Re: "Learning" Language (Long and fierce)

[Mary Howe <howe -at- KUHUB -dot- CC -dot- UKANS -dot- EDU>]

Okey-doke, folks, this thread is driving me crazy.  I think it's way off
topic so I'm not going to go into big detail about this. Amy is right
when she describes the theory as claiming (and I think quite correctly)
that it's not a specific language (or even language family or type) that
humans are preprogrammed to learn, but rather the underlying structures
of what has been termed "universal grammar" (UG).  If you look at
languages across the world, of widely differing types, you see a) that
there are universals and universal tendencies in languages, and  b) that
children learn _their own native_ languages at much the same rate, at
much the same developmental stage.  Of course there are vast differences
between the languages themselves (although fewer than the non-linguist
might think), but the principles of acquisition do not differ much.

But, Amy is incorrect regarding the family in England where it is claimed
that a gene for language was isolated.  This is Myrna Gopnick's research
for which there was lots of early publicity and she has since admitted
that an actual gene has not been isolated.  Unfortunately, the initial
announcement made a big splash in the press, but the subsequent
qualifications did not.  (So, Amy, I don't blame you for only seeing the
first bit.) However, work in this area is ongoing and very promising.
Much of the research involves children with specific language impairment,
which is a disorder that very often runs in families.

OK, I'm almost off my soapbox.  If you want any further discussion with
me about this, let's do it off the list. (Please note my signature line
to see why I know about this.)

My last comment, bringing this almost back on topic, or at least back to
the topic that started all this wild and almost entirely anecdotal
speculation (although I'm definitely not including in that remark those of
you who cited Chomsky), is that it is a mistake to equate the grammar
learned in school (grammar school, English class, etc.) with the grammar
acquired by young children.  The school grammar is a set of prescriptive
rules imposed by society (the academy, the "experts", etc.) aimed at
maintaining a standard language.  The standard language can have little or
nothing to do with spoken language, and is generally based on social
judgements rather than linguistic judgements.  As technical writers we are
generally expected to uphold the standard, and that's fine, because that's
what is in generally accepted use and generally understandable to all.
However, nonstandard varieties, while not acceptable for written
documentation, are every bit as valid as language varieties as the
contents of _The Elements of Style_.

OK, I'm done.  Sorry for the heat.  But if you flame me for this, be
prepared for more (but privately. please).


Mary Howe
Assistant Research Professor
Child Language Program                  Phone (913) 864-4789
University of Kansas                    email howe -at- kuhub -dot- cc -dot- ukans -dot- edu
1082 Dole Center
Lawrence, KS  66045            email after Dec. 15: thunder -at- idir -dot- net


On Thu, 7 Dec 1995, Amy Welden wrote:

> Karen Mayer writes:

> >  The current research by *some individuals* does not necessarily mean the
> >  theory is fact. Have they isolated a gene for this "hardwiring?" If so,
> >  can they distinguish which language the gene is hardcoded for? Do Chinese
> >  people have a gene for a Chinese "dialect", whereas French people have a
> >  gene for learning French? Since the grammar of Chinese is so different
> >  from French, one would suspect, based on the hardwiring theory, that the
> >  genes would be easily distinguishable...
> >

> The more I read about Chomsky's Deep Structure, the less I understand about
 it.
> But, this much I do understand: It's not a particular language's grammar
 that's
> hardwired, it's the underlying categories that allow us to learn the
 particular
> grammar that we're exposed to.  The deep structures are more abstract than
> specific rules of grammar.

> There's some compelling evidence that all children, regardless of the
> particular
> language they speak, go through the same phases while they are acquiring
> language.
> They go through a one-word stage, then a two-word stage with the same
> construction
> of object and then the verb. They then start putting endings on the verbs
> in the same
> order: progressive endings (ing), tenses, etc...

> As far as isolating a gene for this, they have isolated a gene in a family
> in England
> whose members can't learn grammar.  They have trouble creating sentences that
> make sense and can't put the proper endings on verbs.  It's a very specific
> defect
> that they can trace through the family.  It goes beyond just exposure to
> improper
> grammar.

> Amy Welden
> awelden -at- mind -dot- net