From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!sun-barr!ames!elroy.jpl.nasa.gov!sdd.hp.com!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!bronze!chalmers Mon Mar  9 18:33:30 EST 1992
Article 4104 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!sun-barr!ames!elroy.jpl.nasa.gov!sdd.hp.com!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!bronze!chalmers
>From: chalmers@bronze.ucs.indiana.edu (David Chalmers)
Newsgroups: comp.ai.philosophy
Subject: Re: making sense of eliminative materialism
Message-ID: <1992Feb27.234116.18196@bronze.ucs.indiana.edu>
Date: 27 Feb 92 23:41:16 GMT
References: <1992Feb21.200219.3773@ux1.cso.uiuc.edu>
Organization: Indiana University
Lines: 55

In article <1992Feb21.200219.3773@ux1.cso.uiuc.edu> rpoldrac@s.psych.uiuc.edu (Russ Poldrack) writes:
>
>I am faced with the onerous task of defending eliminative materialism
>at at lab meeting next week with a bunch of psychologists.  The one
>stumbling block I have to understanding and making a case for it has
>to do with Fodor's 1975 argument against reduction.  He claims that
>reduction to neuroscience will be impossible because the resulting
>science of mind would fail to capture interesting generalizations
>(i.e. functional states like propositional attitudes).  How do the
>Churchlands answer this criticism?  The most obvious answer to me ist
>that the intuitions that we think must be captured (that is, the
>generalizations that we think must be captured) come from within folk
>psychology and thus the criticism presupposes the foundations of folk
>psychology that we are trying to destroy.  But I can't see how
>psychologists are going to buy this.  

Well, an argument against reduction doesn't carry any weight against
eliminativism, because reduction presupposes the truth of the
higher-level theory, whereas elimination presupposes its falsity (or
inadequacy), whuch is more or less what you say above.

Even so, the Churchlands aren't nearly such radical eliminativists as
they're usually made out to be, and they certainly want to allow the
possibility that at least some parts of folk psychology will be
reducible (I think they prefer the term "revisionist materialism" to
"eliminate materialism", but the second caught on).  So presumably
they still don't buy Fodor's argument.  There are various ways in
which it might be addressed.

The most obvious is to say that they're not looking for an explanation
of "universal psychology", i.e. the psychology of all possible
organisms -- whatever that might be (I think it would have to consist,
more or less, of analytic truths about psychology, so it wouldn't be
very interesting).  Rather, they want an explanation of *human*
psychology, so multiple realization arguments (or at least cross-species
realization arguments) are irrelevant. 

What still might be relevant is the possibility of multiple
realization of high-level states *within* a species; e.g.
psychological state P might be realized by neural state X in this
person, and neural state Y in another, and maybe even by different
states in the same person at different times.  But the reductionist
might answer that simply by casting their neural states at a more
abstract level.  Of course neuropsychologists don't *have* to
reduce to specific neurons, or even neural areas; they can cast 
their neural states at a more abstract level, e.g. in terms of
certain kinds of patterns, or of functional mechanisms such as
tensor computation, all of which might in turn be multiply
realizable at the very lowest level.  I think it's something like
this kind of explanation that the Churchlands favour.

-- 
Dave Chalmers                            (dave@cogsci.indiana.edu)      
Center for Research on Concepts and Cognition, Indiana University.
"It is not the least charm of a theory that it is refutable."


