[From: Bruce Nevin (Wed 930331 08:37:59)]
Ken Hacker, I can sympathize with your views (though I would urge
that it causes as much trouble to take offense as it does to give
it). Bill, Rick, and others here have presumed that linguistics
prior to PCT was just another case of S-R IV-DV behavioral
science. Avery remarked that it is not. (Refer also to
Chomsky's classic review of Skinner's _Verbal Behavior_.) Bill's
rejoinder was that this could be, if linguists just happen not to
do any lab work.
Now I do agree with the view of some linguists that more
experimental work is needed, as opposed to Chomsky's stated
aversion to "data bound" linguistics. But linguistics is
concerned with a different order of problems, distinct from the
problems before any study of behavior. I believe this
distinction also applies to all the social sciences. I have been
working to bring this different order of problems under the PCT
umbrella.
I stated it yesterday (Tue 930330 08:36:29):
Language is not behavior. Speech is behavior. Language is
socially standardized reference perceptions, the control of which
has speech as a byproduct. The reference perceptions are
socially available without recourse to the Test because they are
learned (and taught--cf. Bruner) as socially standardized
perceptions.
There is good evidence that this distinction between competence
and performance (to use Chomsky's terms) applies to the social
sciences generally. That is, behavior that is properly studied
in the social sciences is a byproduct of control according to
socially standardized reference perceptions. The appropriate
subject matter of the social sciences, including linguistics, is
not the behavioral outputs but rather the socially standardized
reference perceptions and the processes by which individuals come
to calibrate their private reference perceptions according to a
perceived public standard. Dan Miller described some early
(1973) investigations of "redundant sequences and patterns".
Some years ago there was a William Pierce at UMass Amherst whose
study of what he called constitutive rules in social interaction
still seems to me to be very fruitful. Perhaps I should dig out
some papers of his that I got in the 1970s.
To the point here, the study of socially standardized reference
perceptions is not subject to the same experimental techniques as
the study of behavioral outputs as byproducts of experimentally
determined reference perceptions. The former has a "meta-"
relation to the latter. Other sorts of questions must be
addressed, questions about how people establish and change
reference perceptions, how the learning process (or "acquisition"
process, the question-begging term preferred in linguistics) may
relate to higher-level perceptions such as the perception of
being in a cooperative relationship, being able to rely on
another, being perceived as reliable by others, whether or not
some of these perceptions are "intrinsic", and so on.
It is not clear to me how experiments regarding these matters
might be designed. Part of the problem is my lack of experience
and training in experimental methodology, but part of it is that
there is no obviously direct transfer of methodology from PCT
modelling of behavioral outputs to PCT modelling of calibrating
one's reference perceptions to a social standard. My time and
energy are severely limited, and my imagination probably as well.
I would greatly appreciate help in this from anyone.
Bruce
bn@bbn.com