Newsgroups: comp.ai.philosophy
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!europa.eng.gtefsd.com!howland.reston.ans.net!pipex!uunet!EU.net!CERN.ch!dxal18.cern.ch!hallam
From: hallam@dxal18.cern.ch (Phillip M. Hallam-Baker)
Subject: Re: What's innate? (Was Re: Artificial Neural Networks and Cognition
X-Nntp-Posting-Host: dxal18.cern.ch
Message-ID: <D2EoD8.Fyp@news.cern.ch>
Sender: news@news.cern.ch (USENET News System)
Reply-To: hallam@dxal18.cern.ch
Organization: Wot!!! Me ????
References: <3erqrv$va@mp.cs.niu.edu> <3esaig$6h5@mp.cs.niu.edu> <3f6ep0$r5f@oahu.cs.ucla.edu> <3f6il0$io6@mp.cs.niu.edu> <D2D2DK.6CL@spss.com>
Date: Sat, 14 Jan 1995 17:24:44 GMT
Lines: 117


In article <D2D2DK.6CL@spss.com>, markrose@spss.com (Mark Rosenfelder) writes:

|>In article <3f6il0$io6@mp.cs.niu.edu>, Neil Rickert <rickert@cs.niu.edu> wrote:
|>>In <3f6ep0$r5f@oahu.cs.ucla.edu> colby@oahu.cs.ucla.edu (Kenneth Colby) writes:
|>>>   I have never been able to buy the "poverty of stimulus" argument.
|>>>   Children are immersed in language from day one ( and even before -
|>>>   prenatals recognize their mother's voice) with parents, siblings,
|>>>   peers, TV, etc. In their dreams they hear people talking, they
|>>>   talk in their sleep, they practice language aloud or silently
|>>>   before falling asleep. By age four, they have had at least 20,000
|>>>   hours of experience with language - hardly "poverty".
|>
|>Children are driven about in cars from day one (and even before; pregnant
|>women drive too).  Young children are fascinated with cars, which are a
|>frequent subject of their conversation and play.  By age four, they have
|>had thousands of hours in cars or thinking of cars.  Naturally they are
|>themselves able to drive by this time.

This is not a usefull analogy. The child is merely a passive object within the
car and has no interaction with it. Indeed the child is almost certainly
shackled to the car seat. 

When learning language the child is the focus of attention and the focus of 
concentration of both parent and child is the aquisition of language. 

|>No one has a really satisying theory of language acquisition.  Your account
|>seems true in outline.  English-speaking children, for instance, go through 
|>a period when they imitate irregular past tense forms, then *drop* these
|>and apply -ed to everything, then start to pick up irregular forms again.
|>It's not hard to see in this three successive stages in understanding how the
|>language works.  What's tricky is figuring out when children decide 
|>to replace their grammars, and how.  They can be astonishingly unaware of
|>striking differences between their speech and adult speech; many transcripts
|>exist of adults trying to correct a child's speech, and the child completely
|>ignoring them-- another reason to distrust the "20,000 hours of experience"
|>handwaving, as well as simplistic stimulus-response theories. 

Children can  be strikingly unaware of many things. Telling them not to do 
something can be as good as telling them to do it. Attempts by a parent to 
`correct' a child's speech failing would appear to count against Chomskys 
argument. If it were merely a case of deciding which grammar to apply then the
child could simply switch from one to another. Attempts by parents to change 
their childs speech are notoriously unsuccessful, as any parent knows children 
learn language from their peers as well as from their elders.


It is important to remember that a child is not simply learning a language but 
an entire world view. Before the child can understand future tenses the concept
of time must be understood. Failure to learn grammar of future tenses in the 
young child does not necessarily indicate that the `grammar' engine is faulty
or derailed (oops someone set the selector switch to German). For the child to
be able to use the future tense the child must have an understanding of the 
concept of future and the use of the future tense, furthermore these concepts
must be integrated within the belief system.



As too Chomskys famous `reverse negation language' let us consider how it
might work in practice. We start with the question "Did you take beer to the
party?" To which I reply "I took beer". OK?

Well not quite because I come from the north of my country where there is a
dielect form of English which these days tends to sound Shakesperian.
In this dielect I state "Beer took I". 

Consider the situation in Elisabethan England. There were several hundred 
distinct dielects in use. A dielect in which negation was implied by order would
be incomprehensible. If there was a predisposition to a particular universal 
grammar then negation in this style would be more likely, not less. Hearing 
phrases in a form that was contrary to our inate understanding of language 
should create resonances at the poetic level at least. There would be a 
universal reference point to be differentiated against. 


I think this Universal language position is unnecessarily ideological and 
dogmatic. Our understanding of the brain is certainly in now way advanced 
enough to consider this approach as definitive. One crucial factor that has
been overlooked is that the grammars we use today are not pure, they have been
tampered with by many centuries of standardisation. If you want to discuss 
universal grammar then you can't use modern english which has been critically 
defined by the opinions of ninteenth century schoolmasters and standardised 
by the BBC. We do not use organic english, we use a synthetic copy that has
been rebuilt on the basis of renaissance understanding of Latin grammar. That
European grammars have similar features is hardly a suprise since they
are the product of a common ontology*. Worse still the classifiers have squashed
almost every other language into their own forms. 

Even if similarities between language grammars are found I would ascribe 
it more to the activities of those attempting to enforce social conformance 
and the definition of social rank than any `inate' capability. I find it 
suprising that Chomsky squares his linguistic ideas with his political thought.
His political thought seems to reject platonism and its implied hierarchical
social order, yet his linguistic theory would seem to imply a universal in
language which seems quite unnecessary.




[* Is that the right word or should I use epistomology, Roy Bhaskar wrote a 
400 page book on dilectics insisting on the imporance of the distinction 
between ontology and epistomology without mentioning what he considered 
the difference to be. Such actions are hardly excusable in a book on 
communication. Either a concept is simple in which case the difference is
obvious or it is complex in which case differences of opinion can arise
even between experts. If one accepts the Hermaneutic axiom then one should
also explain ones terms. I am overly suspicious of the use of such words since 
they are often used more to exclude the lay person from a discussion area where 
their thought may be as usefull as the `professional'. Complex language always
has the hint of being a membership sign for a trade guild.]

[Comming soon, HTML footnotes and news in HTML]

--
Phillip M. Hallam-Baker

Not Speaking for anyone else.
