accusative and ergative languages

[Jens Elmegaard Rasmussen]

On Thu, 27 May 1999, Patrick C. Ryan wrote:

> [...] I would go an unpalatable to some step further, and, agree with G.
> A. Klimov, that an ergative form *must* precede an accusative type.

There are some quite obvious counterexamples: Modern Indic is very close
to being ergative; Kurdish too; Lithuanian is well on its way; and
Eskimo has accusative remains in the pronouns. Also, the "have/be"
periphrasis of perfect tense in Western Europe have passed through a stage
of plain ergative syntax with this part of the verb. In all of these
examples that ergativ has plainly arisen from a passive transformation:
The event was described from the point of view of  what happened to the
object, and in course of time this nuance was suppressed and the
"ergative" structure generalized to the point of ousting the old finite
(subject-centered) verb. This view is often criticized for being based on
the analyst's inability to think in terms of foreign categories, a
criticism that completely ignores the fact that the categories concerned
are in this instance not at all foreign to us.

Jens