Ergativity & Chibchan

[Diego Quesada]

On Mon, 16 Feb 1998, Jose M. Garcia.Miguel wrote:

> Diego Quesada wrote:
>
> > In my work on Teribe, Rama, and Boruca -all three Chibchan
> > languages of Central America- a similar pattern appears (only that the
> > word-order types in running discourse are not as flexible as in Spanish):
> > SOV discourse-initially, OSV/OVS -almost- elsewhere.

> I'm doing right now some work on textual data of Chibchan languages of
> Costa Rica -Bribri, Guatuso. They are OV languages, with variable order
> of A, resulting as far as I now in AOV and OVA, **but not OAV**.
        [starrs mine, DQ]

        I never said that Bribri or Guatuso's alternative orders are OSV;
in fact, I did not mention those two languages, cf. above. The alternative
OVS order is for Boruca and Teribe; the alternative OSV is for Rama, from
Nicaragua.

> I'm
> interested in the discourse factors that correlate with the position of
> A, and especially in the factors that correlate with the presence or
> absence of the ergative morpheme, which in this languages is sometimes
> optional. I guess this has to do with topicality and with the
> distinction between given and new information. Could you give me some
> tips?
        I'm baking that cake too; so I might not wish to cut the cake
before it is baked. But here is the basic thing for Boruca and Teribe:
full NP's in the SOV and SV orders tend to appear discourse-initially. In
most other instances both 0-anaphora or the alternative orders
(VS/OVS) are commonplace. Just like in Spanish, full NP realization of
referents here and there within a text is used for emphasis or
desambiguation. A collection of Teribe texts (with Spanish glosses) by
me is in press.
        As for the specifics of the invarible ergative-marking of Bribri,
you'll have to contact A. Constenla:

        aconsten at cariari.ucr.ac.cr

I'm not so sure that Guatuso has such a pattern of invariable ergativity-
marking; Guatuso differs from the other Chibchan ergative languages in
that it works more like the Mayan languages (it even has various
antipassives -absent elsewhere in Central American Chibchaland- like those
Mesoamerican languages). The reason, I suspect, has to do with the fact
that ergativity marking in Guatuso (like in Mayan) is expressed in the
agreement -cross referencing- system, while in the other Chibchan
languages it is expressed by direct marking. Now, given that most Chibchan
languages are highly "discourse-run" (lots of 0 anaphora; most tests for
subjecthood such as gapping in coordination, etc. fail there; there is
intermittent marking of grammatical categories such as plural, ergativity,
even person in Teribe -which has a highly grammaticalized person
agreement system), which means that marking of categories is
intermittenty, applied only when needed, the presence/absence of "erg" in
Bribri (and in its closest relative, Cabecar) -and to a lesser degree in
Guatuso- depends, as you correctly suspect, on discourse-pragmatic
aspects.
        For Bribri, too, you will have to contact Constenla, and also C.
Jara

        cjara at cariari.ucr.ac.cr

It goes without saying that the Journal Estudios de Linguistica Chibcha
will be of great help.

Saludos,
Diego