The appeal of P&P

[Borsley R D]

I'm sure Andreas is right that P&P people feel that they are engaged in an
explanatory enterprize whereas the rest of us are merely describing facts,
but I think perhaps I draw different conclusions from Andreas. Obviously
P&P talk about explanation relates to acquisition. It is important, then,
to ask whether P&P derives any support from facts about acquisition. It
seems to me that P&P acquisition work in fact has the same sorts of
weaknesses as other P&P work: handwaving, dubious argumentation and sloppy
handling of data. My colleague, Martin Atkinson (who is a committed
minimalist), has a useful paper here (M. Atkinson, 'Now, hang on a minute:
some reflections on emerging orthodoxies', in H. Clahsen (ed.), Generative
Perspectives on Language Acquisition, Cambridge: Cambridge University
Press).

It seems to me that it is important to distinguish between P&P and P&P
work on acquisition on the one hand and the idea that there are innate
parameters and that acquisition is largely a matter of parameter-setting.
Objections to the former are not necessarily objections to the latter.
Thus, Janet Fodor, who is hardly an advocate of P&P, defends the latter in
a review of Peter Culicover's Syntactic Nuts, forthcoming in JL. It is
possible in fact that the two are in conflict. It is possible, that is,
that P&P assumptions about syntactic structure pose problems for
parameter-setting. This is suggested by Janet Fodor at the end of her
contribution to the Baltin and Collins book. Moreover, even if there are
objections to parameters and parameter-setting, it doesn't follow that
radical empiricism is right. As Ackerman and Webelhuth stress in their
work on archetypes there are intermediate positions that deserve
attention.

One thing that is very unfortunate here is the fact that there is hardly
any acquisition work drawing on HPSG. Criticisms of a body of work are
always more persuasive if there are alternatives available. It seems to
me, therefore, that we should either be looking at acquisition data
ourselves or bringing the virtues of HPSG to the attention of those who
work on acquisition.


Bob

Prof. Robert D. Borsley
Department of Language and Linguistics
University of Essex
Wivenhoe Park
COLCHESTER CO4 3SQ, UK

rborsley at essex.ac.uk
tel: +44 1206 873762
fax: +44 1206 872198
http://privatewww.essex.ac.uk/~rborsley