Newsgroups: comp.ai.philosophy
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news3.near.net!noc.near.net!paperboy.wellfleet.com!news-feed-1.peachnet.edu!gatech!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: <D3nB2v.H6q@hpl.hp.com>
Date: Tue, 7 Feb 1995 19:49:43 GMT
References: <3gtu3i$rf3@mp.cs.niu.edu> <D3GBKw.F5D@spss.com> <3guoku$bci@mp.cs.niu.edu> <D3LG9D.G18@hpl.hp.com> <3h69pv$en3@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: 71

Neil Rickert (rickert@cs.niu.edu) wrote:
: I have seen no argument which is persuasive that we have a UG.  There
: seems to be good evidence to the contrary.  A UG should serve to
: stabilize the syntax of a language.  Yet we see that regional
: dialects can form quite readily, and these can include syntactic
: variations.

All known stable languages (i.e. excluding Pidgins) have many
commonalities of sytactical relations. These constitute,
or arise from, or are constrained by the UG. Obviously,
there is considerable variation within the limits fixed
by the UG, and observations of such variation are not
evidence against a UG. Chomsky does not claim that
we all carry an English Grammar book in our heads - that's
just a straw man.

: >Consider vision. We know that considerable preprocessing of
: >retinal images is hardwired into the brain.

: No, we do not know that.  Many people presume that this is so, but we
: are far short of the understanding of visual processing which would
: be required to say that we know this.  Moreover, the visual system
: evolved over a far longer time period than did our language
: capability, so that even if your claim is correct it would give no
: persuasive support to the claim that an innate UG could have
: evolved.

Well, I think we know that, and many details besides. Yes, the
system is more primitive than the language systems, and more
generally useful, and differs in many other ways. See below.

: >                                            Many features of
: >images are abstracted out of the raw data, before we are
: >conscious of anything which could be a subject of learning

: Just a minute.  You are presuming that we must be conscious of
: something in order to learn.  There is much evidence to suggest
: otherwise.  Arthur Reber, in his book "Implicit Learning and Tacit
: Knowledge," presents some of the evidence.

I thought I qualified this adequately below. If you want to argue
this in detail (which might be interesting - I actually know
rather more about vision than I do about grammar) let's
take it to another thread. Consider the (rough) analogy for
yourself, ignoring (if you would be so kind) the rather
crude and incomplete expression I have given it here.

: >(I understand that some learning/imprinting takes place at
: >all levels of the visual system, but it is in some sense
: >"pre-conscious", "local", and "non-semantic").

: What do you have against "pre-conscious" or "local" learning?  As far
: as I can tell, that description fits most learning very well.  Rote
: memorization may be an exception, but the quality and retention of
: knowledge acquired by rote memorization is usually quite poor.

I don't have anything *against* it. But I think there is a
useful distinction to be made between "imprinting" and
"semantic learning", related to how localized the learning
is to a single region of the brain. Learning to distinguish visual
contours, or to map tactile stimuli to motor responses, are
more localized to specific brain subsystems than are higher
cognitive functions, including many aspects of language use.
I used the scare quotes in the hope that you wouldn't take
the phrases too literally - I don't know the magic words
for this distinction (if there are any), and developing it
ab initio would take far too long to do here. Think about
this a bit for yourself. It's another argument, anyway -
perhaps the analogy is just an "intuition pump".

Bo
