From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!sun-barr!ames!elroy.jpl.nasa.gov!usc!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!emory!gwinnett!depsych!rc Tue Jan 21 09:26:28 EST 1992
Article 2810 of comp.ai.philosophy:
Xref: newshub.ccs.yorku.ca comp.ai.philosophy:2810 sci.philosophy.tech:1881
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!sun-barr!ames!elroy.jpl.nasa.gov!usc!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!emory!gwinnett!depsych!rc
>From: rc@depsych.Gwinnett.COM (Richard Carlson)
Newsgroups: comp.ai.philosophy,sci.philosophy.tech
Subject: Re: Semantics of thoughts
Message-ID: <s9XNeB2w164w@depsych.Gwinnett.COM>
Date: 16 Jan 92 22:13:15 GMT
References: <CHANDRA.92Jan14132526@cannelloni.cis.ohio-state.edu>
Lines: 105

chandra@cannelloni.cis.ohio-state.edu (B Chandrasekaran) writes:

> Richard Yee ((yee@cs.umass.edu) says:
> 
> The difference between formal and semantic processing is profound: In
> the former case, basic symbols cannot *represent* (literally
> re-present) anything... to the processor.  In the latter case, they
> can.  Given a symbol, two semantic processors must agree as to its
> formal properties, but they may differ, to a greater or lesser extent,
> as to its associations or content.  In semantically processing true
> re-presentations (as contrasted with formal tokens), each step holds
> the possibility of interpreting the basic symbols---using them to form
> connections with subjective information.  This can yield inferences not
> derivable solely from the intrinsic properties of the manipulated
> symbols.  The point is that such interpretations and inferences are
> available *within the processor itself*.  A formal symbol processor has
> no such leverage with regard to its basic symbols: all additional
> interpretation and inferencing, i.e., all additional *semantics*, must
> lie elsewhere (e.g., in an external agent's use of "wishful mnemonics"
> :-)
> 
> 				
> BC: The above seems on the right track to me, and it is also related
> to the idea that conceptual symbols are anchored on perceptual
> "symbols" which are in fact processed by a special purpose (i.e.,
> modality-specific) machine, which is symbolic in many interesting
> senses of the term, but not in the sense of a UTM.  Hari Narayanan and
> I have been working on an image-representation and manipulation
> approach along the abobe lines, which resolves the paradox about
> whether reasoning with images is done "propositionally" or in some
> "picture-like" fashion.  The bottom line for the discussion on the
> Chinese Room is that we can think of an alternative to Searle's rule
> book, an alternative in which most of the intermediate symbol
> structures have a natural interpretation as images corresponding to
> intermediate sematic structures that are built as the sentence is
> being understood.  
> 
> Having said this, I should hasten to add that I am not entirely sure
> that this eliminates the need for nonperceptual anchoring for
> some symbols.  That is, I am not sure if it can be argued that *all*
> concepts are anchored only on perceptual and sensory-motor symbols.
> It is possible that "noumenal" perceptions will remain which will
> be outside the sensory and motor domains, such perceptions requiring
> additional "semantics" which we don't know how to give computers (yet).

The notion that semantics is ultimately perceptual is, I think, on
the right track.  Being just a little more general, I think you
might agree that characterizing semantics as _peripheral_ (as
opposed to having a _central_ nervous system locus) might be even
better as you could then add preparatory set for "responses" or
"operants," or the fractional mediating responses which Hullian
psychology represented by r(subscript g), what used to be called
an "attitude" in the 19th century.

(It has always seemed to me that the prototype of language has
been the _grunt_, which is in the imperative and has the
illucutionary force of warning a competitor off of a claimed
territory.  Declaratives came later when the winner was trying to
figure out how to brag to his friends that the loser ran away. The
move of the grunt from unconditioned response to verbal operant
was probably very gradual.)

So semantics is most likely built from sensory percepts of one
kind and another, and from the proprioceptive sensations
associated with various behaviors.  But before you move from a
graphics to a propositional representation, you need one more
consideration -- _contrast_.  The structuralists, from Saussure to
Piaget, have emphasized the importance of the contrastive and
assimilative aspects of a stimulus (to some "test" stimulus.)  The
psychologist, Joseph Rychlak, in an article, "The Missing
Psychological Links of Artificial Intelligence: Predication and
Opposition," in the _International Journal of Personal Construct
Psychology_, argues that predication, in a psychological sense,
requires contrast. _Formally_ it can be expressed in terms of
sets, but _semantically_ predicates seem to require 3 units, two
of which share the predicate and one of which does not, the
contrast.  I would add that for real basic meanings the most
effective contrast is contrariety rather than contradiction, i.e.,
hot vs. cold rather than hot vs. lukewarm.  Greimas builds his
semiotic square on this dialectical-like notion of contrast to the
point of contrariness.

Notice how different this all is from "syntactic" or algorithmic
thought.  A formal, "syntactic" manipulation of a syllogism like,
"All criminals should be segregated from society, minor traffic
violators are criminals, therefore all minor traffic violators
should be segregated from society," can't look inside the word
"criminal" and see the chains of signifieds, each one itself made
up of contrastive images.  The fact that minor traffic violators
are subsumed under the set of criminals is almost irrelevant to a
"semantic" "understanding" of that syllogism.

A capacity for "conceptual" thought at the logical level (rather
than the semantic one) may just be thought that ignores the chains
of signifieds and focuses on only one metaphor, the spatial or
container metaphor of set theory, and adds some notion of
"conceptualness" or "abstraction," which is itself a visual image!
(I always get a visual image of shards of clear, clean jagged
glass for the phrase "clear and distinct idea.")

--
Richard Carlson        |    rc@depsych.gwinnett.COM
Midtown Medical Center |    {rutgers,ogicse,gatech}!emory!gwinnett!depsych!rc
Atlanta, Georgia       |
(404) 881-6877         |


