Re: Greenberg's Word Order Universals

Raymond Brown wrote:

>Yes, one would like to see Greenberg's actual evidence for his 'universals'.
>
>And to say that SVO may occur as an alternative does not IMHO tell us much.
Except the point that VSO languages cannot have a rigid order.  There are SOV
languages that do not allow S or O to occur after the V, so why should that be
the case for VSO?  I find it significant, even if somewhat vague, and with
much
more to be investigated.

>>60 langs from all corners of the globe come to mind, but I'm
>>really not sure.  There was a 300 lang one (IIRC), but I don't know who
>>did that one.
>How much credence would a statistician give to results obtained from
>Greenberg's sampling?
There's been lots of work on that exact topic since Greenberg started this:
how
to create a statistically significant sampling.  I take this as an implication
that perhaps Greenberg's sample was not significant enough.  If the 60 figure
is right, then it surely is not.  300 may be, if the selections were random,
and distributed across families and areas so that there is no clumping.
Clumping would introduce the possibility that similarities are due to
borrowing
or genetic inheritance.  The sample also has to include all language types
(eg., agglutinative vs. isolating; SOV vs. VOS; ergaive vs. nominative vs.
active; etc).  There's an excellent discussion of this problem in Croft's book
"Typologies and Universals".






===============================
Marcus Smith
AIM:  Anaakoot
"When you lose a language, it's like
dropping a bomb on a museum."
  -- Kenneth Hale
===============================