From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rutgers!cs.utexas.edu!uunet!boulder!csn!cherokee!dakota!ken Mon Dec  9 10:47:45 EST 1991
Article 1834 of comp.ai.philosophy:
Xref: newshub.ccs.yorku.ca rec.arts.books:11028 sci.philosophy.tech:1271 comp.ai.philosophy:1834
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rutgers!cs.utexas.edu!uunet!boulder!csn!cherokee!dakota!ken
>From: ken@dakota (Kenny Chaffin)
Newsgroups: rec.arts.books,sci.philosophy.tech,comp.ai.philosophy
Subject: Re: Zeleny (was Re: Searle
Message-ID: <1991Dec3.181458.18420@cherokee.uswest.com>
Date: 3 Dec 91 18:14:58 GMT
References: <1991Dec1.113406.6027@husc3.harvard.edu> <12538@pitt.UUCP> <1991Dec2.110629.6077@husc3.harvard.edu>
Sender: news@cherokee.uswest.com (Telegraph Row)
Distribution: world,local
Organization: U S WEST Advanced Technologies
Lines: 35
Nntp-Posting-Host: dakota.uswest.com

In article <1991Dec2.110629.6077@husc3.harvard.edu> zeleny@zariski.harvard.edu (Mikhail Zeleny) writes:
>In article <12538@pitt.UUCP> 
>geb@dsl.pitt.edu (gordon e. banks) writes:
>
>>In article <1991Dec1.113406.6027@husc3.harvard.edu> 
>>zeleny@zariski.harvard.edu (Mikhail Zeleny) writes:
>
>GB:
>>>>There have been some connectionist programs (such as Sejnowski's)
>>>>that have learned to speak words.  These symbols were learned
>>>>by reinforcing certain stimuli, so I suppose you could say they
>>>>"refer" to the stimulus in some way.  In a more complex computer,
>>>>such as the brain, the referents would be multiple and complex.
>
>MZ:
>>>I would argue that the program's words merely refer to its internal
>>>representation (under our interpretation, of course) of the stimulus, as,
>>>regardless of how such a representation occurred, the program's response
>>>would be the same.
>
>GB:
>>And how would that contrast with our words refering to our internal
>>representation of the objects and abstractions that we are symbolizing
>>by them?
>
>Unlike internal representations like sense-data, abstract objects like
>propositions, expressed by our words, or numbers, denoted by them, can be
>meaningfully used in communication with those not enjoying privileged
>access to our mental processes.
>
	Right, but that is only a level of abstraction, a method of 
communicating, and a shared semantics. It has no bearing on the fact that
symbols either represent something or not.

KAC


