[An-lang] Austronesian preposition-demonstrative connection

[Don Killian]

Hi all!

Sorry to bother the list, but I recently joined to get some feedback on 
some work I've been doing, especially if there are any of you who are 
diachronically minded!

I'm not an Austronesianist, but rather interested in deixis and 
demonstratives, which can be very interesting in that part of the world.

A number of Austronesian languages, even those distantly related, appear 
to show a connection with demonstratives and adpositions, although the 
precise connection seems to vary depending on the specifics.

The most frequent type is that like Buru and Lewotobi Lamaholot, where 
locative demonstratives can optionally take adjuncts, e.g.

Da kadu-k na huma
3s come prox house
‘He came (here) to the house.’ (Grimes 1991: 171)

Da puna huma naa
3s do house prox
‘He made this house.’ (Grimes 1991: 171)

Da kadu-k na huma naa
3s come-k prox house prox
‘He came here to this house.’ (Grimes 1991: 171)

I view this as something of a structural parallel in English, e.g.

The man went inside.
The man went inside the house.

However, Indo-European languages do not allow for this type of 
construction with locative adverbs.

The man went there.
*The man went there the house.

I'm not entirely confident that field linguists have always checked to 
see whether constructions like this would be possible, but I have a 
database of over 1000 languages and I have barely seen it outside of 
Austronesian so I can conclude that it is at the very least likely 
rather rare.

A second type is like Begak, where adpositions are derived from locative 
adverbs, and they show a deictic distinction (di' contrasts with nong in 
Begak).

jadi panow kat bano ino di’ dapur adi
so go cdm husband yonder prep.dist kitchen over.there
‘So her husband walked to the kitchen (..)’ (Goudswaard 2005: 91)

A third type is like Paulohi and Thao, where prepositions originate from 
deictic verbs with a meaning of "be here, be there:"

na ka-tue wei wesie nei
let 1pi-sit prox.at woods here
‘Let us sit here in the woods.’ (Stresemann 1918: 61)

Aside from their verbal nature, they seem to show a number of parallels 
to the type seen in Buru.

A fourth type is a bit more marginal, discussed in Lichtenberk 1991. In 
languages like these, you get some sort of deictic distinctions in 
adpositions, but they originate from the verb mai, 'to come', (and 
sometimes go) rather than demonstratives, like in Tetum Dili:

Nia haruka surat ida mai Dili
3sg send letter indef prox.to Dili
‘He is sending a letter to Dili. (speaker is in Dili)’ (Hull & Eccles 
2004: 145)

Nia haruka surat ida ba Dili
3sg send letter indef dist.to Dili
‘He is sending a letter to Dili. (speaker is NOT in Dili)’ (Hull & 
Eccles 2004: 145)

Lichtenberk discusses parallel examples in Fijian with static locations 
and mai/e.

Why I'm writing to the list here is that I can't see how this could be 
reconstructable, but at the same time, I also find it incredibly 
suspicious that such constructions have popped up so many times 
independently even though almost the entire rest of the world shows 
nothing remotely similar. Thao, Buru, and Begak are neither part of the 
same geographical area, nor are they closely related. I could possibly 
accept Buru, Paulohi, and Lamaholot having some sort of geneological 
explanation with Central Maluku sharing a closer connection to 
Bima-Lembata than had been previously thought, but the person arguing 
for such a connection is going to then have to explain why so many other 
Central Maluku and Bima-Lembata languages do NOT have this feature.

"di" I believe is reconstructed as a preposition for Austronesian, and 
it seems to also pop up as some sort of distal demonstrative in any 
number of languages (https://www.trussel2.com/acd/acd-s_d.htm#8796). 
It's not reconstructed as a verb for be/exist that I'm aware of.

Are the two di's connected?

And can any historical linguist give any thoughts or opinions about 
these constructions? The only thing I can think of is that the origin of 
all of these is in locative verbs like 'be here, be there', which 
somehow lost their verbal nature in most languages. This would allow for 
the flexibility in grammaticalizing into either verbal adpositions or 
then demonstratives, depending on the language. Under this theory, these 
four types could be collapsed into 2, the "come" languages and the "be 
(t)here" languages.

The problem is that I haven't seen a great deal of evidence for di, na, 
etc. to be reconstructed as be there, be here.The closest I can think of 
is 'i' (https://www.trussel2.com/acd/acd-s_i1.htm) in Paiwan meaning be 
at, in.  'inay' in Thao also can be a verb meaning 'be here', but I'm 
rather hesitant to draw conclusions about lexical categories in any 
language of Taiwan.

Apologies for the long email, and I welcome any thoughts or discussion! 
As to not to spam the list, you can also send in private additional 
languages which also fit into some of the types I mention here.

Best,

Don

-- 
Don Killian
Researcher in Linguistics
Department of Languages
PL 24 (Unioninkatu 38 B)
FI-00014 University of Helsinki
+358 (0)44 5016437
_______________________________________________
An-lang mailing list
An-lang at anu.edu.au
https://mailman.anu.edu.au/mailman/listinfo/an-lang