Re: THEORY: unergative

Jörg Rhiemeier wrote:

> Hallo!
>
> On Sun, 22 Feb 2004 13:44:38 -0600,
> Herman Miller <hmiller@...> wrote:
>>      But in Dyirbal for instance, which is one of the
>>standard examples of an ergative language, first and second person
>>pronouns are nominative/accusative. In my own language Kazvarad, which
>>was originally a human language (and is currently in an undefined state,
>>depending on whether there turn out to be humans in the Azirian
>>universe), the pronoun prefixes on verbs are nominative/accusative,
>>while nouns have ergative/absolutive morphology.
>
> Which, AFAIK, violates a universal that states that if there is
> a split on the referential hierarchy, it's the other way round,
> as in Dyirbal.
The difference between Dyirbal and Kazvarad is in the third person
pronouns; Dyirbal only uses nominative/accusative for the first and
second person pronouns, while Kazvarad also uses them for the third
person pronouns. Thomas E. Payne gives a convenient hierarchy chart for
this:

1 > 2 > 3 > 1 > 2 > 3 > proper names > humans > non-human > inanimates
agreement > pronouns                            animates

                              definite   >   indefinite

Things farther to the left on this diagram are more likely to be
nominative/accusative, while things farther to the right are more likely
to be ergative/absolutive. Kazvarad just happens to fit; I designed it
long before I'd heard of this universal.