======== WDL Dis. #14 (Bcc: all WDL participants) ======
>From stockman@cse.msu.edu  Sat Mar  4 11:04:39 2000
From: "Dr George C. Stockman" <stockman@cse.msu.edu>
To: weng@cse.msu.edu
Subject: Re: WDL Dis. #9

I agree with Fodor, that higher level human thought has the character of
symbolic representation and this must be so. As John Weng points out,
this has advantages and disadvantages for robots.

RE: the issue of learning of language.

Humans changing from one culture to the other learn language in different
ways depending upon age. Older transplants usually never master the "natural"
pronunciation or all of the syntax, whereas young transplants might master
both. Older transplants have an advantage in learning languages, however, they 
have mastered very general protocols, concepts, and world knowledge -- e.g.
one usually pays for services, water runs downhill, etc. One hypothesis here
is that the young learn "bottom-up" and the old learn "top-down".  These 
observations point against a single developmental approach.

I assume that cognition implies computations on internal representations; and,
that the representations used are symbolic in character. Humans are often
conscious of the cognitive mechanisms they use and can even repair methods that 
don't work in practice. The wonder to ponder is "how does this internal 
representation and processing develop in an organism?"

This DOES NOT IMPLY that the symbols used are pre-programmed. Perhaps some are 
and some are created through development (so, how?). Transmission of culture to 
the young does imply programming. 

The most interesting issue to me is: How much automatic symbolic knowledge
representation can/does a single individual do relative to what is "programmed"
by the culture? 

Recently, we have learned that even chimps pass along problem-solving
strategies through their culture and that there are different chimp cultures.
Culture is a library of programs for meat machines capable of reading them.

I conclude with the opinion that a completely developmental approach is not 
optimal and is not followed by nature, human or otherwise. Cultures provide 
animals with high-level programming, which increases both their functionality 
and biases. Higher level organisms do not have enough lifespan to learn
everything they need from development.

George Stockman