THEORY: Active case-marking natlangs

Hey. I don't know if there should be a THEORY tag in the subject
but since Tokana is still mentioned, I think I'll leave it out.

Marcus wrote:
> > You could call it by Dixon's term "Fluid-S". I think that active
languages
> > are a subset of Fluid-S langs, though I think people tend to use the
terms
> > as equivalents. Thus, Tokana, Georgian, Chickasaw, and Mohawk would all
be
> > Fluid-S, but only Chickasaw and Mohawk are active (the latter moreso
than
> > the former). As with any artificial division, I'm not exactly sure where
> > the line separating active from the rest should be drawn.
Matt wrote:
> I guess I have absolutely no idea what you mean by "active", then, and how
it
> differs from the kind of case-marking system found in Tokana.  Please
define!  Your
> claim that Old English is an active language makes no sense to me, I'm
afraid...
> As for Russian, other than the presence of some 'quirky case' subjects,
and the
> fact that animacy affects the object case marking of masculine gender
nouns, I fail
> to see how the Russian case-marking system is anything like Tokana.
This is the way I define "active alignment" in my BA-thesis:

"An active language is a language which organize its core grammar so that
the argument of some one-place predicates is marked like the A of a
two-place
predicate, while the argument of the other one-place predicates is marked
like the P of a two-place predicate (Blake 1994). Or put another way: Some
one-place predicates get their arguments marked like A, and others like P.
And sometimes some one-place predicates can get their arguments marked like
either A or P. Yet another way of putting it is that one can think of an
active language as pursuing a middle course compared to ergative and
accusative
languages; it marks some S like A and some like P."

Then I talk a bit about Dixon's "split-S" and "fluid-S" and why I can't use
them:

"Dixon claims that there is a major difference between split-S and fluid-S
languages. Split-S languages are just like ergative and accusative
languages,
having syntactically based marking of core constituents. The case marking or
agreement marking of a predicate is always done in the same way, no matter
what the semantics is in a particular instance.
         [ ... ]
Fluid-S languages have syntactically based marking for transitive verbs,
but semantically based marking for intransitive predicates, depending on
what
the semantics are in a particular instance of use. Hence, in a fully fluid-S
language, every predicate has the possibility of taking either A or P as its
marking.
         [ ... ]
There are two problems with this. First, the division of the intransitive
core role (S) being marked as either A or P has a semantic basis which is
similar regardless if it is variable ('fluid') or not. Second, there are
languages which employ semantic marking for transitive clauses as well."

As you can see, I have a very vague definition of "active". I think the
main thing that distinguishes active langs from ergative and accusative
ones is that there is some kind of semantic reason for marking the S
argument as either A or P. The most common ones being Control vs. non-
control and event vs. state.

I don't know if my view on things really matters in this discussion
(Marcus doesn't quite agree, having a narrower definition of "active"
and Matt having a very broad definition (as it seems) and Jörg basing
his definition more on animate vs. inanimate) but at least you know what
I think about active langs (at least for the moment, my view changes all
the time).

Comments?

daniel