DPs: I got it backwards

[ROOD DAVID S]

Hi, Rory,
	It's flattering to be taken so seriously.  I'll try to respond to
your comments below.

David S. Rood
Dept. of Linguistics
Univ. of Colorado
295 UCB
Boulder, CO 80309-0295
USA
rood at colorado.edu

On Mon, 20 Feb 2006, Rory M Larson wrote:

> David,
>
> If I am following you correctly, you propose two grammatical categories of
> functional morphemes that qualify or constrain noun phrases in Lakhota.
> Category 1 is the set of 'specificity' markers (he/le/ka), which have also
> been called "demonstratives".  Category 2 is the set of 'definiteness'
> markers (ki/waN/cha/eya/etaN, etc.), some of which  may also be called
> "articles".

	I tend to call all of the items in Category 2 "articles". For
those who don't know the Lakota details, ki is definited and everything
else is indefinite; the choice depends on the number and count/mass
category of the noun and the realis/irrealis context of the NP; "cha" is
an idefinite article used only on clauses.


>
> To make a sentence conveying the information "The boy who saw the horses
> told us about them", three assertions are implied:
>
> 1. There was a certain boy;
>
> 2. He saw the horses;
>
> 3. He told us about them.
>
> In Lakhota, this works out to something like:
>
>   Hoks^i'la waN he   s^uN'kawakHaN' wic^a[saw]  ki  [he.told.us].
>   Boy       a   that horses         he.saw.them the  he.told.us.
>
> (My Lakhota, of course, is rusty!  Please correct!)
>

	The is exactly right. 'Saw them' would be waNwichayaNke.  If the
sentence were 'A boy who saw the horses told us about them', the "ki"
would be replaced by "cha".  So the article at the end of the clause
indicates the definiteness status of the head of the clause.


> The article/definiteness-marker (waN, ki) ties tightly to the end of the
> preceding material to wrap it up into a noun phrase.
>
>   [[Hoks^i'la] waN]
>   [A [boy]]
>
>   [[Hoks^i'la waN he   s^uN'kawakHaN' wic^a[saw] ]  ki]
>   [The [boy who saw the horses]]
>
> Its order is fixed.
>
	Yes.  Perhaps a little more literally, you could try the
paraphrase:

	[the [a certain boy saw the horses]]


> The demonstrative/specificity-marker (he) is more loose in where it
> appears.  It can function either as a noun modifier (like Japanese sono) or
> as a representative of the noun phrase itself (Japanese sore).  In this
> respect, it is bi-functional, like the English word 'that'.  ("Did you see
> that horse?" vs. "Did you see that?")

	This is true, but not in the relative clause construction just
cited as an example.  So in simple noun phrases you could say

	He hoks^ila ki 'that boy'

or 	Hoks^ila ki he 'that boy' or 'the boy, he'

and it's also possible to say just "hoks^ila he" for 'that boy'.  I don't
know the functionarl difference between using or omittingthe "ki", but you
cannot omit it if you start with "He".

>
> In the construction above, the function of the demonstrative he is
> arguable.  There are two possibilities:
>
> 1. It modifies the preceding NP as "that (newly introduced) boy".
>
> 2. It is a stand-alone noun representative placed in apposition to the
> preceding NP as "A boy, he saw the horses..."

Actually, if by "the construction above" you mean the relative clause
construction, I don't think either of these is true.  I would argue that
the "he" functions to mark the 'boy' as particular or specific, although
as yet unidentified or indefinited (signalled by waN) IN THIS
CONSTRUCTION, NOT IN GENERAL. In the relative clause construction,
both the waN and the he modify the preceding N; the pronoun reading
seems to me to be excluded.  But I'm not sure about this.
	This is where I get the idea that "ki/waN"  and "he/le" indicate
definiteness and specificity separately; I try to reflect "specific" by
using the English word 'certain".  The two choices you just gave would be
available to regular NPs of the form N ki he.
>
> >            There are also a lot of cases where the sequence N+ki+dem
> looks as
> > if the "dem" were really some kind of resumptive pronoun, kind of like
> > "the boy, he told us about it".  I have no solid evidence for that
> > intuition, however.
>
> That would fit with possibility 2.  I have the same intuition, both about
> Lakhota he and OP e.  I think the "resumptive pronoun" construction may be
> a common grammatical feature in these languages.
>
> Is the above discussion a fair paraphrase of your argument?

	Very much so, and very clearly described.


>
> Best,
> Rory
>