[Corpora-List] Identity crisis, re: Quantitive Corpus Linguistics

[Bob Parks]

I'd like to throw in an additional question about the way corpus 
linguistics is developing, in relation not only to linguistics, but 
to other disciplines.

My question is about the status of relations between corpus 
linguistics and social scientists who apply  "qualitative methods" 
with text analysis software (AtlasTi or NUDIST etc.). In other words, 
how important are the numbers; and at what stage of analysis are they 
important? Are there any references to researchers who combine 
approaches?

This question seems important, I think, in developing corpora. Must 
all corpora be chosen from "natural" sources - i.e., sources 
generated autonomously? or is it permissible to include solicited 
text, or text generated in response to researcher questions, etc. To 
what extent, in other words, could we envision a research approach 
that acknowledges the importance of the social context of the text 
that is analyzed?  Halliday, Bernstein and others seem to show how 
important that approach might be; and Wikipedia shows how powerful 
"solicited text" can be.  I'll try to be more specific: could anyone 
comment on the value of a Wiki developed with the corpus/qualitative 
researcher in mind?  What might that entail?

Bob

At 12:21 PM +0100 8/22/08, J Washtell wrote:
>Dear corpus users,
>
>I have just taken the time to read Wolfgang's "My version of corpus 
>linguistics (in 25 theses)". I should have done so earlier, as it 
>renders my own recent points (and some of the present discussion) 
>rather impotent. Clearly, there are many like me, who value corpora, 
>and who have linguistic pretensions, but who do not subscribe to the 
>strict (perhaps occasionally intrinsically questionable??) 
>disciplinary branch described therein; but that is certainly not to 
>challenge that it is a valid pursuit.
>
>As a PhD student with little (cognitive-)linguistic background, this 
>is all very interesting to me. The trouble I am left with then, is 
>knowing what to call myself. I suppose I am one of these 
>"computational linguists". Only my interest in language does not 
>necessitate the use of a computer. Where it does I use corpora quite 
>intensively because my guilty inclination is to "trust the text" 
>(especially when there it so much of it readily available).
>
>I don't suppose any of this is a problem as long as nobody asks what 
>my field is. I wonder if this situation is common amongst subscribers 
>to this list?
>
>Justin Washtell
>University of Leeds
>---
>"This is an integral part of this, but this is not an integral part of this."
>"What do you mean by 'this'?"
>"That."
>"No, I mean the other this."
>"The stool."
>"Ah! Sorry, I thought I'd flushed that."
>
>
>_______________________________________________
>Corpora mailing list
>Corpora at uib.no
>http://mailman.uib.no/listinfo/corpora


-- 
*  The best dictionary and integrated thesaurus on the web: 
http://www.wordsmyth.net
*  Robert Parks - Wordsmyth - (607) 272-2190
* "To imagine a language is to imagine a form of life."  (LW)
* "Philosophers have only interpreted the world. The point, however, 
is to change it." (KM)
*  Community grows as we communicate, honing our words till their 
meanings tap the rich voice of our full human potential. 




_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora