Agreement with animate direct objects in Dani

[Everett, Daniel]

In my grammar of Wari' (Amazonian) with Barbara Kern, I discuss Wari' verbs that break into separate paradigms of synthetic and analytic along these lines, for plural agreement.

Dan


On 4 Dec 2010, at 13:43, Enrique L. Palancar wrote:


Dear Giorgio,
Although does not count as exactly the same phenomenon as the one you're pointing out, Basque verbs came up to me when you mentioned Dani, as it has two types of verbs with two different paradigms, one synthetic (currently about two dozen of them, both transitive and intransitive) and the rest (which are inflected periphrastically), number agreement markers involved are at times different, e.g. synthetic "ekarri" 'bring'  "d-akar-tza-gu" 'we are bringing them' [3.abs-bring-pl.abs-1pl.erg] in contrast to "ikus-i d-it-u-gu" ‘we saw them' [see-perf 3.abs-pl.abs-aux.tr-1pl.erg], but see Rijk's (2008) "progressive" grammar of Basque (MIT), (I would myself call it "impressive" rathern than "progressive"). There are also Mayan languages where the absolutive is a enclitic hosted on an preverbal TAM auxiliary, but as far I recall now this marking pattern applies to all verbs, and there are no subtypes like in Basque.
Best,
Enrique

________________________________
Date: Fri, 3 Dec 2010 16:14:08 -0700
From: giorgio.iemmolo at UNIPV.IT<mailto:giorgio.iemmolo at UNIPV.IT>
Subject: Agreement with animate direct objects in Dani
To: LINGTYP at LISTSERV.LINGUISTLIST.ORG<mailto:LINGTYP at LISTSERV.LINGUISTLIST.ORG>

Dear typologists,

I'm working on my dissertation on Differential Object Marking and Differential Object Agreement, and I found a quite peculiar case of DOA in a Trans New Guinea language, Dani.

Dani has a DOA system based on animacy: animate direct objects (as well as indirect objects and beneficiaries) are signalled on the verb via an affix. The intriguing fact about Dani is that, as many other Trans New Guinea languages, it does allow the object prefixes to be directly attached to the main verb only with a restricted subset of verbs, namely (w)at- “hit, kill”, hei- “put”, and ha- “perceive”. With other verb roots, direct object affixes must be attached to post-cliticised 'auxiliaries', such as hei- “put”, -ha- “see”, -et- “give” or -ap- “do for” (the last two are used to introduce recipients and beneficiaries) that are usually compounded with the main verb stem, forming thus a sort of serial verb construction (cf. Bromley 1981: 157 ff. in his Dani Grammar).

To my knowledge, this feature seems to be unique to Papuan languages: while it seems to be likely that verb serialisation is the source of these constructions (see Foley 1986), I don't know of any other language in which the animate direct object is indexed on an 'auxiliary', nor do I know of  any studies dealing with this rare pattern from either a synchronic or a diachronic point of view. For instance, I find it quite interesting that only the verbs  “kill, perceive, put”, which are quite heterogeneous from a semantic point of view, can freely host the object affix.

I would be very grateful for any relevant suggestions or bibliographical hints on this matter.

Thank you very much in advance,

Giorgio Iemmolo
--
Giorgio Iemmolo
Dipartimento di Linguistica
Università degli Studi di Pavia
Strada Nuova, 65
I- 27100 Pavia, Italy
e-mail: giorgio.iemmolo at unipv.it<mailto:giorgio.iemmolo at unipv.it>
http://lettere.unipv.it/diplinguistica/pagina.php?id=169

*Please do not print this e-mail unless you really need to*



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20101204/e8769889/attachment.htm>