From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!uwm.edu!psuvax1!psuvm!jpe1 Thu Apr 16 11:33:41 EDT 1992
Article 5012 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!uwm.edu!psuvax1!psuvm!jpe1
Organization: Penn State University
Date: Thursday, 9 Apr 1992 12:56:06 EDT
>From: <JPE1@psuvm.psu.edu>
Message-ID: <92100.125606JPE1@psuvm.psu.edu>
Newsgroups: comp.ai.philosophy
Subject: Re: syntax and semantics
References: <92098.170625JPE1@psuvm.psu.edu>
 <1992Apr8.010858.5398@news.media.mit.edu> <92099.165657JPE1@psuvm.psu.edu>
 <1992Apr9.040057.4784@news.media.mit.edu>
Lines: 60

In article <1992Apr9.040057.4784@news.media.mit.edu>, minsky@media.mit.edu
(Marvin Minsky) says:
>
>In article <92099.165657JPE1@psuvm.psu.edu> <JPE1@psuvm.psu.edu> writes:
>>In article <1992Apr8.010858.5398@news.media.mit.edu>, minsky@media.mit.edu
>>(Marvin Minsky) says:
>>>
>>>Nonsense. What the computer does is what it does, just like a brain.
>>
>>   What does this mean?  That a computer does whatever it does (which tells
>>us nothing) or that a computer does whatever a brain does (which is a central
>>issue here, not something that can be simply assumed or stated without
>>support)?
>>
>>>It is _you_ who may interpret what it is doing as formal or precise.
>>
>>   Does this mean that you reject the isomorphism that many philosophers
>>find between computers and formal systems?  And if so, on what grounds?
>>(Directing me to an article would be an acceptable response here, although
>>some stated reasons would be easier to discuss on the net.)
>
>I mean that a computer is a machine, in practice finite state.  Its
>behavior can be described in many ways, formal and informal.  And of
>course one could use, in particular, various kinds of monogenic
>production rules.  But my point was simply to object to using terms
>like "syntactic" and "formal" -- along, I fear, with their various and
>curious implications -- as though these were substitutes for simpler
>and less ambiguous terms like "finite-state" or "deterministic
>machine".  To be sure, if you yourself don't subscribe to those hidden
>assumptions, then no harm is done.  However, there appear to be those
>who are convinced that "syntactic" means "devoid of semantic" --
>without substantial justification for the implication that
>finite-state (or, in Penrose's view, "algorithmic") machines cannot
>have any degree of meaning, understanding, consciousness, etc.  What
>do you mean when you say a machine is "syntactic"?

     I hope it's not too impolite to direct you to my other posts in this
thread (my discussion with Neil Rickert) for what I'm trying to say by
calling computers syntactic.  Basically, I mean no more than is meant
by calling them formal systems.  I believe that 'finite state' or
'deterministic machine' are also equivalent terms in this context.
I am at this point unconvinced either way as to whether this means that
computers could ever have consciousness.  Clearly, I think, humans, as objects,
are also deterministic machines.  The question then is how these particular
deterministic machines manage to instantiate subjectivity, and whether such
conditions could be recreated in a computer.  This I don't yet know.
But I don't think that distinguishing between the formal, syntactic, or
deterministic operations of a machine (or human) and the fact that there
is _also_ meaning, intentionality, or semantics, forces us to take a stand
either way on the prospect of intelligent (self-conscious) computers.
I do, however, believe that semantics is not reducible to syntax, or to put
it another way, that meaning is not reducible to mechanism, or again, that
subjectivity is not reducible to objectivity.


-------------------------------------------------------------------------------
 John Emmer               "...reason has no dictatorial authority; its verdict
 Philosophy Department     is always simply the agreement of free citizens..."
 Penn. State University                   - Immanuel Kant, CPR, A738/B765
 JPE1@psuvm.psu.edu       Now playing:  Amiga: BC  Genesis: KC  SNES: Smash TV


