Date:  Mon, 28 Feb 1994 09:32:00 EST

From:  "Dennis.Preston" <22709MGR[AT SYMBOL GOES HERE]MSU.BITNET>

Subject: Re: there's



T. Daniel Seely (Eastern Michigan University) and I gave a paper on

'Disagreement' at the Michigan Linguistic Society in 1990. We worked with both

conversational data and traditional theoretical instruments (i.e., our

intuitions). We arrived at three principles which encourage disagreement:1)

Phonological processes a) Contraction (e.g., 'what's' versus 'what is'), b)

dissimilation ('where're' versus 'how're'); 2) Word order markedness (e.g.,

'In the shed is some cats' versus 'Some cats is in the shed'), and 3)

Processing (linearity) constraints (e.g., 'There is one man and two

women...'). These principles were confirmed by our conversational data, but

the sample was so small that we are not happy with the holes it left and with

the lack of opportunity to look at other (more subtle) constraints. Since

there is no extensive corpus of spoken AmerEng in which such itmes can be

studied, we'll have to wait around I guess.

Dennis Preston <22709mgr[AT SYMBOL GOES HERE]msu.bitnet>