LINGUIST List 11.110

Fri Jan 21 2000

Disc: Newmeyer: Language Form & Language Function

Editor for this issue: Anthony Rodrigues Aristar <aristarlinguistlist.org>

Directory

1. Andrew Carnie, Reply to Hilferty

[Moderator's note: The following posting is authored by a moderator of the 
LINGUIST list. However, this posting reflects the personal views
of the author, and does not in anyway reflect the position (editorial or
otherwise) of the other moderators or the LINGUIST list]. 

Colleagues,

I would like to thank Joe Hilferty (henceforth H) for his interesting 
response (LINGUIST 11.105) to my review of Newmeyer's _Language form and 
Language Function (LINGUIST 11.57). I think it is very important to hear 
the "other" side of on issues as divisive as the functionalist/formalist 
split, and I hope his message will generate a lot of dicussion. 

I would, however, like to respond to some of H's criticisms -- both of
the review and of the ideas underlying Newmeyer's work. 

First to the matter of the critical content of the review. H complains:

> I was surprised to see that the
> review does not even wince at, much less consider, statements like
> the following
>
> There is little reason to believe that the conveying of
> information is a central function of language to begin
> with. (p. 133)

I'm not entirely sure that such this is a fair criticism of a review
(whether it is a fair criticism of Newmeyer is another matter). I stated
up front in my review that I had a bias in favor of Chomskyan grammar,
and further that I wasn't in any position to evaluate accuracy of Newmeyer's
representation of functionalist assumptions and claims. The goal of my
review was to introduce the much more fundamental idea underlying
Newmeyer's book: That functionalist and formalists *do* have something
to say to one another. 

Turning now to the more contentful issues, H observes:

>> Synopsis.
>[snip!]
>> He [= N] also points out the bizarre
>> attitude among some functionalists to view certain varieties of formalist
>> grammar (such as HPSG or Categorial Grammar) as being more congenial to
>> functionalism than GB/minimalist approaches (presumably because such
>> views share a non-transformational/unificational approach to language.)
>
>How could this attitude be bizarre? If functionalists aren't attracted
>to the proverbial "six-foot trees" of Chomskyan syntax, why can't they
>feel more at home with surface-based theories?

H quotes me out of context here, and leaves out the crucial part of this
paragraph. I was pointing out that in many respects Chomskyan syntax
(in particular minimalism) is *more* functionalist (uses external 
explanation, invokes a strong semantics-pragmatics-syntax interface, etc)
than the much more formal unification-based theories. The size of
Chomsky's trees (which, by the way, are no longer trees, and no longer 6
foot long), is quite irrelevant to the more basic underlying assumptions,
which are shared by functionalists and minimalists, but are excluded 
by the more formal CatG and HPSG. 

I'll skip several of H's comments here, leaving them for others to 
discuss, and turn to the issue of the so-called psychological 
reality of the various theories. Here is my quote, followed by
H's response.


>> [snip] N soundly trounces Lakoff (1991) for simply
>> misrepresenting the "cognitive commitment" of Chomskyan linguists.
>> Lakoff claims that generativists reject a view of linguistic theory situated
>> within cognitive neuropsychology
>
>Let's go back to the mid-1960s. Did generative grammar stop
>recognizing the legitimacy of transformations when they were shown
>to be psychologically implausible? No.

This is a very important issue, and reflects what I believe (I make
no claims about what Nemeyer believes) to be a problem with the 
scientific behavior of many linguists (both generative and otherwise).
The psychologcial experiments of the 1960s, may have suggested that 
the transformations *of the 1960s* had little psychological 
reality. (I wish H had cited some of these experiments, so that 
I could go and look at them and evaluate more accurately whether 
this is what they actually did). There are two responses to this
kind of result. The decendants of Generative Semantics (many of
whom went on to become functionalist practicioners of Cognitive Grammar)
took these results as evidence that the entire transformational enterprise
was misguided and abandoned it wholesale. As far as I am concerned,
(again, this is my *personal* opinion), this is simply bad science. 
When one recieves a result that goes against your hypothesis, you don't
abandon your hypothesis -- you REVISE it. This is what happened
in the 70s within "mainstream" Chomskyan circles. Transformations
today look *NOTHING* like transformations of the 60s (in fact, it
is probably inaccurate to call the operation "move" a transformation
at all -- it is more accurately a subpart of the structure building 
operation merge -- and does very little structure changing) -- I might
add that transformations of today actually bear little remblance to
transformations of the 80s either. Turning back to the issue
of the psychological reality of such operation, extensive and extremely
sophisticated work on parsing, priming experiments, language acquisition
has been performed since the 1960s (by such people as Bever, Crain, 
Pinker, Nicol, McKee, etc) that show that there is evidence
supporting the more modern notion of transformation. My (& Newmeyer's)
original statement holds: Lakoff's claim that Generative grammar 
ignores results from psychology is *simply* and *utterly* false. 

H continues:

> Now, let's come forward to
> the year 2000. Is the Chomskyan search for "perfection" (non-
> redundancy in syntax) motivated by any findings in cognitive
> psychology?

Let me respond to this with a question. Do all results in linguistics have
to be motivated by cognitive psychology? Of course not. 

H concludes his discussion with the concern that N's book is an inappropriate
"first contact" work for formalists approach functionalism.

> Finally, I should say that I don't disagree with everything that
> N says in _Language Form and Language Function_. Many of his
> criticisms of functionalism are no doubt well-founded. I do find
> it unfortunate, however, that the first contact that many linguists
> will have with functionalism will be with this book. I really
> don't see how it could stimulate many "formalists" to look any
> further at the movement.

I have to disagree greatly here, the whole point of my review what to
point out that N's work is a resounding endorsement of the idea that
Formalists should abandon their traditional dismissive attitude towards
functionalist results, and that functionalists should look more carefully
at the claims and assumptions of formalists, to see that they aren't so
distant after all. It is an unfortunately artifact of the long-standing
- almost tribal-- division among linguists, that it takes a self-confessed
insider like Newmeyer to get formalists to even look at functionalist work.
Similarly, I suspect it will take a book by a functionalist insider
to get functionalists to stop dismissing Chomskyan work as completely 
without merit. 

After posting my review, I received a number of private responses from
people observing that Newmeyer's book comes now at a very appropriate
time. While the historical origins of the functionalist/formalist split
may go much deeper than the 1960s, it is clear that that is when
the split got its nastiest and most polarized. There are many younger
linguists who intellectually "grew-up" long after the issues that 
initially divided the camps have become non-issues. I strongly believe
that many people are now, after 30-40 years of strife, are now willing
to talk to one another, and Newmeyer's book -- at least for the formalist
camp-- is a great starting point.

Best,

Andrew Carnie
Asst. Prof.
Department of Linguistics,
University of Arizona.
Tucson AZ 85721
carnieu.arizona.edu