drift from OV to VO

[DAVID GIL]

Bingfu,

I think these are very interesting questions.
Did Fritz Newmeyer ever post a summary to his query?
(If he did, it might be nice to have it put on the
lingtyp list, to trigger further discussion.)

Some thoughts of my own on the drift issue:

First a question:  Is it actually the case that
there's more OV>VO than VO>OV amongst ATTESTED changes,
or only amongst RECONSTRUCTED changes.

If the latter, then the reason for, say, Proto-Indo-
European being OV in the eyes of some historical linguists
*may* (I'm only throwing this out to be provocative) be
the same as the reason for, say English being OV, underlyingly,
in the eyes of this or that flavour-of-the-week generative
framework -- namely the product of a desire to get everything
nicely lined up and serialized and pretty and elegant.

However, if the former, then I see no way to avoid the
conclusion that what we have here is a fascinating window
into linguistic evolution, with OV languages being
evolutionarily prior to VO languages with respect to
word order typology.

(An inevitable disclaimer.  To say that, say, OV Japanese
is evolutionarily prior to, say, VO English with respect
to word order typology is not to rule out the possibility
that, say, OV Japanese might be more advanced evolutionarily
than VO English with respect to other linguistic properties
not correlated with word order.  And needless to say, such
claims say nothing about extra-linguistic matters.)

David Gil