Object-initial Languages

[Dan Everett]

Folks,

Like Spike, I share concerns about using subordinate clauses
as a guideline for determining 'basic word order'. In fact, I am not in
favor of using guidelines like this at all, except as a posteriori
illustrations.

I suggest that we be a bit more Sapirian in our approach to these matters
and not allow ourselves to be misled by superficial similarities. The
following points might be worth bearing in mind:

1. In almost all languages, 'word order' is a misnomer for 'constituent
order'. What kinds of constituents are we talking about for a given
language? How did we arrive at this inventory of constituents? What
assumptions underly this inventory?

2. The term 'basic' has no status in linguistic theories. And as soon as
we attempt to make it more precise, we immediately expose the vast
differences between approaches and what one or the other will consider to
be subsumed by the informal term, 'basic'. For example, are we talking
about 'underlying'? 'input'? 'unmarked' in elicitation? 'unmarked' in
discourse? most frequent? etc. Comparing these terms (e.g. most frequent
vs. underlying) is likely to be unfruitful, falling into what Popper
called 'essentialism'.

3. Some theories (e.g. RRG) assert that there are no such things as
universally applicable grammatical relations. For such theories
classifying languages as SOV, VOS, etc. is always going to be misleading.

4. Languages should first be described carefully before they are plundered
for theoretical points. Does each language in a survey have a well-argued,
detailed grammar? What kind of field experience was that grammar based on?
What kind of corpus?

I suspect that most on this list already agree with most of what I just
said. Perhaps it is useful, though, to remind ourselves of such things
from time to time.

Best,

Dan Everett



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

Daniel L. Everett
Professor of Phonetics and Phonology
University of Manchester
Oxford Road
Manchester, UK
M13 9PL
dan.everett at man.ac.uk
44-161-275-3158