butterfly

[R. Rankin]

> What's of special interest, of course, in the
present contect, is that
> element nic^a in the Santee form.  I'm pretty
sure that's nic^a, not
> nic^ha, and I suspect it's a contextual variant
of *yiNka 'little'.  I
> think that's regularly c^hiNc^a' 'child' in
Dakotan.  I suspect that in
> contexts that allow *y to be intervocalic, it
can be rhoticized and the
> resulting *riNka would appear as ni(N)c^a in
Dakotan.
> This element could also be nic^a < *riNke 'to
lack', but I don't see how
> that would work.

I'd have to say that neither works for me.
Dhegiha doesn't seem to share in any of these
varieties of 'little' with /n-/ as far as I can
tell.  Doublets /z^iNka/ 'little' and /z^iN/
'diminutive' are common, but I can't recall any at
all with /n/.

> Winnebago, which merges *r and *y regularly, has
niNk as a diminutive, as
> in wake'(niNk) 'raccoon', s^uNuNgniNk 'puppy',
and so on.

But not in the South.

> Of course, whether or not nic^a is 'little',
it's clearly a good match for
> the Quapaw nikka.  It would be a regular match
if Quapaw had nika.

I think that is entirely possible, especially
given Osage /hkihtanika/ with the single /k/,
written <g> by LF.  It's a bit strange, given that
Dorsey didn't often confuse the lax with the other
series of stops (very often confusing the
aspirated and tense ones), but he does it on
occasion, even in Kansa.  Just last week I was
reading one of the historical texts aloud for the
language program and ran across several instances
of /ga:yo/ 'then', written <kayu>.  So I suspect
you're right and that Quapaw has /-nika/.

BTW, Biloxi has /tiNskana'/ 'jay', with simple
/tiNska/ meaning 'back of the neck'.  But Ofo is
/teska itho:hi/ 'bluejay' = 'bird' + 'blue'.  And
tiNska/teska may somehow be related, given the
Dakotan and other forms.

> Presumably something like this explains ppiza,
too.

Naw, that was just my mistake for 'sand', which
has /pp/.  My memory is fading....

Dorsey's "philosophy" of transcription was
slightly different from ours.  I explained it in
my (yes, unpublished) paper on Biloxi stops.  But
it was similar in that he tended to use unmodified
letters for sounds he considered the least
"marked", and, for him, those were the sounds most
like English.  So unmodified ptk were used for
aspirates (which they mostly are in English).  The
ones he wrote with diacritics were those he
considered most remarkable, i.e., least like
English.  He wrote them with the little x beneath
(LaFlesche's dot) and they were printed upside
down.  The bottom line is that the diacritic is
always used on the most lenis voiceless series.
Unfortunately this means that the languages with
voiced lax stops (Omaha, Ponca and Kaw) use the
subscript x for voiceless tense stops, while
Quapaw and Osage, which lack a voiced series, use
the x for the voiceless lax stops.  Confused?
Well, anyway, I think John is right that Quapaw
probably has /nika/ in 'jay'.

Bob