Date: Fri, 11 Feb 1994 00:13:36 -0700

From: Rudy Troike RTROIKE[AT SYMBOL GOES HERE]ARIZVMS.BITNET

Subject: Re: An English Grammar Text



Tim,

Here at the University of Arizona, we give students a pure diet

of 1963-era Transformational-Generative grammar, building that on a

required course in the History of English. We do try to show them how

tree diagrams relate to Reed-Kellogg diagrams, and try to disabuse them

of simplistic elementary-school notional definitions of sentence,

subject, and parts of speech, etc. (including debunking the notions that

gender is a matter of sex and that "possessives" are really that

--genitive remains the better term). Last year we used a text by Kaplan

which is good but requires a good bit of supplementation, so this year

we have gone back to Veit, Discovering English Grammar, which has much

better coverage of the grammar and lots of exercises. It has its

careless moments, as most texts do. We supplement it with material on

English phonology and lots of transcription practice, leading into

spelling rules and the Great Vowel Shift, and with sociolinguistic stuff

using Peter Trudgill's Sociolinguistics. It is a pretty heavy load for

one semester, and really almost too much for summer, though we do manage

to cover most of it even then.

I think early TG grammar still gives the best insights into

things like passives, relativization, question formation, and

nominalizations, and makes coherent sense out of it all in a way that no

other model does, particularly for practical teaching purposes, since it

builds on native-speaker intuition. While I am currently trying to keep

up with Chomsky's Minimalist model (which has replaced GB), I don't

think it is any more usable for classroom teachers than quantum

physics is for introductory high school science. I retain a certain

fondness for Reed-Kellogg diagrams, though they are pretty stultifying

and non-dynamic, unlike transformational-generative grammar.

Rudy Troike