Competence vs. Performance: Summary

[Adele Abrahamsen]

Hello, all.

I came late to this discussion but just enjoyed reading the whole sweep of
contributions all the way through Lise Menn and Lois Bloom’s post-summary
reflections. Clearly the topic hasn’t been exhausted—each message makes
thought-provoking points—so I will add a mention of the framework I worked
out after years of poking at the question of how psychology and
linguistics should, and do, interact.  See:

Abrahamsen, A. A. (1987). Bridging boundaries versus breaking boundaries:
Psycholinguistics in perspective. Synthese, 72, 355-388.

Abrahamsen, A. A. (1991). Bridging interdisciplinary boundaries: The case
of kin terms. In C. Georgopoulos & R. Ishihara (Eds.), Interdisciplinary
approaches to language: Essays in honor of S.-Y. Kuroda (1-24). Dordrecht:
Kluwer Academic Publishers.

Although linguistics, psychology, and their relationship have become more
diverse and complicated in the years since I wrote these papers, I still
find the basic framework useful especially in considering classic notions
like competence. In brief, I emphasized that linguists study language as a
cultural product—as an abstract structure resulting from behavior—and
psychologists focus on behaviors in real time. I also suggested these
enterprises could be connected in ways that bridge boundaries (e.g. a
psycholinguist might reformat a linguistic account to make it a suitable
tool for designing an experiment; a linguist might find in real-time
processing an explanation for aspects of language structure), but they
could instead be connected in ways that break boundaries (e.g. Chomsky’s
notion of competence and its relation to performance).

That is, Chomsky was acting as a linguist in working out various versions
of his generative grammar, and switched to psychology to make his
(influential and controversial) claims about competence-performance and
language acquisition. I wrote (p. 381) “There is nothing inherently wrong
in this; innovative thinkers often wear multiple hats…[but] …[m]any
psychologists, frustrated with a distinction that did not find a natural
fit within their own meta-theory, concluded that a ‘performance’ theory
was all the theory that one needed; essentially, they decided that there
need be no distinction between linguistic and psychological perspectives.”
Arguably, many behavioral scientists and computer modelers pursuing this
path are actually doing their own linguistics (vs. abandoning
linguistics). Others use linguistic theories other than Chomsky’s as a
tool (e.g. cognitive grammar), minus the boundary-breaking view that these
are competence theories.

The 1991 paper suggests how seemingly conflicting analyses and data on kin
terms can be understood in terms of boundary-bridging connections, finding
a role for linguistic abstraction that does not carry all the baggage of
Chomsky’s competence notion. Nonetheless, I find the notion powerful in
that thinking about whether or how it might apply nudges my thinking
beyond bland assumptions. For example, Optimality Theory can at least
provisionally be taken to be a kind of competence theory (albeit one based
on constraint satisfaction). The nature of the relation between OT and
connectionist networks seems to be boundary-bridging, not
boundary-breaking. Does that suggest new directions to go in developing
the notion of competence—or should we just not go there?

Adele Abrahamsen
Center for Research in Language
University of California, San Diego
9500 Gilman Drive, CB 0526
La Jolla, CA 92093-0526