moving beyond

[Liz Bates]

Like Talmy, I have mixed feelings about Bruce Ricman's "Beyond Chomsky"
initiative, although my concerns are a bit different from Talmy's.  I agree
that this proposed initiative seems to be quite naive regarding the long
list of linguists who have provided critical alternatives, for many years,
to Chomsky's particular variants of generative grammar.  So my response to
each of these mailings has been "Good idea, but hasn't it already been
done?"  But I have to disagree with Talmy about the magnitude of Chomsky's
contribution in bringing down the hated behaviorists.  Talmy seems to be
quite naive regarding the long list of psychologists who have provided
critical alternatives to behaviorism, for many years, before and during and
after Chomsky's own contributions.

The truth is that radical, "black box", Watson/Skinner behaviorism was a
flash in the pan.  Its primary role in history (it now seems) has been to
serve as a scarecrow for generations of generative linguists.  I think it
would be useful for functionalists and generativists alike to understand
this a little bit better.  It is deeply human to want to look inside any
box that is placed before us.  There were physiological psychologists
trying to look inside that box throughout the 20th century, and
experimental psychology has never been without a high proportion of
influential mentalists.  Donald Hebb comes to mind: his 1949 book "The
organization of behavior" was really about the organization of the
mind/brain, and if there is one psychologist in the history of our field
who has been proven SOUNDLY AND FULLY right, it was Donald Hebb (in case
you have ever heard the term "Hebbian learning", it comes from Hebb's
conjecture that "the neurons that fire together wire together", an
unabashedly associationist principle that has been repeatedly confirmed and
elaborated in neuroscience since his time -- Eric Kandel's Nobel Prize two
years ago in many respects represents an acknowledgment of Hebb's victory).
And then of course there was Edward Tolman, the Berkeley psychologist for
whom the Berkeley Psychology building is named.  Tolman believed that the
rat presses the bar because he EXPECTS to be reinforced -- rats have
expectations, hopes, dreams and aspirations, and psychologists have to deal
with those facts and build a theory that contains them.  If you would like
to take a look at his classic paper "Cognitive maps in rats and men", here
is a URL that will take you right to it...

http://psychclassics.yorku.ca/Tolman/Maps/maps.htm

But the list goes on.  Do you know when the term "psycholinguistics" was
first applied to a conference?  It was in 1952, five years before Chomsky's
first influential book appeared.  The conference was organized by Charles
Osgood, to put linguists and psychologists together to talk about
similarities and differences in their mental models of language as well as
their empirical methods.

To be sure, there was a big non-linear acceleration in ideas and works of
this kind after 1957, but I get a bit tired of hearing people say that
Chomsky started the Cognitive Revolution.  The return to mentalism was
inevitable, for two reasons: (1) accelerating progress in neuroscience
brought with it a renewed hope that a mechanistic base could be found for
mental phenomena, and (2) the birth of computing machines (symbolic and
connectionist, born around the same time) meant that we now had a truly
mechanistic (as in 'machines') set of metaphors for exploring mental
phenomena.  If you look at fields like ethology, developmental psychology
and physiological psychology, you will find that the same big non-linear
blip was happening everywhere.  There was a return to mentalism all across
the behavioral and neural sciences (that would eventually become the
cognitive sciences), promulgated in many cases by people who had never
heard of Chomsky, or if they had, they knew that he was an influential
linguist who ALSO (like them) thought that we could investigate the
contents of mind.  It is no coincidence that Piaget's works were translated
into English or widely read by developmental psychologists in the 1960's
(many years after they were written).  Piaget's influence on developmental
psychology after that point was one more sign of the return to mentalism
that characterized the Cognitive Revolution.  But here is where Chomsky and
his followers hijacked the mentalist movement (or at least their sector of
it): mentalism does not have to go hand in hand with (a) nativism, or (b)
autonomy of domains, but Chomsky's particular version of mentalism
contained (then and now) strong assumptions about the innateness and
autonomy of language.  And in the "take no prisoners" atmosphere that
flourished in that particular community, anyone who did not buy the *WHOLE
PACKAGE* was roundly denounced as a behaviorist.  Piaget is the best case
in point: as nativists got more and more control of the agenda in
developmental psychology, Piaget was pilloried for his belief in
construction and emergence (neither innate, nor learned, a category unto
itself).  Piatelli-Palmarini's edited book "The Piaget-Chomsky Debate" was
a high water mark (and low point in my own career, it was such a depressing
patricidal moment). And yet Piaget was the consummate mentalist, someone
who deeply believed that mind is rooted in biology, and can be studied with
experimental methods like any other biological phenomenon.  His crimes, it
seems, were his belief in emergence (as opposed to a strictly deterministic
form of innateness, one that takes the form of a priori
representations/knowledge), and his belief in the fundamental unity of
cognitive phenomena (as opposed to domain specificity and autonomy).  Those
beliefs were and are unacceptable to card-carrying members of the Chomskian
orthodoxy.

We live in an era today in which the vast majority of neurobiologists are
convinced that cortex is largely constructed, the result of plastic and
bidirectional processes that include genes-->structure but also
experience-->structure.  Activity dependence and plasticity are
acknowledged as primary processes in setting up the brain, including forms
of experience that are going on in utero, with the body teaching the brain
via exactly the same mechanisms that mediate what we have to call
'learning' later on.  The emergentist approach is clearly on the rise in
biology, and in computational neuroscience (in the various branches of
neural network research).  Unfortunately, I think that Chomsky's strong
insistence on a preformationist kind of nativism and a compartmentalization
of the mind are now obstacles to change.  At one time he was an important
part of the cognitive revolution (though he did not do it alone, and it
would have happened anyway even if he had not been around).  Right now, I
think we do need to move beyond those aspects of his views that have been
eclipsed in neuroscience but are still embraced in linguistics.  But it is
hard to make that point, because there is still a very strong sociological
tendency in generative grammar to belittle anyone who promotes emergence,
plasticity, or (God forbid) the kind of complex, rich and neurally valid
associationism that Hebb understood long ago. If you don't buy the whole
package, you are a behaviorist.  Talmy indicates in the following quote
that he is indeed still Chomsky's student in that particular sense:

>Who of the Bloomfieldians would have challenged the Watson/Skinner extreme
>empiricist view of language learning by--ONLY-- rote, memory, immitation
>and S(timulus)-R(esponse)? Who would have challenged Bloomfield's
>anti-meaning and anti-mind dogma? Who would have raised the possibility
>that beyond the surface item-and-arrangement strucures that Bloomfield
>urge us to catalogue, classify, disect AND THEN QUIT, there lay a system
>that 'supported semantic interpretation'?

Who indeed?  My answer is: a whole lot of people, in linguistics and
psychology and neuroscience and computer science, people were challenging
the Watson/Skinner extreme for many many years.  Give up on that scarecrow.
It doesn't exist, and in reality, it was never more than a kind of
radical-chic stance taken by a handful of psychologists just to see how far
they could go.  In some respects, I think that Bruce Richman's call for us
to move "Beyond Chomsky" has come in a little late.  Many of us already did
that a long time ago.  What we need to do now is to sort through a complex
landscape of beliefs (emergence vs. learning vs. unfolding of innate
knowledge; modularity vs. interactionism; autonomous syntax vs. embodied
grammar) and figure out how they can be reconfigured and recombined for a
new era.   -liz bates