From:    Carsten Becker <[log in to unmask]>
> > I would have to second this.  Fluid-S languages have
> > gradient marking depending on how volitional, animate,
> > etc., the argument is.  Thus, an intransitive like "fall"
> > would normally take O-marking since you don't normally
> > choose to fall, but if you did, you could use A-marking.
> > My understanding is that prototypical trigger languages
> > have no such split-behavior among intransitive
> > predicates.
> OK, OK, so I misunderstood the term. But what about one
> daughter lang developing from Trigger to Ergative and
> another one from Trigger to Accusative? It should be
> possible, shouldn't it?

I'm by no means an expert on trigger languages, but I don't
see any obvious problem with this.  One word of warning:
most, if not all, languages have mixed systems.  Even a language like
English has some manifestations of ergativity.  The -ee suffix,
borrowed from French but now independent, can be used for any
intransitives (on buses in Chicago one can e.g. see signs
refering to the duties of 'standees') or the patient roles of
transitive verbs (as in employee).  So, you may want to consider
keeping vestiges of the trigger system around.  (Ask me about
Mingrelian sometime, which is a really neat example of vestiges
like this.)

> The first is called "absolutive" IIRC (with
> the absolutive being called "subject"?!) and the second is
> called "nominative" (with the nominative object being
> called "subject").

There are two issues here:  what we call the cases, and how that
case morphology patterns in the syntax.  As you probably know,
many morphologically ergative languages, like Basque, have accusative
syntax, and so one cannot properly speaking call the absolutive
case a subject unless you show syntactic criteria that it behaves
as such.

> Would the Ergative lang make more use of
> causatives as well? (Causative in Ayeri = someone/-thing
> caused someone/-thing to do something)

AFAIK causativization has nothing to do with the subjective
alignment system a language has.  One possible exception might
be split-S systems, given that in some languages a so-called
unergative intransitive cannot be causativized, while so-called
unaccusative verbs can be.  The implication is that in a split-S
language, A-marking intransitives may not be able to be causativized
while P-marking intransitives can be.  But this doesn't apply to
ergative languages as such, but to languages in general.

> I mean like in the
> example I gave, "to invent" -> "being invented", where
> "being invented" is "invent.CAU".

This sounds more like a passive to me than a causative.

Thomas Wier	       "I find it useful to meet my subjects personally,
Dept. of Linguistics    because our secret police don't get it right
University of Chicago   half the time." -- octogenarian Sheikh Zayed of
1010 E. 59th Street     Abu Dhabi, to a French reporter.
Chicago, IL 60637