Craig,

You present a view of language to support a perspective on language teaching. I question your view of language and therefore reject some of your teaching suggestions.

If what you report below is "convincing," then share with us some of the convincing explanations.

>>> Craig Hancock <[log in to unmask]> 02/06/09 12:17 PM >>>

Cognitive linguistics  has taken on that question very seriously, and I find their explanations 
very convincing.

****
I presented a number of examples from my experience teaching both native and non-native speakers that don't confirm some of your claims about language.  

Here is a simple piece of evidence about knowledge of English grammar that is clearly learned/acquired by ALL native speakers without any instruction or mentoring or whatever term you want to use.  English makes a distinction between count and non-count nouns.  

I have NEVER seen a dictionary for native speakers that identifies whether the noun is count or non-count.  Think about what that means for the following claim:

If, in fact, language is learned and not just "activated", and if 
interaction is a huge key to that, and if children differ radically in 
the kinds of language they bring to school, then there are huge reasons 
for rethinking our current practices.

Why don't we have to teach NATIVE speakers of English the count-non-count distinction?  After all, children "differ radically in the kinds of language they bring to school," yet NO ONE seems to have noticed a variation in groups about what nouns are count and what nouns are non-count.  Perhaps, children with respect to this feature of English regardless of background get this distinction for free.

Now, look at a dictionary designed for non-native speakers.  ALL the nouns are identified as being count or non-count.  If learning a first language is just like any other kind of learning, why do non-native speakers have to be explicitly taught the count-noncount distinction and native speakers don't?  Certainly, it cannot be the case that ALL non-native learners of English are stupider than native speakers.

***
There is a problem with the following claim.

   I use the term "mentoring" very carefully, as opposed to 
"instruction", which probably brings in workbook images and the like. It 
may make more sense to say that language is "learned" than it is that it 
is taught, but it is always learned interactively. It involves intention 
reading and pattern finding, which are normal cognitive processes. I 
like Tomasello's notion of a "joint attentional frame" to describe 
moments in which two people are attending to the same phenomena, perhaps 
finding ways to embody it in words. Those of us who raise our children 
well do that all the time.

If language is learned interactively, then we have two problems.  

Problem I: How do we know what isn't possible in the language?

Both sentences 1 and 2 are good echo questions:

1) Bob likes ice cream and what?
2) Bob likes ice cream with what?

However, only (4) is possible and not (3).

3)  *What does Bob like ice cream and?
4)  What does Bob like ice cream with?

What interactions have we all had that explain these judgments that ALL native speakers have?

Problem 2: How do we know a sentence is possible when we have never been exposed to it?

The only time I have EVER encountered the following relative clause is as an example in a linguistics class.

5) There is the woman whose daughter my daughter is prettier than?

I maintain we have never had ONE interaction that has a relative clause like (5).  Yet, every native speaker knows it is possible sentence in English.

Finally, let's not confuse correlations of various structures in a particular kind of genre and grammatical judgments. Craig writes:

I see a strong interconnection between grammar and genre. Even the corpus
grammars are finding empirical correlations between the kind of text and
the kind of grammar that is likely to occur. It is possible to see those
as functional patterns.

First it is not surprising that certain grammatical constructions tend to occur in certain kinds of texts.  As someone who has taught ESL, I use such facts to determine what kind of tasks I want my students to do to practice a particular grammar structure.

However, "empirical correlations" are very different from grammatical judgments.  Go back to my sentences 1-5.  ALL native speakers agree on the grammaticality of those sentences.  Exactly how far from an "empirical correlation" must a text be before it is not an example of a particular genre.   For example, let's assume that in genre X, there are 5 passive voice constructions per 1000 words.  If I write a text that has only 3 passive constructions/1000 words, is it no longer part of that genre?  10 constructions/1000?    
 
Bob Yates, University of Central Missouri

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/