Regarding query: http://www.linguistlist.org/issues/17/17-411.html#1 
 
Regarding sum: http://www.linguistlist.org/issues/17/17-615.html#1 
 
A few weeks ago I posted a query about the order of subjects and predicates 
in nonverbal clauses, such as that seen in (1), from Indonesian. 
 
(1) Dia guru 
 3sg teacher 
 ‘S/he’s [a] teacher.’ 
 
Crucially, I was interested in cases in which the order of the subject and 
predicate in the nonverbal clause was different to the order of the subject 
and predicate in a verbal clause. In Indonesian this is not the case; but 
some languages have, for instance, V-initial order in verbal clauses, but 
S-initial order in non-verbal clauses. 
(And, of course, many languages have variability in the order of elements, 
depending on some semantic property or other. More on this later.) 
 
And I got some responses, and posted a summary that included them, as well 
as a clarification of the original question. 
All well and good; but since the sum, I’ve received more information from 
LL readers, enough to think I should post a follow-up sum. (Advice to 
future LL query-posters: wait more than just a couple of weeks to post your 
summary!) 
 
We now, thanks to Dick Hudson, Philip Davies, and Bruno Estigarribia, have 
some more information. And it’s interesting, and a 
confirm-a-hypothesis-that-wasn’t-there-before way. 
 
Bruno points out that (verb-initial) Mixtec or Zapotec show similar 
behaviour; and Dick mentions Arabic as a rigidly VSO language that 
nonetheless has S PRED order in nonverbal clauses. 
 
(in the previous sum another language that featured strongly was Ge’ez, a 
classical language from Ethiopia, which is VSO. Verb-initiality is important?) 
 
Philip mentions Palauan as a language that shows a lot of SVO, but is 
underlyingly VOS, yet has S PRED in nonverbal clauses. On its own this is 
just another example; but combined with the fact that Enggano and Tukang 
Besi also show the same pattern, and that these languages are not only all 
related to each other (they’re all Austronesian), but are all found on the 
very fringe of the Western Malayo-Polynesian zone (Palauan: far east; 
Enggano: south-west; Tukang Besi: far south-east), we have an interesting 
distribution. 
But Palauan, Enggano and Tukang Besi are widely separated; there’s no 
question of mutual interference here. How is it that they all ended up with 
the same (typologically rare) pattern? 
 
Now, some more relevant information: Givón (1977), and Payne (1995) 
point out that, when you have VS and SV as alternatives, the SV option is 
used for more stative, less eventive predicates: and, surely, that will 
very nicely characterise the semantics of nonverbal predicates, almost by 
(a) definition. 
 
This suggests a possible historical scenario for the Austronesian languages 
mentioned earlier: predicate-initial languages have a subject-initial 
option, used (among other reasons) when you have less eventive predicates. 
This becomes grammaticalised first with nonverbal predicates, and later 
spreads to be a general subject-first parameter (the languages in the 
‘inside’ of the southern half of the WMP zone are usually SVO, S PRED). 
 
Outside the Austronesian cases, we have V-initial languages with an S PRED 
order in nonverbal clauses; no cases of V-final languages with PRED S in 
nonverbal clauses. Partly this reflects the lack of motivation for 
postposing an S (though, if my speech is representative, the number of 
postposed putative ‘antitopics’ has been growing in the last couple of 
decades in English: things like ‘It’s hard, that homework.’; and see the 
note on Sanskrit in the earlier sum). 
But the fact that, if anything, a smaller proportion of V-initial languages 
lack copular verbs than for V-medial and V-final languages means that 
there’s a definite pattern emerging here. 
 
Thanks again to all who wrote in. 
 
-Mark 
 
References: 
Givón, Talmy. 1977. The Drift from VSO to SVO in Biblical Hebrew: 
The Pragmatics of Tense-Aspect . Charles Li, ed., Mechanisms of Syntactic 
Change: 181-254. Austin: University of Texas Press. 
Payne, Doris. 1995. Verb initial languages and information order. In Pamela 
Downing and Michael Nooning, eds., Word order in discourse: 449-485. 
Amsterdam: John Benjamins.