From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!neat.cs.toronto.edu!maione Mon May 25 14:07:11 EDT 1992
Article 5847 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!neat.cs.toronto.edu!maione
Newsgroups: comp.ai.philosophy
>From: maione@cs.toronto.edu (Ian Christopher Maione)
Subject: Re: Comments on Searle - What could causal powers be?
Message-ID: <92May22.150955edt.47997@neat.cs.toronto.edu>
Organization: Department of Computer Science, University of Toronto
References: <1992May17.225236.3588@oracorp.com> <1992May18.185935.13789@psych.toronto.edu>
Date: 22 May 92 19:10:28 GMT
Lines: 46


In article 6671 Michael Gemar writes .....

>I take it then that you agree with Harnad that, without such real-world
>connections, there can be no semantics.  Thus, Searle is at least right
>in saying that symbol manipulation in and of itself is insufficient for
>semantics.

>This position still seems to leave SHRDLU's Dilemma, the problem of a
>program whose "world" is "a purely mathematical object".   Could SHRDLU
>ever have semantics?  As I interpret your claim above, it could not,
>simply because it has no interface with the "real" world.  However,we can
>imagine that SHRDLU's internal "World" can mirror ours to an arbitrary
>level of accuracy.  This seems to lead to a problem - the "real" world
>generates meaning, where a reproduction of that world to an arbitrary
>level of detail in a computer does not.  Something seems wrong somewhere..

   Perhaps it would be illuminating to consider a different situation,
where we imagine going into a person's brain and disconnecting all the
connections between their brain and their various senses, and connect
them all up instead to our Cray version 1Million, which has the capability
of creating nerve impulses corresponding to any sensory experience we
wish.  We can now make our poor victim experience any 'reality' we
choose.  That world, of course, does not exist in the same way the
'real' one does, but it can (in principle) be described in terms of
the requisite states of the computer.

    Now our poor fellow may say "I see a 30ft tall purple goose in
front of me", but he is of course mistaken.  If he were to say
"The Cray is in state X" for the proper X, he would be making a true
statement.  Actually, I suppose almost all of his beliefs would be false,
except for any perceptual ones he has.  The point is that the person is
experiencing the "real world", just in such a strange manner that he
is really talking about states of a computer when he thinks he is
talking about giant purple geese.  Now perhaps a similar, but less
dramatic account could be applied to SHRDLU.  I don't think one can
say that SHRDLU's virtual world is a "mathematical object", for the
same reason one can't say a running program is (at least wholly) a
mathematical object.  SHRDLU has a 'world', which is not one of
blocks, but one of electrical states of a computer.

Regards,
Ian





