I will let Johanna explain what perspective she speaks from.
However, many of the claims in the following are just plain wrong.
>>> [log in to unmask] 09/03/06 7:37 AM >>>
Johanna speaks from the transformational/generative perspective which
*assumes* that *native* speakers of a language know the *grammar* of
that language. Unfortunately, this theory has never been supported with
evidence. There is evidence, though, that children never exposed to
language will never speak a human language. Language, as cognitive
linguistics affirms is learned, and not innate.
***
For the teaching of grammar to native speakers, this debate has
important consequences. If Eduard is right, then a lot of grammar
categories have to be explicitly taught because students don't know
them. If language is innate, then a lot of grammar instruction can be
dispensed with (does any native speaker get taught how the article
system in English works?), and for those categories that are important
for students to know explicitly, we can define those categories in such
a way as to show students how their innate knowledge can identify them.
Much of Eduard's claims that language cannot be innate go to the poverty
of the stimulous argument that is made by innatists. Essentially, all
native speakers know things about their language that cannot possibly be
in the input. There is evidence for the poverty of the stimulous
argument.
I) We all know that children produce utterances that they have never
heard. If language is not innate and the language we use crucially
depends on the input, we left trying to explain how that is possible.
Moreover, even those of us who have full native-speaker competence can
make judgements about possible sentences that we have never been exposed
to. An example I like to use is relativizing the genetive of the object
of a comparative.
1) There is the woman whose daughter my daugher is prettier than.
That is a perfect good sentence in English; it is not possible in most
languages of the world; and I am confident most of us have never heard
or read it. Eduard's claim is that linguistic categories are not
"innate." It will be interesting to read an explanation about how we
"learned" that (1) is possible without any reference to abstract
linguistic categories.
II) It should be observed that dogs are talked to all the time and
never learn the language that is address to them. Obviously, children
need some kind of input, but the serious issue is what does the nature
of that input have to be. Contrary to what Eduard claims, there is
evidence that that input can be very, very minimal.
First, there is the work on home sign by Susan Goldin-Meadow and
colleagues. Home sign is the gestural system of deaf children born in
hearing households in which no one has learned or will learn America
Sign Language. I recommend the following article by Goldin-Meadow and
Mylander in Language.
Goldin-Meadow, S and Mylander. (1990). Beyond the input given: The
child's role in the acquisition of language. Language, 66, 323-355.
Goldin-Meadow and Mylander discuss one child who developed agreement
morphology (hand shape was different depending on the object being
manipulated) that the child regularly used. NO ONE in his family
regularly used hand shape in this way. In other words, this child
created a morphological system without any input. This is predicted by
the innatist claim. I have no idea the story that someone who believes
language is not innate would tell.
Second, there is the work of Jenny Singleton with Simon. You can find
this research cited in Chapter Two of Pinker's The Language Instinct.
(An aside: Sampson makes no mention of this work in his book refuting
Pinker.)
Simon was born deaf to deaf parents. Simon's only ASL (American sign
language) was from his parents who had learned ASL late and were not
native-like in their signing. Singleton found that Simon's ASL did not
look like his parents' ASL but like kids born to native ASL signers. In
other words, Simon went beyond the language he had been exposed to and
regularized it. These results are predicted by the innatist hypothesis
and not by a claim that language must be "learned."
III) I found nothing particular insightful in Sampson's book refuting
the innatist hypothesis. I will consider an example that Sampon spends
a lot of spacing on: yes - no questions in which there is a tensed
clause in the subject.
Sampson is correct to observe Chomsky has cited such sentences, and
beginning linguistics texts, use such sentences as evidence that we have
knowledge about language that is not in the input. Here is the example
used by Sampson. Notice that in the following sentence only "will" can
be moved to make a good yes-no question and "are" can't.
2) Will those who are coming raise their hands?
3) *Are those who coming will raise their hands?
For innatists, the kind of knowledge we have to know (2) is possible and
(3) is not cannot come from the input. Try to explain why (2) is
possible and not (3) without any grammatical categories. Remember if
language is not innate, then children don't have those categories. Note
that (4) is possible.
(4) Are those [children] coming?
It is instructive to observe how Sampson refutes the poverty of the
stimulous claim about such yes-now questions. On page 82, he provides a
number of examples from a written corpus of 90 million words. He says
he found many examples, but doesn't tell us how many. Children are not
really exposed to those written sources, so this really is not any
refutation.
Sampson reports he could not find any examples of such questions in a
spontenous speech corpus. It is very revealing to read his speculation
on why he was unsuccessful:
Certainly the possibility exists, as always, that my failure to find the
[relevant examples in the spoken language corpus] was because of some
shortcoming in the search patterns I used used and there really are
examples which my automatic search missed. (Sampon (2005), The
'language instinct' debate, p. 82)
For those of us how are native speakers, we don't have any shortcomings
in our search patterns for such sentences, and more importantly, we
don't even know what we are searching for to determine whether the
search was successful or not. Of course, this is not problematic if
language is innate, but Sampon's own speculation here shows how
difficult it is to connect just input to what we know is possible in a
language.
Much of grammar is innate. We can use this fact to help us teach
grammar to native speakers. This is not a new observation. Both
DeBeaugrande and Noguchi have made this proposal and shown how native
speaker intuitions can help students understand grammar.
Bob Yates, Central Missouri State University
To join or leave this LISTSERV list, please visit the list's web interface at:
http://listserv.muohio.edu/archives/ateg.html
and select "Join or leave the list"
Visit ATEG's web site at http://ateg.org/
|
