Sentences once more

[Wallace Chafe]

Dan Everett's message shows how easily a brief attempt at expressing
something that's really very complicated can lead to misunderstandings. I
hope this won't muddy things further. I didn't mean to say that spoken
sentences are always incoherent, or that there's always a mismatch between
syntax and prosody. In prototypical cases, I think, syntax and prosody do
match, and sentences do express what I would call coherent subtopics in the
flow of thought. I just wanted to point out that the cognitive status of
sentences can be unstable in two ways. First, opportunistic or premature
sentence closure, maybe followed by afterthoughts, is certainly something
that happens fairly often. Second, the subtopics sentences express may
change from one telling to the next. But I didn't mean to say that
sentences are irrelevant or unimportant in speaking. Among other things
they can show the subtopic structure of discourse, and they can show the
variety of ways speakers can decide when they've come to the end of
something, whatever it is. But we can't expect them necessarily to be doing
the same thing all the time. The question I wanted to raise is whether they
should be treated as THE basic unit of linguistic study, as seems for a
long time to have been the case.
By the way, I've never before been accused of ignoring the existence of
fieldwork. How strange!
Wally