From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!bronze!chalmers Tue Mar 24 09:56:10 EST 1992
Article 4494 of comp.ai.philosophy:
Newsgroups: comp.ai.philosophy
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!bronze!chalmers
>From: chalmers@bronze.ucs.indiana.edu (David Chalmers)
Subject: Re: The Systems Reply I
Message-ID: <1992Mar17.020503.9967@bronze.ucs.indiana.edu>
Organization: Indiana University
References: <1992Mar14.213045.21776@mp.cs.niu.edu> <1992Mar15.011107.7828@news.media.mit.edu> <1992Mar16.171520.15584@psych.toronto.edu>
Date: Tue, 17 Mar 92 02:05:03 GMT

In article <1992Mar16.171520.15584@psych.toronto.edu> christo@psych.toronto.edu (Christopher Green) writes:

>What troubles
>me is a lot of CompSci types sitting around ragging on Searle (who
>is well aware of these matters -- hell, he developed some of the issues)
>without knowing what syntax and semantics are. Its one thing to say,
>'Gee, I don't really understand the difference between syntax and semantics.
>Perhaps I better go find out about it if I'm to understad Searle's argument.'
>It's quite another to say, as many on this group seem to say (and I'm *not* 
>specifically including Minsky -- who got me started on this -- among these),
>'This sytax/semantics stuff isn't very clear to me. Therefore it must be a
>lot of pseudo-intellectual mumbo-jumbo.' That just sheer anti-intellectualism.

Throwing around various references to the philosophy of language is all
very well, but the fact is that even in the most sophisticated
philosophical discussions there's no sufficiently well-developed
understanding of the notions of "syntax" and "semantics" that allows you
to say that "syntax is not sufficient for semantics" is some kind
of widely accepted truth.  As far as I can tell, the question is rarely
discussed.  Of course there's no doubt that in the philosophy of
linguistics, say, syntax and semantics are understood to be very
different things, but its very unclear how issues about sentences'
static syntactic form have much bearing on the dynamic manipulations
performed by computers.  To subsume all these things under the loose
term "syntax" is simply to obfuscate the issues.

>For the heart of the current debate on semantics, take a look at
>Fodor's "Methodological solipsism" (BBS, 1981?), _Psychosemantics_
>(in which he changes his mind about solipsism), and _Theory of content_;
>Putnam's _Representation and reality_; and Millikan's _Language and other 
>biological concepts_. (or is it "categories"?). An interesting
>side trip consists of Dretske's _Knowledge and the flow of information_,
>followed by Fodor's "Semantics Wisconsin style".

It's very amusing that you bring these up, because common to most
of these authors is the notion that semantics essentially comes
down to correlation, or covariation (plus or minus some bells
or whistles).  Computers that were appropriately programmed, and
causally connected to the world, would have little trouble
satisfying these theories of semantics (certainly on Fodor's and
Dretske's accounts, and probably on Millikan's, if we allow a
little simulated evolution).

These are just the kinds of theories that Michael Gemar would scream
something like "but that's a *syntactic* notion of information"
at (as I saw in a recent post).  Very well, but then you'll have
to accept that most of the explanations of semantics that are
currently on offer base semantics on these "syntactic" grounds.
Whether these are adequate or inadequate, you certainly can't appeal
to anything like "accepted philosophical wisdom" in support of your
case.

-- 
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."


