A question about ergative markers

[Johanna Nichols]

Bill McGregor asks:

>Dear typologists,
>
>I am wondering if anyone knows of any language with more than one ergative
>marker?
>
>Obviously I'm not thinking of more than one allomorph, but where there are
>two morphemes both marking ergative, but with some nuance. For instance,
>Warrwa has -na @ -ni and -ma. It seems that the first is an ordinary
>ergative marker, while the second is a focal ergative marker, marking
>unexpected agents.
>
>If anyone is aware of anything like this in any other languages I'd
>appreciate hearing about it.
>
>Thanks,
>
>Bill McGregor


In contemporary spoken Ingush (Northeast Caucasian) one ergative allomorph
is expanding at the expense of another, and there seems to be some semantic
or syntactic conditioning going on at the frontier of the expansion.  The
ending -(a)z was originally found only on a few human nouns and -uo on many
others (human and non-human).  -(a)z is expanding and is now common on on
many human nouns and on other animate nouns.  Sometimes it is possible to
elicit minimal pairs in which an ergative noun with the -(a)z allomorph is
clearly subject while one with the -uo allomorph feels more like the agent
of a passive (e.g. it is favored by the perfect tense, which is resultative
and evidential/non-witnessed in its semantics and easily occurs without an
overt ergative subject).  I don't have any text support for this
conditioning yet, only limited elicitation.

There is a separate instrumental case as well, so if the conditions on
ergative ending choice prove non-ephemeral in Ingush we may have an example
of a language with a three-way opposition of true ergative case,
agent-of-passive case, and instrumental case.  Or, put differently, two
different ergative cases, one for true or prototypical subjects and one for
demoted subjects or less prototypical subjects.

This wording shouldn't be taken to suggest that Ingush actually has a
passive.  The perfect tense is a tense, not a voice.  And I really don't
expect the allomorphy to be frozen into a case opposition.  Still, the
conditions that mediate this diachronic expansion of an allomorph may have
typological interest.

Johanna Nichols

* * * * * * * * * * * * * * * * * * * * *
Johanna Nichols
Professor
Department of Slavic Languages
Mailcode 2979
University of California, Berkeley
Berkeley, CA 94720, USA

Phone:	(1) (510) 642-1097 (direct)
	(1) (510) 642-2979 (messages)
Fax:	(1) (510) 642-6220 (departmental)
* * * * * * * * * * * * * * * * * * * * *