Re: CHAT: Pima determiners

On Wed, 22 Nov 2000, dirk elzinga wrote:

> Cool. Okay; fellow U of A student Colleen Fitzgerald (now at SUNY
> Buffalo) wrote a little paper ("Prosody Drives the Syntax") in which
> she claimed that the Tohono O'odham determiner _g_ was deleted when
> sentence initial for prosodic reasons--something like "can't begin a
> sentence with a stressless element." In other words, prosody takes
> precedence (in an OT sort of way) over the syntactic requirements of
> noun marking by a determiner.
That certainly makes sense, as the determiner is always unstressed.
However, in Pima, the form is _heg_ (following normal Uto-Aztecan
orthography patterns. This actually isn't what I find wierd. The
determiner is optional in post-verbal situations. For example, I have
received both of the following sentences.

Hascu aak am ha-nolav Melissa heg komkjidh?
Hascu aak am ha-nolav heg Melissa heg komkjidh?
'Why did Melissa buy a tortoise?'

I see not syntactic or prosodic reasons for the presence/absence of _heg_
in these examples. When pressed, he prefers the forms with _heg_
present.  Perhaps there is some subtle difference in topic/focus.
I just don't know, but it is a point I'm going to keep my eye on as I work
on the more interesting portions of the grammar.

> Akimel O'odham (aka Pima) may be different but the pattern you
> described sounds about the same. Now if you're talking about
> syntactic/semantic behavior, I can't help you.
I am a syntactician. So the phonology is almost irrelevant to me. (Sorry.)
On the other hand, I really like the problems with reduplication; but that
is practically morphology. :)

Marcus