Newsgroups: comp.ai.philosophy
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!swrinde!sdd.hp.com!hplabs!hplntx!curry
From: curry@hpl.hp.com (Bo Curry)
Subject: Re: What's innate? (Was Re: Artificial Neural Networks and Cognition
Sender: news@hpl.hp.com (HPLabs Usenet Login)
Message-ID: <D310KE.5vp@hpl.hp.com>
Date: Thu, 26 Jan 1995 18:55:26 GMT
References: <3fosrd$2if@mp.cs.niu.edu> <D2q93L.JHA@spss.com> <3fq5ih$hkb@mp.cs.niu.edu> <D2vEL5.9u8@gpu.utcc.utoronto.ca> <3g169u$s8o@mp.cs.niu.edu> <D2xL1y.MzF@hpl.hp.com> <3g42bk$jco@agate.berkeley.edu> <D2zD5B.F00@hpl.hp.com> <3g6j2i$fhi@mp.cs.niu.edu>
Nntp-Posting-Host: saiph.hpl.hp.com
Organization: Hewlett-Packard Laboratories, Palo Alto, CA
X-Newsreader: TIN [version 1.2 PL2]
Lines: 49

Neil Rickert (rickert@cs.niu.edu) wrote:
: Keep in mind that patient instruction on language doesn't seem
: to help either.  I suspect that the approach used in patient
: instruction is wrong for a child of that age.

: My own children took to mathematics like a fish takes to water.
: Perhaps that is because we skipped the instruction, and made it into
: a game.

Really? The level of sophistication with which a 4-year-old deals
with language (counterfactuals, nested intentionals, subordinate
clauses) is comparable in complexity to quite high-level
algebra.

: Kids learn lots of things at a remarkable speed.  There is a tendency
: to notice this more with language because we have a foolish bias that
: only what can be expressed linguistically counts as knowledge.

There are two separate questions here. The first is
whether humans have specialized brain structures to facilitate
language acquisition. It's not clear whether you are actually
denying this, but I think you have a tough row to hoe if you try.

These specialized brain structures are, of course, composed of
interconnected neurons, which presumably function and learn
much as do other neurons. The specialization arises from the
wiring pattern of the interconnections, and its evolution is
not any more astounding than the evolution of many other
specialized brain subsystems (e.g. visual subsystems).

The second question is whether, among these specialized language
structures, there is one which facilitates syntactical
manipulation (i.e. the Universal Grammar). Evidence for
extraordinarily facile language learning in children does
not by itself support the existence of a UG. It is also
necessary to (a) discover universal "arbitrary" constraints on
learned syntax, and to (b) show enhanced learning ability for
these structures in particular (as opposed to, say, vocabulary).
The former is Chomsky's creole argument (e.g. the Nicaraguan
signers discussed by Pinker), and the latter is the
"poverty of the stimulus" argument.

Can we agree to answer the first question in the affirmative?
Then we could focus the discussion a bit on the strengths
and weaknesses of the arguments for the UG.

Bo


