LINGUIST List 3.649

Tue 25 Aug 1992

Qs: Applied; UG and Neurology; Minimal Commitment

Editor for this issue: <>

Directory

1. "l. valentine", Applied Linguistics Text
2. , Universal Grammar and Neurology
3. , query: minimal commitment

Can anyone recommend a reasonably current text for an
upper division course in applied linguistics? I'm particularly
interested in something having a broad, survey approach,
rather than one that focuses only on L2 learning.

Lisa Valentine
valentinjulian.uwo.ca

Is there anybody out there who works on the neurological side of Universal
Grammar? I am working on the accessability of UG in Second Language
Acquisition, and I would like to know whether anybody has any results on
how parameterization would look like, spoken in neurological terms.

I have an idea that probably does no more than sound good, but has no
significance for the discussion at hand. Here it is: In many animals we
find "learning" processes that involve the implementation of knowledge
(loosely spoken) of a very specified kind, e.g. the recognition of its mother
by a young goose. This kind of "knowledge", once acquired, is unforgettable
and irreversible, and it must be acquired during a sensitive period in
the bird's life. Similar results obtain for the acquisition of
sight. Kittens exposed to polarized light during a certain period never
learn to see properly in the real world, i.e. some structures in the brain
have been trained in an irreversible manner. As far as I have found out
on the pop-sci level I can handle, as a non-biologist and non-psychologist,
this kind of learning corresponds to some kind of window between the
genetically fixed behavior determined by the structure of the midbrain (or
whatever the English term is) and the cognitive processes of the cortex.
In this way one could model the instantiation of UG and parameterization as
a more advanced process of essentially the same kind. This would entail that
first language acquisition, as far as grammar is concerned, is an
irreversible process that happens ONCE and consists in "fleshing out" of
an innate structure with language-specific morpho-lexical material, and a
parameter is a "variable" in an innate program. Thus, UG can be accessed
once, and must be accessed during a period that is genetically fixed
 If the truth is to be found only roughly in this direction, the whole question
 of UG in Second Language Acquisition is pointless, and we have to asked
ourselves whether second language learners can build up anything like a
native speaker's representation of grammatical structures.

Who knows anything about these things? I would be mst grateful for comments,
corrections, references, etc.

Achim Stenzel
Universitaet Hamburg

I'm looking for literature on minimal commitment or assertion set
parsing, both in a computational and in a psycholinguistic context.
If it also relates to coordination, it is extremely welcome.
Please reply directly to me.

Marjan Grootveld