THEORY: Construct state and/or genitive case in Semitic langs -- was: Re: THEORY: Morphosyntactic Alignment (again?), and Milewski

On 5/17/06, Eldin Raigmore <eldin_raigmore@...> wrote:
> On Tue, 16 May 2006 18:13:10 -0400, Jim Henry <jimhenry1973@...>
> wrote:
> >On 5/16/06, Eldin Raigmore <eldin_raigmore@...> wrote:
> >>On Mon, 15 May 2006 11:22:43 -0400, Jim Henry
> >>><jimhenry1973@...> wrote:
> >Of course this analysis wouldn't work for a language
> >that really has a construct state identical to a
> >particular oblique case and also has a distinct genitive,
> >or vice versa; but if there are any such natlangs
> >I don't know about them.  As far as I know
> >the construct state as such is attested only in
> >the Semitic language family (see below), and there
> >construct state is orthogonal to case.
>
> [er]
> I believe it was Akkadian; at any rate the very first Semitic language I
> looked at specifically for the purpose of investigating "construct state",
> was indeed analyzed has having _both_ a genitive _and_ a "construct state",
> by that author.  (I think if you look back at "Carsten's Birthday" in the
> archives, you may see where someone referenced that language.)
According to the Wikipedia article I cited, Arabic has a
genitive case as well as a construct state.  I'm not sure
about Hebrew, but surely one or more persons on this list
will know.  This article

http://en.wikipedia.org/wiki/Hebrew_grammar

is unclear on the genitive and/or construct state and I suspect
(based on its apparent inconsistency with what I've heard
elsewhere) that it may be inaccurate.


> >>Is "construct state" a "case", as it seems at the moment?  Or is it
> >>like "definite" and "indefinite", whatever they are?
>
> [JH]
> >I treated the contra-genitive like a case in Pliv-Rektek,
> >but according to this Wikipedia article,
> >
> >http://en.wikipedia.org/wiki/Status_constructus
> >
> >the construct state in Arabic and Hebrew is a
> >kind of definiteness marking, not a case.
>
> [er]
> I'd love to know how they know that.  I'd love to know how certain and how
> accurate they are about that.
That I don't know.  If for instance a noun can be marked as both nominative
and construct, or accusative and construct, but can't be
both construct and definite or both construct and indefinite, then I
would say that construct is a kind of definiteness marking
in a given language.  But I don't know if that is true of Arabic
and/or Hebrew.

For instance, if in a nonce-conlang definiteness is marked by
mutating the initial consonant and case is marked by various
suffixes, we might have

paf - a cat (absolutive)
baf - the cat (absolutive)
maf - the cat of (absolutive)
pafek - a cat (ergative)
kes - a woman (abs.)
tus - mouse (abs.)
sasu - to purr
tevu - to eat

then we might have:

sasu maf ges - the woman's cat purrs
tevu mafek gesek tus - the woman's cat eats a mouse

--
Jim Henry
http://www.pobox.com/~jimhenry