syntax bigotry

Stephen M. Wechsler wechsler at mail.utexas.edu
Wed Mar 24 13:10:05 EST 1999


I'd be interested in receiving any examples you may have of
ignorant/misinformed/dismissive/etc comments about HPSG and other
non-transformational theories.  A common source is reviews of manuscripts
and conference abstracts.  Here is an example, recently sent to me by a
colleague; this is an excerpt from an abstract review:

"In HPSG, the different features - one syntactic, one semantic - seem rather
trivial, not surprising at all, because HPSG features are free to be almost
anything one wishes."

A moment's thought reveals the error here:  while it is true that the
formalism allows one to posit any feature at all, a particular variant of
the theory  involves a specific feature architecture (indeed, "geometric
prediction"-- predictions following from the way the features are set up--
is a key theoretical tool).  The point is that the same situation (actually
a much worse situation) obtains in P&P/MP.  I doubt there is a single
theoretical principle which is adopted in all variants of, say, P&P.

In my experience these stupidities tend to fall into two categories:
1. "HPSG is descriptive, not explanatory."  The above quote is an example
of this.
2. "HPSG is a notational variant of GB."  See my "Balinese Bind" paper in
the Webelhuth et al 1999 volume for discussion of this point.

So, if you have any examples, please send them.  Direct quotes are ideal,
but rough descriptions are OK too.  If I get any responses then I'll post
them back to the list (keeping your identity confidential).  If I get a lot
of responses then I may use them for a paper on the sociopathology of our
profession.

best,
Steve

P.S. If you have seen similarly uninformed comments about transformational
theories then send them too.

Stephen Wechsler, Assoc.Prof. |   http://uts.cc.utexas.edu/~wechsler/
Linguistics Dept.             |   Calhoun Hall 403
University of Texas           |   ph. (512)471-1701
Austin, TX 78712-1196         |   fax (512)471-4340



More information about the HPSG-L mailing list