Re: ergative? I don't know...

Sally Caves wrote:

>         I thought I had understood this, but I guess I'm still
>         operating under a misconception about S.  The term "S"
>         HAS TO MEAN INTRANSITIVE SUBJECT, doesn't it?  Now I
>         finally understand where I have gone wrong in labeling
>         my terms for T.  Under the system that I now have, it
>         is incorrect to call the non-volitional subject an
>         "S," because this only really properly refers to
>         intransitive subjects.  I then lead everyone astray when
>         I apply the term to non-volitional transitives, as in
>                 Galleyli li zef oua--"the man (S) hears music."
>
>         What *I* mean by S is "participant," but a participant
>         is not necessarily an intransitive subject in Teonaht.
Yes.  The term S simply means "intransitive subject".
It does NOT refer to a type of noun case, and it does
NOT refer to a semantic role.  NOT NOT NOT!  :-)

How ergative, accusative, and active languages
differ from each other is in the way morphological
cases are mapped onto the three types of arguments,
A, S, P.  In accusative languages, A and S get
the same kind of marking, and P gets a different
kind of marking, while in ergative languages, P and
S get the same kind of marking, and A gets a
different kind of marking.  The *semantics* of
A, S, and P (vis a vis any particular verb) are
completely irrelevant in such systems.
A could be an Agent, an Experiencer, etc.
P could be a Patient, a Theme, etc.  S could
be just about anything.  That much depends on
the verb.

It's only in active languages where semantics seems
to play a role.  In active languages, A gets a
particular case marking (call it X) and P gets another
case marking (call it Y).  S may appear with either
X-marking or Y-marking, depending on whether
it's semantics are more typical of A arguments
or more typical of P arguments (according to some
language-specific criterion or criteria).  Suppose
the criterion is volitionality, as in Lakhota.  In such
a language, S arguments that are more volitional will get
X-marking, on analogy with A arguments, which
TEND TO be volitional (recall that A does not equal
"agent"; it is merely typical for A arguments to be
associated with the semantic role Agent).  S arguments
which are less volitional, on the other hand, will tend
to get Y-marking, on analogy with P arguments.

So, to summarise, there are three ontologically
distinct classes of objects here:

(1)  A, S, P
(2)  semantic roles (like Agent, Experiencer...)
(3)  case markings (like Nominative, Ergative...)

How these three map onto each other determines
whether you're dealing with an accusative, ergative,
or active language.

Matt.