Newsgroups: comp.ai.philosophy
Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!news2.near.net!MathWorks.Com!europa.eng.gtefsd.com!howland.reston.ans.net!cs.utexas.edu!news.unt.edu!hermes.oc.com!internet.spss.com!markrose
From: markrose@spss.com (Mark Rosenfelder)
Subject: Re: Folk psychology (was: Is Common Sense Explicit or Implicit?)
Message-ID: <Cwswyp.Inz@spss.com>
Sender: news@spss.com
Organization: SPSS Inc
References: <1994Sep21.131455.3228@oracorp.com> <35q0l5$mgr@mp.cs.niu.edu> <CwJKq8.7n9@spss.com> <35sop3$1v1@mp.cs.niu.edu>
Date: Tue, 27 Sep 1994 18:35:11 GMT
Lines: 52

In article <35sop3$1v1@mp.cs.niu.edu>, Neil Rickert <rickert@cs.niu.edu> wrote:
>In <CwJKq8.7n9@spss.com> markrose@spss.com (Mark Rosenfelder) writes:
>>Whoa.  The size of the gap between these statements is itself remarkable
>>and calls out for explanation.  How can Daryl think that folk psychology
>>has "enormous" predictive power, and Neil think it's "negligible"?
>>It's hard to believe that the two of you have wildly different personal
>>experiences with prediction; it seems more likely that you're talking
>>about two different things somewhere along the line. [...]
>
>Clearly, there is a miscommunication here.  We must be talking about
>quite different things.  I agree that we are reasonably good at
>predicting other people's behavior.  The question is whether folk
>psychology has anything to do with this ability.
>
>I take folk psychology to be the theory that we predict the behavior
>of others by using beliefs, desires, and other propositional
>attitudes.  That is quite separate from the question of whether we
>are able to predict.  

Hmm, I think your definition is a bit quirky.  I don't see that folk
psychology is committed to the notion that its postulated entities
are all propositions.  We might explain someone's behavior by saying
that she's in love, or believes in God, but that doesn't mean that
in folk psychology love or faith are propositions.

>If folk psychology really has any substantial
>predictive power, it should be possible to test it with controlled
>experiments.  The experiment would have to be done so that those
>making the predictions only had access to written representations of
>the propositional attitudes, and had no visual or other contact with
>the people whose behavior they are predicting.  If some good
>controlled experiments have been conducted, I would be quite
>interested in references.

What you're testing here is so far from the way that people actually make
predictions that it's hard to see why you call it "folk psychology";
you should find another, clearer term.

The analyses or predictions of folk psychology can be *reported* in
language, of course, and people can even make further predictions based
on these reports.  But folk image analysis-- that is, sight :)-- can be
described in words, too, and it's even alleged that people can form
mental images of what they read about in books.  (I don't know that I
can, but I don't have a very visual imagination.)  The faculty itself
doesn't depend on language in either case (unless you're Sapir and Whorf :).

>[...] Similarly, folk psychological
>explanations of human behavior might be only a just so story.

Very possibly.  However, it's a much better story than "nature abhors
a vacuum".  And why shouldn't it be?  We vertebrates have been observing
each other far longer than we've been thinking about vacuums.
