accusative and ergative languages

[Jens Elmegaard Rasmussen]

On Fri, 16 Jul 1999, [in continuation of a discussion with Par Ryan]
Larry Trask wrote:

> [...]
> But briefly: if a transitive sentence in an ergative language were
> "really" a passive, then its absolutive NP would be the subject and
> would exhibit subject properties.  But this [is] not the case in most of
> the ergative languages I have heard about.
[...]
> The "passive" view of ergative languages in general is indefensible.

I guess some key words are "really" and "split". If ergative syntax is
allowed to be limited in a language and may still be called ergative, why
can't the passive interpetation of it be allowed a similar latitude and
still be called passive? Is it not possible that some syntactic structures
that were originally passive (with all the trimmings that word entails)
_lost_ their passive value and simply became the normal way of saying
things (like a joke that gets repeated and isn't even meant to be funny
anymore) - and that, subsequent to this change, reflexive reference and
all other subject properties were transferred to the new pragmatic
subject? It does not make the analysis pretty, but if the linguistic
history consists of a period in which people changed their minds about the
basic roles of ordinary sentences, should an accout of the truth not be as
ugly as what really happened?

I do not know of many ergative-structured languages; but all I have seen
can be safely or highly probably analyzed as being passive circumlocutions
in origin, but none of them is completely so in terms of synchronic syntax
anymore. But that does not change their prehistory as passives. Is A
asking about origins, and B answering about descriptive synchrony?

Jens