Corpora: Chomsky and corpus linguistics

[James L. Fidelholtz]

On 27 Apr 2001 ramesh at clg.bham.ac.uk wrote:

[snip]
>Corpus linguists are more interested in explaining "what is common
>or frequent", which is closer to "what is probable", hence I suppose
>the attraction of statistics. "What is possible" seems to require
>a binary yes/no type of answer, "what is probable" suggests a
>cline or spectrum. Language is a part of human behaviour, and
>almost everything seems to be possible within human behaviour.
>However, corpus linguists are happy to say "this type of (language)
>behaviour is rare" because we have little or no evidence for it,
>but we would not say "it is impossible".

[a little more snip]

	Hmmm.  Maybe I'm not cut out to be a 'real' corpus linguist, if
this is true, since my principal interest is in relatively 'rare'
phenomena.  Or maybe it's an indication that I am basically a
descriptive linguist (I've even been known to do basic Chomskyan
armchair linguistics).  However, I still have trouble really feeling the
much-discussed opposition, or even really much tension aside from a few
people's comments, between corpus linguistics and (Chomskyan or
non-) descriptive linguistics.  Basically, it seems to me that only an
idiot practitioner of descriptive linguistics would a priori exclude
evidence from corpus analysis.  How one analyzes the evidence, of
course, can produce good or bad linguistics; from where I sit, this 
depends more on the analyst's insights than on their theory, assuming
they have some sort of reasonable theory.
	Have a nice weekend.
		Jim

-- 
James L. Fidelholtz			e-mail: jfidel at siu.buap.mx
Posgrado en Ciencias del Lenguaje	tel.: +(52-2)229-5500 x5705
Instituto de Ciencias Sociales y Humanidades	fax: +(01-2) 229-5681
Benemιrita Universidad Autσnoma de Puebla, MΙXICO