Date: Sun, 20 Oct 1996 18:53:05 -0700
From: Robert May <RMAY@uci.edu>
Subject: Remark on Zlatic/Seely/Reeves
To: linconf@tamvm1.tamu.edu
Message-Id: <326AD781.463B@uci.edu>
I'm not quite sure I'm following this discussion about
pragmatic/processing. Zlatic said "the majority of my informants (8 out
of 11) did not accept binding . . ." OK, so the issue is about the
other three, who presumably did accept binding by the subject of a
non-process nominal. The issue is to say what their grammar is, such
that this is so, since I presume that for these speakers this is just as
much their honest report of the data. Now, if I understand Reeves'
point, what is going on here can't be of the processing/pragmatic sort
Zlatic seems to envisage, since these 3 speakers presumably make a
judgement over a particular interpretation. This is a good point, and
indicates that they have a different rule - for them, non-process
nominal subjects are arguments. The question here just seems to be why
can there be divergent grammars of Serbian at this juncture?
As a general matter, I think that Reeves makes a very good point. It
really doesn't do just to say that a majority of speakers get a
judgement (and indicate this through the affixing of '?*', which is
completely misleading), and dismiss the judgements of the minority. The
point is that that the minority are also speakers of the language, and
their judgements have to be accounted for in just the same way as the
majority. We cannot just dismiss the minority as suffering from some
pragmatic or processing deficiency - they as much speak and make
judgements in accordance with the rules of their grammar as the
majority.
-Robert May