Competence vs. Performance: Summary

[Lois Bloom]

Reliving History. . .

First, in his posting on October 12, Matthew Saxton shared his "depression" over the "internecine struggles that have lasted for decades" and the "relative lack of empirical maturity in the field of child language."  He then described as "derisory" the "sample sizes of many studies in child language" . . .  "compared with other branches of the human sciences."  WELL . . .   having spent more than 40 years and published 7 books and very many papers in the peer-reviewed journals, I own up to spending all those years and all those printed words on studying the language acquisition of fewer than 25 children.  And so I sighed.  

And then on October 14, Joe Stemberger shared his experience "a few years ago" in which his work was derided because it "clearly had more to do with performance, and why would a linguist be interested in analyzing such data using the mechanisms of linguistic theory," and I sighed again.  The criticism was very familiar to me having heard it often, many years ago.  And I suspect it was familiar as well to others who have also spent hours, days, years listening, watching, and interpreting the sounds, words, sentences of very small children --yes, unabashedly studying their performance in the sincere effort to learn something about what they *knew* about language and how that *knowledge* changed over time.  In my view of history, that criticism is the very heart of how the competence-performance distinction did, indeed, hold back progress in the field.  It has been used as a club to knock the research of those of us who do not rely on introspection and grammaticality judgements, or who do not do experiments.

And now today on October 16, Matthew has shared a very useful summary of what I found to be one of the more insightful discussions on the InfoCHILDES talkbank.  I thank him for the summary and for starting the exchange as well.  Please add one more citation to the list: 

Bloom, L., Miller, P., & Hood, L. (1975). Variation and reduction as aspects of competence in child language. In A. Pick (Ed.), Minnesota symposia on child psychology (Vol. 9, pp. 3-55). Minneapolis MN: University of Minnesota Press.   Reprinted in Bloom, L. (1991). Language Development from Two to Three (pp. 88-141).  Cambridge University Press. 

in which we presented data (from performance, but data nevertheless) and proposed a model or 'theory of knowledge' to account for the variable length of early sentences, building on the "probabalistic grammars" of Patrick Suppes (1970) and the variation paradigm of Bill Labov (1969) and Cedergren and Sankoff (1974).  (Really digging into ancient, pre-internet history here!)  

And so I applaud Matthew's conclusion that promotes a level of complexity beyond the simplistic C-P distinction: "the acknowledgement that an additional, and perhaps more interesting, source of linguistic variability stems, not from physical breakdowns, but from numerous other factors that could correctly be seen as part of speaker's knowledge of language."  In our 1975 study, more than 30 years ago, my colleagues and I  grappled with the "empirical headaches" he refers to, "trying to identify and explain what [those factors] are and how they function."   

Rather than continuing to escape into the antiques business. . .  it's a gorgeous day today, and I'm heading out to the golf course.

Cheers!  

 

LOIS BLOOM
    (Edward Lee Thorndike Professor Emeritus, Teachers College, Columbia University)
95 Wilson Road
Easton CT 06612

phone: 203-261-4622
mobile: 203-673-7021
fax: 203-261-4689
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/info-childes/attachments/20071016/dbef8008/attachment.htm>