DPs and Demonstratives

[Koontz John E]

On Fri, 24 Feb 2006, Mirzayan Armik wrote:
> I should have stated in my email that the content had nothing to
> contribute directly to the Dhegiha discussion!

It's interesting seeing the data in parallel, though, and I don't think
anyone were confused.  It's interesting to see the Dakota data discussed
in terms of pitch accent, too.

> I do realize that the article system is different from Lakhota and that
> the dem-det order is impossible in Lakhota. I only thought the
> information might illuminate a way of looking at the Dhegiha
> pitch-accent in the same types of phrases, though maybe others have
> already done so.

I think that the whole issue of pitch accent at any level above the word
is an unexplored one.  Even within the word I'm not sure it's been well
explored in those languages where it has been adopted as an explanation.
I've never gone beyond a very impressionistic applicaiton in Omaha.

There are definitely limits to analogizing one language's syntax with
another, as has been pointed out, but one interesting consequence of
thinking of the Dhegiha definite articles historically as secondary
concordial elements co-occurring with a now missing definite marker
analogous to the Dakota definite article is that it tends to explain how
they can co-occur with both the noun and a preceding demonstrative.  If
they are gender markers that occur only with definite NPs, then they can
easily be extended to a preceding demonstrative that has a definite
reference, by analogy with what would happen with an independent
demonstrative, or with one following the NP.  In other words, the Dhegiha
articles replace definite articles functionally, but retain the syntactic
behavior of a predicate.

Indicating a definite element with X(def) and using gen for gender to mark
the location of the article, then given the patterns

dem(def)-gen
N(def) dem-gen

the development

dem N(def)-gen => dem-gen N(def)-gen

is natural, at least if the pattern dem N(def)-gen is at least potentially
analyzable as two NPs in a way that N(def) dem-gen is not.  I think this
would be just the sort of concordial marking that Catherine suggests.

I don't know if we'd want to call dem N constructions apposition.  It
seems to me more likely to be a focus construction.