LINGUIST List 2.70

Wednesday, 10 Mar 1991

Disc: Quechua, Cognitive, Tromso

Editor for this issue: <>

Directory

1. , Re: Quechua
2. "Michael Kac", Quechua
3. Bill Poser, cognitive linguistics
4. Marit Westergaard, Contacts in Tromsoe
5. John Goldsmith, Re: Tromso
6. , Re: Tromso

Regarding the query on Quechua, I recall that there are several books
of introduction to the language. I do not have the books with me here
so I cannot check the titles:but I recall

Clodoaldo Soto-Ruiz, *Gramatica Quechuana* (Published, I think, in Lima)
Soto-Ruiz, *Quechua* (Lima(?), this is an introduction w/ lessons etc)
Conrad Phelps, The Grammar of Quechua.

Sorry to be so vague about the titles, etc but I am a long way from home
where I have the books stored and am a specialist in Old Testament studies,
not South American languages. It might be a good idea to contact Cornell
University for further help, as they have considerable background in 
Quechua instruction.

Mike Cheney
Teologiska Inst
Lunds Universitet

One possible source of information is Carol Klee, Dept. of Spanish and
Portuguese, U. Minnesota, Minneapolis, MN 55455. (No e-mail address that
I can find.)

I would like to add my objection to the usurpation of the term
"cognitive linguistics" for work done by advocates of a
particular set of views in a particular subfield of the domain
that this term more properly covers, and to address several
issues that have arisen in the discussion thus far.

It is true that linguists are not, in general, particularly well
versed in psychology and up on the literature on modularity.
However, it is not clear to me that those involved in "cognitive
linguistics" are, in general, any more familiar with this area.
Indeed, I am struck by the fact that what from the literature is
an active area of controversy is regarded by them as settled. I
recommend that anyone who regards the issue as settled consult
the anthology:

 Modularity in Knowledge Representation and Natural Language
 Understanding, edited by Jay L. Garfield (1987).

What linguists do have to contribute is their knowledge of what
language is like and what phenomena there are to be accounted
for. Not surprisingly, linguists know a great deal more about
this than non-linguists. Perhaps more surprisingly, and
certainly unfortunately, even those non-linguists who address
linguistic questions are often rather naive about language
and unacquainted with theoretical work. Two examples:

 (a) a well-known figure in Artificial Intelligence has been
 a proponent of the view that natural language understanding
 does not require parsing, that is, that it does not
 require syntactic analysis or any knowledge of syntax, but
 can all be done from semantics. In its pure form, this theory
 is incapable, as a matter of principle, of accounting for the
 fact that the English sentences (1) and (2) are not synonymous:

 (1) Mary saw John.

 (2) John saw Mary.

 (For discussion I recommend Mitch Marcus' paper in the book
 _Talking Minds_.)


 (b) a well known figure in work on neural networks gave a talk that
 was touted as describing a neural network that "learned syntax".
 As it turned out, it was able to learn the difference between
 transitive and intransitive verbs. That is not quite all there
 is to syntax.

To most linguists such claims are so far out as to look like the work
of cranks, but their proponents are not regarded as cranks by
people in AI and psychology. The gulf in knowledge of language between
linguists and non-linguists is huge.

Now, this is not by any means to say that all non-linguists are
ignorant, but it is to say that that it is very important, in
any psychological investigation of a behaviour, to have a good
idea of what that behaviour is like, and that linguists play a
crucial role in investigations of linguistic behaviour by
providing most of the facts and generalizations about the
structure of language that need to be accounted for.

 Secondly, I would like to second David Pesetsky's point
that anti-modular views have been rendered plausible to a large
extent by ignoring the sorts of linguistic behaviour that need
to be accounted for. It won't do to look only at certain areas
of semantics and pragmatics - how about syntax, morphology, and
phonology? The same thing is true of functionalist attempts at
explanation. For most of the generalizations posited by
linguists, especially formal ones (e.g. locality principles), I
have seen no attempt at functional explanation. (Note that I do
not mean to identify anti-modularism and functionalism: while a
functionalist explanation can be faculty specific only
vacuously, it is logically possible that there could be general
principles of cognition and/or generalized cognitive abilities
that lack any functional explanation.)

 Third, let me point out a subtlety that seems often to
be missed. There are two different claims involved in the
modularity vs. generality debate. One claim has to do with
informational encapsulation, to use Jerry Fodor's term. This is
a claim about modularity of mental processing. The other claim
has to do with the faculty-specificity of explanatory
principles, that is, with whether there are specifically
linguistic principles or whether they are all consequences of
more general cognitive principles. These two claims are not the
same. It seems to me that a lot of what I have read by people
like George Lakoff bears on the faculty-specificity of
principles but, at least not directly, on processing modularity.
(If this is not true, no doubt George will object.)

 Finally, I'd like to respond to Margaret Fleck's
suggestion that the input of people in computer vision and
robotics is crucial. While the input of anyone with relevant
knowledge is welcome, I don't see the special relevance of these
two areas, for two reasons. First, they don't have much to do
with language, and if it is the relationship between language
and cognition in general that is at issue, it would seem that
they aren't particularly relevant, unless perhaps she just means
that they are among the many people who may have something to
contribute to our general knowledge of cognition. Second,
by the very definition of these fields, people in them
are not concerned with understanding how human cognition works,
but rather with making machines do somewhat similar things.
It doesn't matter to them whether the machines emulate human
abilities (if they could do better than humans, they would
be delighted), and they don't care whether they perform
these tasks by the same mechanisms that humans do. This is true
in general of much of AI work - what one does to solve an engineering
problem MAY provide an idea as to how humans might do it and may
provide some tools for studying humans, but the tasks of engineering
and of understanding how human minds work are by no means the same.


 Bill Poser

To Ed Battistella

RE: Contacts in Tromsoe, Norway

Your colleague can contact the following linguists in the Linguistics
Department at the University of Tromsoe:

 Tarald Taraldsen knuttmack.uit.no
 Ove Lorentz lorentzmack.uit.no
 Toril Fiva fivamack.uit.no

However, if your colleague is our Fulbright professor for next year,
he will be working in the English Department, and the only linguist
there with an e-mail address is me. I am on sabbatical at UC Santa
Cruz until July -91 and can be contacted at the following address:

 maritling.ucsc.edu

Marit R. Westergaard

Jan-Terje Faarlund is at Trondheim, and his email address is 
jan.faarlundavh.unit.no
John Goldsmith

To: Ed Battistella, concerning linguists in Norway
I know of one linguist in Norway - Thorstein Fretheim in the
Linguistics Dept. at Univ. of Trondheim. A colleague of mine here
just told me about someone else - Per Moehn, though he's
not sure where exactly he is; maybe also in Trondheim.

Jeanette Gundel, U of Minnesota