Corpus-based, intuition-based or tradition-based teaching?

metal56 · Post by **metal56** » Thu May 25, 2006 7:14 am

Article:

http://www.gabrielatos.com/Grammar-Intuitions.htm

JuanTwoThree · Post by **JuanTwoThree** » Thu May 25, 2006 8:55 am

English grammar, as explained in the better standard grammars now in use , generally took pains to emphasise its roots in usage and purported at least in part to be based on observation. In the past though this was really, I suppose, mostly the intuition of the writer.

An intuition that resulted from the often academic world of the writer, his or her social contacts, family and so on. An idea of rightness and wrongness from what the writer had read and heard in a restricted circle.

In the absence of a top-down academy or such like, all English grammar is bottom-up; the difference is that when a writer in the 70's or 80's wrote "some people consider this to be incorrect" rather than "many people" this comment was based either on incomplete data or the writer's too narrow sociolect, not that there was much difference between the two if you ask me.

Basing conclusions on corpora isn't so very different from the best intentions of earlier writers. The difference is that using the corpora gets more accurate results.

We all know from these threads that most of us have been surprised on more than one occasion by the discrepancies between our much lauded intuitions and what the evidence suggests. What's more, nobody has ever suggested that everybody is out of step except them. Would the writers of the grammar books presently consulted by ELT professionals have been any different? Would they have insisted on their rightness if it flew in the face of what the corpora suggested?

Stephen Jones · Post by **Stephen Jones** » Thu May 25, 2006 3:09 pm

An old article (2002) but full of common sense.

I think the point is that our intuitions are a starting point. And whilst they sometimes tell us somethng is incorrect when it is in fact correct or even the most common collocation, they are rarely wrong when they say something is correct.

abufletcher · Post by **abufletcher** » Thu May 25, 2006 6:05 pm

Stephen Jones wrote:they are rarely wrong when they say something is correct.

While intuition may be reasonably reliable for grammaticality judgements (and apparently also collocation issues) it is an extremely weak tool for investigation of discourse issues. Harvey Sacks, one of the founder of the field of conversation analysis had this to say about "introspective" or "imaginary" data:

"...from close looking at the world we can find things that we could not, by imagination, assert were there."