LINGUIST List 2.60

Tuesday, 5 Mar 1991

Disc: Cognitive Linguistics

Editor for this issue: <>

Directory

1. "Michael Kac", Re: Cognitive Linguistics
2. , posting for the mailing list

To Hestvik: The wording of your comment suggests that the question of
whether language is autonomous, or part of an autonomous module, is the
same as the question of whether it's innate. Those seem to me to be dif-
ferent issues, at least partly. I suppose it would be hard to claim that
there's an autonomous language faculty that isn't innate, although even
that doesn't seem like a foregone conclusion; but surely one could believe
in an innate general cognitive capacity. It also seems possible to suppose
that this capacity, however general, might be very different in humans from
what it is in infrahuman species.

The debate that's going on in Linguist doesn't seem, in any event, to in-
volve who is right on the issue of whether language is autonomous. It seems,
rather, to involve a battle over turf: the antagonists appear to be arguing
over who has the right to occupy the hallowed plot of ground called cognitive
linguistics. That's not a very worthwhile thing to have a fight over, it seems
to me, though I sympathize with those who object to reserving the term 'cogni-
tive linguistics' for the research program of only one group of combatants.
I say this with some trepidation, since I can now see coming a metafight over
philosophies regarding what it is and isn't right to have fights about. So
in the end, I'm with Hestvik: everybody back to the lab!

Michael Kac

Arild Hestvik (arildadler.philosophie.uni-stuttgart.de) writes, in
response to the Cognitive Linguistics debate on this mailing list.

>Another futile arm-chair debate about general intelligence vs. innate
>linguistic knowledge is raising its head--let's instead have the real
>arguments and *research* decide whether there is an innate language faculty
>or not. In the end that's the important thing and not whether cognitive
>linguistics is a brand-name reserved for people who a priori have made up
>their mind one way or the other.

I agree, but there is a problem with this reply. Lakoff claims that
research has *already* decided these questions. Indeed, he claims that
there is "massive" evidence for his view. So, from his point of view,
he is out of the armchair already. Now it is important to realize how
this claim gets made.

The "evidence", as I understand it, concerns things like preposition
choice and metaphor (e.g. why "armchair"), not the things you or I have
worked on -- e.g. locality conditions on Norwegian pronouns or the
possibility of expletive 'do' participating in inverted counterfactuals.
The obvious question is how evidence about the modularity of metaphor
can tell us anything about how to explain the antecedents of Norwegian
pronouns. The answer comes only by begging the question, it seems to
me. If you start with the assumption that everything called "language"
is a product of general intelligence, then "massive evidence"
concerning one piece of "language" will automatically bear on every
other piece of "language", since all aspects of language have the same
nature and the same origin. By begging the question, you get the
answer to the question that you want.

What Lakoff and other "Cognitive Linguists" have to provide is not
research on other "also interesting" topics, but "Cognitive Linguistic"
research specifically on the same topics that we look at, if they want
to reach any legitimate conclusions about the topics we work on. This
is why I think that the proposed journal, whatever its title, is a good
idea. I look forward to "cognitive" reanalyses of your dissertation, my
dissertation, Rizzi's "Relativized Minimality", recent work on
verb-negation-clitic order, the distribution of parasitic gaps -- all as
functions of general, non-language-specific cognitive proceses. If, as
Powers and Lakoff suggest in recent notes, work in Cognitive Linguistics
has been suppressed by the linguistics establishment; and if, as Lakoff
suggests, there really is massive evidence bearing on his claims; and if
this evidence addresses the question rather than begging it, then the
journal should be very interesting indeed.

David Pesetsky


P.S. Notice, by the way, that there is no reciprocity required. I do
not need to prove anything about the relation of metaphor to other
cognitive systems, because I am not making any claims about it. I'm one
of the guys who doesn't assume everything is conected to everything
else. It may be turtles all the way down, but it's not necessarily
turtles all the way across.