Re: What would you call this?

Quoting Doug Dee <AmateurLinguist@...>:

> >> So, the inflection swaps the positions of the subject and object? Sounds
> >> like inverse voice.
>
> >No, inverse "voice", as traditionally used in Algonkian studies
> >at any rate, is not a voice at all, but the only way to express
> >certain kinds of grammatical relations.  In Algonkian languages,
> >this hierarchy operates something like as follows:
> >
> >   1, 2 > 3 proximate animate > 3 obviative animate > 3 inanimate
> >
> >In such languages, the *only* way to say that a 3 obv, say, is
> >acting on a first, second, or third prox is to use the inverse
> >verb marking.  Otherwise, a direct marker shows that the actants
> >are behaving according to (rather than contrary to) the way the
> >hierarchy predicts.
>
> Hmm.  It seems to me that what we're discussing here is similiar
> enough to a direct/inverse system that I'd call it a variety of
> direct/inverse.
Right.  My argument was not that this is a direct/inverse
system (aka hierarchical), which it is, but that this direct/inverse
system is not a kind of voice distinction. Rather, it runs parallel
to nominative/accusative, ergative/absolutive, Split-S, and Fluid-S
systems. If it were a voice, you'd expect it to be optional, which
it is not, and a variety of other phenomena which Amy Dahlstrom
has described in her grammar of Meskwaki (alas it is still in manuscript
form).

(In fact, as I've said before in another post, I am speaking
specifically of Meskwaki here.  Though other Algonkian languages
are similar, some, like Ojibwe, have changed the inverse markers
to a kind of passive.)

> That is (correct me if I go wrong here), in Algonkian, in a
> sentence with two 3rd person NPs, a direct inflection on the
> verb means "the proximate NP is acting on the obviative NP",
> and an inverse inflection means "the obviative NP is acting
> on the proximate NP."  The proposal here is very similar:
> the "direct" inflection (if I may call it that) means "the
> core NP mentioned first is acting on the core NP mentioned
> second" while the "inverse" inflection means "the core NP
> mentioned second is acting on the core NP mentioned first."
Kinda, except that in Meskwaki, word order is so free that it
not clear whether the more basic word order should be VSO or
VOS.  Word order has essentially no role in determining who
did what to whom.  I think you need to change "mentioned first"
to "discourse-functionally chosen" (see below).

> Or, to put it another way, what we're discussing is a direct/
> inverse system in which the proximate role is always assigned
> to the NP that comes first is the clause.
No, actually that's not the case.  The choice of proximate or
obviative is clearly a function of the discourse, that is, it
is the choice of the speaker simply to divide 3p animate
arguments into two categories for purposes of reference tracking,
not the grammar as such*. (Though it is true that within any clause
with two 3p animate arguments, one has to be proximate and the
other obviative). Once the choice has been made, the choice
may, but need not, be changed but very easily, from one clause to
the next. (Some languages with prox/obv distinctions require the
distinction to be held over a particular stretch of discourse.)
That is to say, the choice of proximate and obviative is autonomous
from, but interacts with, the direct/inverse system.

*(There is one exception to this. Possessed NPs must inherently
be obviative:  neniw-a o-o:s-ani, man-an.prox.sg 3-father-an.obv.sg
"the man's father")

> (This would mean there's no need for morphological marking on
> NPs to distinguish obviative from proximate, as there generally is in
> Algonkian, if I'm not mistaken.)
If proximate and obviative were assigned as you say, I believe
you would be correct. But they aren't.  In fact, though, it is true
that in the case of the so-called further obviative (which divides
two arguments previously marked as obviative into more and less
salient) the only marking occurs on the verb, not on the noun arguments:

 e:hici     "He (prox) addressed him (obv-1)"
 e:hikoci   "He (obv-1) addressed him (prox)"
 e:hikonici "He (obv-2) addressed him (obv-1)"

Here, an explicit argument "man" for both obv-1 and obv-2 would
be _neniwani_, as opposed to _neniwa_ for proximate.

 =========================================================================
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