OV to VO und Verwandtes

[Frans Plank]

According to Fritz, Bingfu, and David:

If (a) there are more OV languages today than VO languages
and (b) there are more attested instances of OV>VO than of VO>OV,
then OV must have been even more predominant originally.

What about this analogy?

If (a) there are more languages with a dual today than languages with no dual
and (b) there are more attested instances of dual loss (dual > no dual)
than of dual innovation (no dual > dual),
then dual must have been even more predominant originally.

Both implicantia happen to be true in the case of the dual.  Can dual
therefore be considered 'evolutionarily prior', even a remnant of
'proto-world'?  If you believe it can you will be glad to know that there
is strong evidence of dual correlating with lower cultural complexity (as
measured by G. P. Murdock, Atlas of World Cultures, 1981;  see W.
Schellinger, Dual und Kulturstufe, 1995).  As we all know, cultures are
getting more complex in history (at least as measured by Murdock).

Still, the diversity of extant dual FORMS, pointing to independent
innovations of duals, couldn't suggest more forecefully that this would be
the entirely wrong conclusion to draw.  But what IS the right conclusion?

One possible conclusion is anti-uniformitarian:  Duals used to be innovated
far more frequently in earlier times, prior to our taking stock of dual
losses and dual innovations and counting more instances of the former.
Today we (ex-agri-)culturalists are content with having numerals for
counting to two and more, and couldn't care less for inflections.

Another possibility, and it will be preferred by those professing
uniformitarianism, is that dual innovations are just as frequent as dual
losses, but dual losses generally take more time than dual innovations.
Grammaticalizing a numeral 'two' is something anybody can do within a
couple of weeks.  Cutting down again on your number inflections may keep
you busy for a couple of millennia--unless of course they are wiped out by
a catastrophe, such as pidginization.  (And uniformitarianism and
catastrophism are not as incompatible as has sometimes been implied;  see
B. Naumann, F. Plank, & G. Hofbauer (eds), Language and Earth: Elective
Affinities between the Emerging Sciences of Linguistics and Geology, 1992.)
This means that in a given time span we will have fewer opportunities to
observe processes which take less time, if instances of them are about as
frequent as instances of processes taking more time.

What this means for O and V I leave for you to ponder.

Frans Plank