Re: THEORY: Non-nom Subj & Nom Obj -- Quirky OVS Word Order Or Quirky Case?

Hello, Henrik, Markus, and everyone.
--- In conlang@yahoogroups.com, Markus Miekk-oja <m13kk0@H...> wrote:
> >From: Henrik Theiling <theiling@A...>
> >Reply-To: Constructed Languages List <CONLANG@l...>
> >To: CONLANG@l...
> >Subject: Re: THEORY: Non-nom Subj & Nom Obj -- Quirky OVS Word
Order Or
> >Quirky Case?
> >Date: Fri, 5 Aug 2005 02:03:45 +0200
> >
> >Hi!
> >
> >tomhchappell <tomhchappell@Y...> writes:
> > > Thanks for writing, Henrik.
> >
> >Pleasure!
> >
> > >...
> > > > IIRC, it was Markus who mentioned this a few days ago.
> > >
> > > I remember Markus's answer as being helpful, informative, ...
> >
> >Nono, he had a posting about different subject structures in German
> >and Icelandic and I asked what the difference was.  One thing he
> >mentioned was coordinated clauses (the other thing was reflexives).
> >But he said by his understanding, German did not allow non-
nominative
> >subjects.  I *currently* disagree, but there are so many strange
that
> >it is possible to convince me with good arguments.
>
> I didn't say it doesn't allow non-nominative subjects, only that
the
> non-nominative subjects have different features from the average
subjects
> (such as different features as far as reflexive binding and such
goes) , and
> that this needs to be accounted for. My proposed solution - which I
at the
> moment don't know how to test - is that the non-nom subjects are
embedded in
> the subject phrase in a way that the nominative subjects are not.
(I have
> been thinking of "agreement" a bit lately too, since I am of the
opinion
> that this is a form of agreement.)
>
> I have been away for a couple of days, and therefore unable to
respond. I'll
> read through the rest of the discussion promptly.
I just read
"Case and Grammatical Relations in English and Serbian" by Ivana
Vidakovic.
http://www.rceal.cam.ac.uk/Working%20Papers/ivana.pdf
She applies some of the subjecthood tests from Keenan E.L.
1976 "Toward a Universal Definition of 'Subject'" in Li, C. "Subject
and Topic" New York: Academic Press.
She uses those criteria which work on Serbian nominative subjects but
not Serbian accusative direct objects nor dative indirect objects,
and applies them to Serbian dative subjects and Serbian nominative
objects, especially when the dative subject and the nominative object
occur in the same clause.  She comes to the conclusion that the
dative subject is more of a subject than the nominative object is,
even though the verb agrees with the nominative object rather than
with the dateive subject.
----------
[PIVOT]
I was interested by the contributor who pointed out that, in ergative
languages, it is frequently the absolutive argument -- the one we
would ordinarily think of as 'the object' -- which can be elided from
a second clause if it is identical to that of the immediately
preceding clause.

In other words, in a situation where Jane fills a bucket then Jane
spills the bucket, in an accusative SVO language we can say

Jane-NOM fills bucket-ACC and [elided nom subj] spills bucket-ACC

but in an ergative SVO language we could say

Jane-ERG fills bucket-ABS and Jane-ERG spills [elided abs obj]

In the former case Jane-NOM is the pivot between the two conjoined
clauses; in the latter case bucket-ABS is the pivot.

In switch-reference systems, is there any need for "pivots" like this?

Especially, if it is possible for a verb to specify that it has the
same "subject", OR that it has the same "object", as the reference
verb?

Are their natlangs attesting such an arrangement? That is, where a
verb can have markings with meanings such as
verbstem-SAME.SUBJECT-SAME.OBJECT
verbstem-SAME.SUBJECT-DIFFERENT.OBJECT
verbstem-DIFFERENT.SUBJECT-SAME.OBJECT
verbstem-DIFFERENT.SUBJECT-DIFFERENT.OBJECT

(Of course, any one of these four markings might be a prefix instead
of a suffix, and they might come in any order, and they might be some
other kind of marking instead of affixes.)

Does anyone know?

Thanks,

Tom H.C. in MI