A defense of beans and counting them

Carl Mills Carl.Mills at UC.Edu
Tue Mar 30 17:50:45 UTC 1999


Aya Katz wrote (in part):
>>
>> That's the magic of language. The concretes don't matter.

And Tony Wright seconded:

>
>Hear hear!  What you have written is--I think--the first sane and actually
>meaningful thing to appear on FUNKNET in ages.  I think everything you have
>said is implicit in the way linguists actually work.  Some are
>willing to admit that they use of abstraction, and some aren't.  And then there
>are those who want to be neurologists.  Frankly, while I'm sure neurology is
>fascinating, abstract linguistic analysis is no less so, to me anyway.
>
>I can't believe people want to give up linguistics for
>bean counting (neuron-counting).
>
>--Tony Wright
>

As a confirmed counter (of beans and everything else (see the character Cameron in Richard Brautigan's The
Hawkline Monster)) and a determined lover of things concrete, I want to reply to the recent posts by Aya and
Tony.  (And yes, I did read Tony's apology, but I think other points of view need to be represented.)

First, magic is fine, but not all of us are interested in magic.  And for some of us the concretes do matter very
much--at least, in science, as a means of keeping score.

And while linguistics is certainly not neurology, there are other ways of looking at the relationship between
neuroscience and linguistics.  Vic Yngve, Syd Lamb, and George Lakoff have, in different ways, explored some
aspects of this relationship.  Personally, I would like to see linguistics become a science, and that means looking
at the relation between concrete structures and specific functions, on one hand, and abstract theory, on the other.
More, it means that linguistics would finally have to pay a long-overdue promissory note to neuroscience (That's
ok:  there has not been much "there" there in neuroscience to receive payment until quite recently.).  Linguistics
needs to become physically realistic.  This does not have to result in the sort of nothing-but reductionism that has
been labeled "greedy reductionism" (Whom am I quoting here?  Is it Pat Churchland?).  But it does mean that
linguistics has to be "responsible to" a neurocognitive substrate--just as chemistry has to accord with the laws of
physics, biology has to frame its explanatory theories in terms of chemistry, and psychology has to ground its
theories in the brain.

It is not, however, a matter of counting neurons.  As I, an observer hoping to learn something useful, of
neuroscience and neurocognitive approaches to things human, understand it, connections--vertical as well as
horizontal, subcortical as well as cortical--among populations of neurons are what we need to look at (and
neurochemistry, too, for that matter).

Linguists do not have to be neuroscientists, but unless we give our discipline a firm grounding in what folks are
discovering about the brain, it will never become a science.  Instead, linguistics will be condemned to drift forever
between the sciences and the humanities (sic).

However, there is more to grounding than mere neuroscience.  We also need to ground out abstract theories in
evolutionary biology.  In The Mind's Past, Gazzaniga says that the first (and last ) question we must ask is "What
is X for?"  As an outside and sympathetic (I think) observer of functionalists who hopes to learn something from
y'all, I believe that functionalists have been asking Gazzaniga's question for a long time.  I just don't think that
"grammar" or "Grammar" or "communication" are sufficiently concrete, sufficiently fine-grained, sufficiently
biological to build a useful abstract theory on.

I have no quarrel with "grammar(s)" as summaries or parsimonious descriptions of observed regularities in
attested languages (with an 's' at the end).  I just see such "grammars" as data to be explained, not as
explanations.  As for big-G Grammar (or UG, if you will) or "Grammars" in the formal sense, I do not see how
they offer anything but a series of dead ends, an endless asking of how many angels can dance on the head of a
pin.    But I should leave such big issues to Lakoff and Yngve and others who are more capable of dealing with
them.

Reductionistically yours,

Carl



More information about the Funknet mailing list