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!emory!dscatl!gwinnett!depsych!rc Tue Nov 26 12:31:25 EST 1991
Article 1491 of 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!emory!dscatl!gwinnett!depsych!rc
>From: rc@depsych.Gwinnett.COM (Richard Carlson)
Newsgroups: comp.ai.philosophy
Subject: Is dialectical thought an "informal logic"?
Message-ID: <5RPRBB5w164w@depsych.Gwinnett.COM>
Date: 21 Nov 91 15:20:03 GMT
Lines: 81

When I was going to high school in the 50s I had an English
teacher who used the Socratic method in her teaching and who
thought that Socrates was the wisest man who had ever lived.  I
also had a history teacher who believed that the adversarial
method of American jurisprudence was the greatest guarantor of
both individual liberty and societal justice.

Both of these teachers also thought that "dialectical thought" was
evil, totalitarian, authoritarian, Communistic, Fascistic,
imprecise, unscientific, foreign, un-american, fuzzy-headed,
Germanic and generally bad.  If I had told them then that both
Socrates and Perry Mason were "dialecticians" I would have failed
their courses and earned their eternal enmity  as a champion of
all that was dark and a foe of everything that was good and
decent. Of course I never did that because I didn't realize that
both Socrates and Perry Mason were masters of the dialectic and
that that was how they did the miracles they did.  To me at that
time a "dialectician" was a wild-eyed man with a long, scraggly
beard wearing foreign, Russian-looking clothes and spouting angry
nonsense.

Several times I have brought up the dialectical nature of
discursive thought, particularly that thought which is usually
thought of as "interpretive," although I haven't used that term.
I have said that statements in a natural language are heavily
implicated in dialectical thought.

On the surface I can see an obvious reason why persons with an
interest in AI would tend to ignore the dialectic.  It can't be
easily formalized.  If you take an "argument" or a "discourse" as
your unit of analysis, it is not possible to write down formal
rules for making a counterargument or a counterdiscourse which is
either the contradictory, or contrary or, more importantly, the
"antithesis" of that discourse.  So people interested in AI take
the "statement" as the unit of analysis, possibly the very worst
choice since human thought most likely does not proceed by the
movement of discrete statements.  In fact if you took the
dialectic seriously it would seem to suggest that AI isn't even
possible, although Aristotle invented logic, which is really a
kind of primitive version of AI, as an attempt to formalize those
aspects of the dialectic which he thought might be formalizable.

I can give you an "algorithm" for constructing a dialectical
response, but is it a "real" algorithm in the mathematical sense
of being precise and completely spelled out so that it can be
taught to a stupid person or a machine?  Here's the algorithm:
Find the problematic term, X, in the opponent's discourse and
reply: "What do you mean by 'X'?"  There are two "rules" for
identifying X.  It is either 1.) the most "abstract" term in the
sense of having the largest chain of signifieds between it and any
possible extensional meaning, or 2.) a term indexing some recently
problematized "theme."

An examples of 1. would be: "What do you mean by 'justice'?"

An example of 2 would be: "What do you mean by 'we'?" (Given the
multicultural discourse which problematizes which "group" anybody
belongs to.)

A young attorney or politician, even one with at best a very
mediocre "mind," can learn this algorithm and its rules.  High
school sophomores can and did learn it in my day.  But how do you
teach it to a computer?  Oh, it could be simulated, like the
famous Eliza program (which I have on my  hard disk in Prolog), by
tagging the "hot" words.  But you couldn't teach the computer a
rule for finding those terms by itself without teaching it
everything the average person knows, including what sorts of
things David Brinkly and his colleagues discuss on Sunday
mornings.

I suspect that one reason for ignoring the dialectic is precisely
because it cannot be formalized.  So we are blind to it, but much
of our thinking -- yes, yours too, especially in those areas which
have to do with meaning and belief -- is essentially dialectical
in nature.

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


