Re: Word Order in typology

Looking at "Describing Morphosyntax" (which I haven't read in a
while).... It says "The grammatical term subject can be defined as S
together with A...." which means that subject means what I always took
it to mean before I got confused. But anyway, using subject in such
universals seems extremely dubious to me unless you can show that in all
languages such a grouping is significant and meaningful... and since all
these universals only seem to be dealing with the word order in
transitive sentences (its not too difficult to imagine a language whose
typical word order is VAP etc in transitive sentences but SV in
intransitives... I doubt its very common, but it doesn't seem impossible
to me), then using a term which covers more than just an agent-like
argument of a transitive sentence seems a bit strange. Indeed, Thomas
Payne himself uses AVP etc as I was arguing for in at least one place in
the book that caught my eye: "Strictly speaking, constituent order
ergativity would only be possible if the verb occurs between its two
core arguments (ie AVP or PVA languages)." P137.
 Also quoting from Payne, talking about GRs "They clearly exist, and
*may* even be universal..." P130. Emphasis mine. The implication is that
Payne isn't even making that claim the GRs are universal, so if they're
not then how can you talk about them in language universals?