning; it's just that he believes, and indeed, has good reason to believe, that people normally grimace when they are happy. (p. 352)

      A person who is obviously a native speaker of (correct) English could fail to know that people who scream are usually in pain, that people who smile are usually happy, and so on, and it might still be the case that he knew the meaning of all these words and used them with the same meaning that we do. That is to say, normal English speakers, when given sufficient information about the kind of culture that this speaker comes from, might decide to account for the difference in what he says not as owing to a difference in the meaning of words, but simply as owing to a difference in the behavior that he is accustomed to. No inference of the following form can be valid: Jones knows the meaning of the word "pain"; Jones knows that if someone winces, screams, writhes, etc., normally, one is entitled to conclude that he is in pain. (p. 353)

      If Putnam is right, then it can't be required to learn the meaning of our words "anger" and "pain" that one take the behavioral evidence we use to be good evidence for ascribing these mental states. At most, it could only be required that one believe of some evidence or other that it's good evidence--or standardly thought to be good evidence--for ascribing these mental states.

      But now we face the question: is it required that one know (or believe) that the evidence in question is in fact good evidence? Or is it sufficient if one merely knows that one's local linguistic community takes it to be good evidence? This is Putnam's next question.

2. Could someone understand the meaning of "anger" without having learnt that certain forms of behavior do in fact justify the ascription of anger to others?

   Putnam argues that the skeptic needn't fail to understand our psychological vocabulary, or use it with a different meaning, just because he refuses to believe that the standard sorts of behavior count as good evidence for ascribing the relevant mental states:

   If a person is confronted with a clear case of another person who is angry--with someone showing unmistakable signs of anger, or egotism, or joy, or whatever--and fails to note that the person is angry, he may be suspected of not knowing the meaning of the word. At first blush, this is very convincing, but only at first blush, for we see when we look more carefully, that the person is only suspected of not knowing the meaning of the word "angry," or whatever, if he is unable to give a reason for not agreeing that the subject in question is angry, or whatever, or if we are unaware that he possesses such a reason... If it turns out that that is not the way in which angry people act in the speaker's culture, then the suspicion that the speaker does not know the meaning of the word "angry" lapses... [If the speaker offers the reason,] "That's only behavior, and no amount of behavior entitles one to apply a psychological predicate to another," then again the suspicion that he does not know the meaning of the word "angry" lapses, and is replaced by the conviction that the speaker is a skeptic. (pp. 356-7)