Autonomous Syntax & Research

Daniel L. Everett dever at VERB.LINGUIST.PITT.EDU
Fri Jan 10 16:26:42 UTC 1997


Folks,

I agree with Fritz on the idea of autonomous syntax, which won't surprise
you, but of course it is a hypothesis, subject to empirical testing like
any other. It has stood the test of time pretty well, though.

But I am writing on something else, namely, what autonomous syntax has to
do with research methodology, which is really how I interpret a lot of the
comments on looking at texts. In my opinion the view one holds on the
autonomous syntax thesis ought, in most cases at least, to have very
little impact on methodology.

Data should come from natural text *and* isolated sentences. In my
fieldwork, I rely crucially on both. Usually, I take sentences from
natural texts and study them in isolation (i.e. creating paradigms based
on them and checking them with a variety of native speakers) after I have
analyzed their role in the text from which they are extracted. But
occasionally I need to look for aspectual (etc.) combinations that are
rare or nonexistent (and whether or not they are nonexistent is often what
I am trying to figure out). In these cases I put on what might look like
little stage plays, working with various informants (I usually only do
fieldwork in monolingual situations, having a bilingual informant is a
luxury I have rarely had) to get at the examples (or not) that I am
looking for.

It is true that a lot of work in formal linguistics has been
methodologically inferior to work done in functional linguistics, so if I
were a functionalist, I might count that against formal approaches. But
bad practice does not make bad theoretical assumptions, just, perhaps, bad
theoretical results. There is no justification for this whatsoever and as
a formal linguist, I am sorry that we have lagged behind (although notable
exceptions come to mind, such as Ken Hale).

-- DLE

P.S. Carson Schutze has recently published a book on methodology
(basically for formal linguists), which hopefully will contribute to
change in formal theoretic work.


******************************
******************************

Dan Everett
Department of Linguistics
University of Pittsburgh
2816 CL
Pittsburgh, PA 15260
Phone: 412-624-8101; Fax: 412-624-6130
http://www.linguistics.pitt.edu/~dever



More information about the Funknet mailing list