LINGUIST List 4.436

Wed 09 Jun 1993

Disc: GB and non-GB

Editor for this issue: <>

Directory

1. , Re: 4.407 Agreement & GB
2. Spencer A J, Re: 4.421 GB and non-GB
3. John S. Coleman, 4.421 GB and non-GB
4. "Sam Wang, 4.421 GB and non-GB

Message 1: Re: 4.407 Agreement & GB

Martin Haspelmath asks how GB theorists manage to agree on so many
assumptions. In the case of agreement, the answer would seem to be that
they don't. From the beginning of the split-INFL proposals, it has been
unclear whether to put TP above AgrP or vice versa, and indeed by no
means all linguists working in the GB model have "bought" the split-INFL
sentence structure. See, for example, the paper by Wexler & Poeppel in
the recent issue of "Language" in which they argue for IP and CP but
not TP or AgrP in child German.
In the same issue, Stephen Anderson (has published in L.I., thus could
be seen as a GB person) takes syntacticians to task for treating
morphology as syntax. What worries me about the split-INFL model is
the way it equates lexical and grammatical categories, which have
traditionally been argued to be radically different kinds of elements.
Also, the evidence for projecting functional categories has often been
tenuous at best, perhaps especially in the case of AgrP which Pollock
motivated on the grounds of adverb placement in negative infinitive
constructions. This is reminiscent of the English "split infinitive"
problem, and I suspect that both cases lie at the limits of core grammar -
see Baker's L.I. paper on English "not" (oops, I'm talking like a paid-up
GB person...)
The obvious alternative to projecting Agr is to treat it as a feature,
as in GPSG and related models. GB has been reluctant to invoke features,
for reasons that are obscure to me. Any offers?

Steve Matthews

Message 2: Re: 4.421 GB and non-GB

In response to Martin HASPELMATH's comments about the homogeneity of the
GB-syntax community - it's not clear to me that GB syntacticians do, in
fact, share a large number of assumptions. It would be interesting, in
particular, to take one assumption cited by Martin, that inflectional
"morphemes" project phrases in the syntax, and find out how many
syntacticians really do believe that (and to what extent they believe it,
for instance, is there a separate CaseP node for case endings in languages
like German or Russian?). I think it would be very difficult to find any
specialists in *morphology* working broadly within the GB paradigm who
would espouse any strong version of that view, for instance.

Andrew Spencer
Department of Language and Linguistics
University of Essex
Colchester
CO4 3SQ tel: (0206) 872188
United Kingdom fax: (0206) 873598

e-mail: spenaessex.ac.uk

Martin Haspelmath recently asked:

> how do Chomskyans manage to agree on so many assumptions
> (e.g. that S-bar is CP, or that inflectional morphology has its own node(s)
> in the tree)?

If Chomsky said that's how it is, then if you disagree, you
are not a Chomskyan. The -an suffix implies agreement on
assumptions, doesn't it?

--- John Coleman

I have an objection to Martin Haspelmath's recent posting when he says
that linguistics in Taiwan is mainly GB. While GB is a significant trend
in Taiwan, there are quite a few linguists in Taiwan who do not work in
the GB framework, and they have also made significant contributions.

 Regards,
 Sam Wang (Wang Hsu)
 swangualtavm.bitnet