Word for 'prairie'?

[Koontz John E]

On Fri, 30 Jan 2004, Rory M Larson wrote:
> > The texts have udhaN'da 'island' (JOD 90:436.8), udhaN'da 'middle of the
> > tribal circle' (JOD 90:601.17), niN'udhaN'da 'island' (JOD 91:39.2). I
> > haven't detected any other uses of the root dhaNda. [...]
>
> Yes, I'm sure that's right.  I looked up 'island' in the
> Stabler/Swetland dictionary and came up with
>
>   niu thoNta  (i.e., ni-udhoNda)

This is the Dhegiha 'island' set, it turns out.

Os  odhaNta 'center', niNo'dhaNta 'island'

LaFlesche refers from the latter to niN'pase which he glosses 'land out of
water'.  Rankin lists (from Dorsey) Ks niba'se 'island in the regular
channel' vs. ni(N)' oga'giye' ~ j^o'gagi'ye (j^e 'lake') 'island in an
oxbow'.

Qu  ni(N)' doNtta', ni(N)' naNtta' 'island'

The general form (where something else doesn't replace it) is *niN' +
odhoNta, where the latter seems to be '(be) in the center'.

The rest of Central Siouan seems to have *wit- 'island', with various
noun-forming suffixes, e.g., Da wi'ta, IO j^eromiNj^e (j^e 'lake'), Ma
wit-ka.

> I'm guessing that the -da is an old positional or
> nominalizer that occurs in conjunction with the
> leading u-:
>
>   u-dhaN-da
>   IN_CONTEXT-GLOB-THING
>
> which probably means a (globular) eminence standing out
> upon an otherwise level surface.

The more general sense of 'be round-shaped and located' isn't attested for
udhaN, but various derivates are:

udhaN' 'to hold, to take hold'
ugdhaN' 'to put in; in between'

I agree that -da is probably a postposition.  Compare dha'tta 'left
(handed)', Dha'tta(=)da 'Left Hand Side Clan', maybe originally 'to the
left'.  The -da form is cognate with Winnebago's locative (e=)j^a, Dakota
=ta.  The latter appears as e=k-ta with e=, and that form -k-ta accounts
for the more common Dhegiha form =tta 'to(ward)' (*kt > tt).

JEK