Ok, I'm going to jump in here, though with some trepidation since I haven't actually read a lot of the messages in this thread. If I wait till I find time to read them carefully (much less go back and check any data) I'll never get to it. Where DO you guys get the time?
Anyhow, I can't resist rising to the bait of "strange article use" in Omaha and the notion of article copying/definiteness agreement vs. apposition.
As some of you know, I once did a paper on Omaha constructions like "Ga-akHa nu-akHa" (this-the man-the) or "wa'u-thiNkHe she-thinkHe" (woman-the that-the), with a demonstrative and a noun, each with matching definite article suffixed. These are fairly common, both in the speech of my consultants and in Dorsey. Extremely common in the speech of a couple of constultants, who happened to be the most fluent, even Omaha-dominant speakers I recorded, so I don't think it's a semi-speaker effect of any kind. My memory is that the DEM-art N-art order is considerably more common than the reverse. Occasionally there are more than 2 parts to the construction, i.e. DEM-art N-art N-art. The obvious analysis of these is that they are appositive constructions (possibly a hesitation phenomenon, buying time with a DEM while searching for a more specific word??) and this looks particularly likely when one of the parts is a proper name ("She-akHa nuzhiNga-akHa Bill-akHa", i.e. "that boy, Bill"). In my paper I tried to make a case for analyzing them instead as defniteness agreement, with article (optionally) spreading to all parts of a complex DP. At this point I can't remember what the arguments were, but in any case it's not an entirely straightforward decision in either direction. If people are interested I could dig it out and try to give some sort of synopsis.
Going back to an earlier part of the thread: For nominal phrases with both demonstrative and article (or deictic & specificity marker or whatever they are), a 2-tier X-bar with the determiner (aka article aka specificity element) highest seems right to me. That is, something like John's tree below. (I've cleaned up the spacing so it lines up right on my screen... hopefully didn't mess it up for everyone else.)
DP
!
D'
/ !
NP D
/ !
DEM N'
!
N
This would, as John said, nicely allow for [DEM N] and [N DEM] orders while keeping the article last; it also works nicely for allowing the "specificity" feature to spread across the whole phrase in case that turns out to be the right analysis of the Omaha doubled-article constructions above. :-)
Catherine
>>> rankin at ku.edu 2/23/2006 3:12 PM >>>
I'm aware of article copyingon Arabic adjectives, but I don't think there is anything like that operating inSiouan. It would at least have to be demonstrated. Modern Omaha doeshave some strange article use, but it's different from the 19th century varietyin this respect. The common pattern as far as I can tell is DEM N-DET, sothe other pattern needs a special translation if it's going to be correctlyrendered.
Bob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/siouan/attachments/20060223/e17f8b9f/attachment.html>