DPs and Demonstratives

[John Boyle]

Thanks Bryan.  I’m not really sure that demonstrative is the right name for these 
things to either. Given what Jan said about Lakota being able to have both:

1) Demonstrative Noun Determiner

2) Noun Demonstrative Determiner

we may want to assume your right and the structure is:

	 DP
	   !
	   D’
	 /   !
       NP   D
       / !
 DEM  N’
           !
           N

Where DEM can either proceed or follow N.  Thus, it would c-command N but 
not D. Would this give us a better analysis?  Jan, what is the difference between 
the word orders in (1) and (2)?

Thanks,

John Boyle

---- Original message ----
>Date: Mon, 20 Feb 2006 13:15:59 -0600
>From: "Bryan Gordon" <linguista at gmail.com>  
>Subject: Re: DPs and Demonstratives  
>To: siouan at lists.colorado.edu
>
>   Personally, I'm sceptical of the designation
>   "demonstrative" in general. It
>   seems to be used to ascribe both deictic capacity
>   and the syntactic
>   behaviour of determiners under the same category -
>   which is clearly NOT
>   appropriate for Siouan. Is it possible that Siouan
>   deictics don't c-command
>   D at all, but are contained within NP?
>
>   - Bryan Gordon
>
>   On 2/20/06, jpboyle at uchicago.edu
>   <jpboyle at uchicago.edu > wrote:
>
>     Hi All,
>
>     I am just looking at noun phrases that have both a
>     determiner and a
>     demonstrative.  I was wondering if anyone else has
>     looked at these besides
>     Randy and Catherine (who should of course feel
>     obligated to reply to this e-mail
>     anyway).  In Missouri Valley the structure is:
>
>     Demonstrative Noun-Determiner
>
>     I think this is true for other Siouan languages as
>     well, correct? Has anyone
>     thought about how to analyze these constructions
>     (specifically in an X'bar
>     framework)?  Are they DPs that have a
>     demonstrative phrase (DemP) in SPEC and
>     an NP complement (as in 1)?
>
>     1) [[Dem P [NP ]]DP]
>
>                 DP
>                 / !
>       DemP   D'
>                 /  !
>              NP  D
>
>     This would make it all left branching, which is
>     what we would assume, right?
>
>     Or are they DemPs that take a DP Complement that
>     then take an NP complement
>     (as in 2)?
>
>     2 [Dem P [[NP DP]]]
>
>               DemP
>                   !
>               Dem'
>                 /  !
>          Dem  DP
>                     !
>                     D'
>                    / !
>                 NP D
>
>     This structure would be both right branching and
>     left branching (possible but
>     not as pretty).  Is there any evidence for either
>     analysis?
>
>     Thanks
>
>     John Boyle