butterfly

[Koontz John E]

I definitely agree with Bob Rankin and Henning Garvin's notion that
Winnebago miNmiNke looks like an Algonquian loan.  Miner gives it as
miN'iNmiN'iNke.

On Tue, 28 Oct 2003, Kathleen Shea wrote:
> Oops!  I'm still mixing Net Siouan with the practical Ponca orthography.
> 'Hummingbird' is /wac^c^i'ninikka waz^iN'ga/.  I know that no one thinks
> there's an aspirated voiced dental fricative in Ponca!  However, the word
> for 'butterfly,' /wac^c^i'ninikka/, compared to /wac^hi'gaag^e/ 'to dance,'
> does exemplify the phonemic contrast between /c^c^/ and /c^h/ in Ponca.

Swetland lists wati'ninika (1991:30), i.e., wattininikka, without the
diminutive affrication.  I think the source here is Fletcher & LaFlesche
(1972:106) wati'nini ka, a rendition that confirms kk.  There's a tendency
to hear geminates as word boundaries that appears in both the Dorsey texts
and in Swetland's own dictionary work.  I think explains why the future is
always written a separate word, thoug this might also be the influence of
Riggs's usage on Dorsey.  Dorsey's ms grammar began as an adaptation of
the Riggs grammar and often follows it word for word, with OP morphemes
substituted for Santee ones.

The LaFlesche Osage dictionary lists "dsi-oN' dsi-oN" 'butterfly'
(1932:239b) and dsiN-tha t.oN-ga 'butterfly (large)' (idem), along with
dsi-oN'-dsi-oN wa-p.o-ga 'owl butterfly' (idem).  These forms are given
again (but with a dash between the two repetitions inthe first) in the
Osage side.  I would summarize these in something more like Osage
phonetics (revising LaFlesche's use of OP-based spelling) as tsiaNtsiaN
(maybe spelled ciaNciaN) as Bob almost does.  He omits the reduplication -
fairly, I think - in comparing this with Quapaw ti'na, but I've mentioned
the precise forms in case somebody notices and wonders what happened.

I suspect that ciaN(ciaN) is already one step from ciNdha toward Kaw giaN.
That's just the Kaw spelling of Dhegiha forms *kiaN 'to fly', but in the
context of 'butterfly' it represents a degree of folk etymologically based
revision of presumed earlier *tiNdha (or *tiNra, depending on the point at
which we introduce PMV *r > Dh dh ~ ...).  I suspect it's nasalized,
because "dsiNtha" ciNdha in Osage is, and because Quapaw has ti'na, and
not ti'da.

The question is, of course, whether the Omaha-Ponca form is in any way
connected.  The -tti- part isn't, of course, because tt < *ht, not *t.
However, *R merges with *t in Osage, Kaw, and Quapaw, and *R becomes n in
OP.  So the PDh form might be *RiNdha, potentially yielding *nidha in OP.
In this context, the ...ninikka part might be relevant.  -nikka is
'person, man' in compounds.  I don't think it occurs alone in OP.  The
closest thing to watti... is a nonce form watti'dha 'rank of warriors' in
Dorsey.  I can't make anything of all this except that the ...ninikka part
looks a lot like Winnebago miNiNmiNiNke form.  The -kka matches -ke
regularly, but the reduplicated part has ni(N)ni(N) instead of miNmiN.  I
still can't explain watti... If it were wati... it might be some sort of
compound *tiNdha (not *RiNdha) and niNinNkka, with the latter perhaps
reformulated from *miNiNkka.

I looked at Ioway-Otoe to see if it might be helpful, but it's rupa'n~i ~
ropa'n~i, i.e., something like ropa'niN (n is enye before i).  This seems
to be a compound involving 'Pawnee', but if ro- (or ru-) is ro 'body', it
seems to be in the wrong place for 'Pawnee body'.  I looked for *miNmiNhka
or *RiNdha like forms without any luck (might expect *j^iNna for the
latter) and also checked 'moth' and 'owl', the two likely places for such
forms to wander off to.  'Owl' is maNkoke, makoge, which looks like maN +
'box'.  MaN looks helpful (see below), but it's probably 'earth', since
the owl in question is the burrowing owl.


If the Winnebago form is an Algonquian loan, we have to wonder about the
Dakotan forms, too.  Williamson gives Santee ki'mama, kimimana and Teton
kimimala.  Riggs has Santee ki'mama and ki'mamana, with Teton kimimela and
gmimela.  Among Teton sources Buechel has kimi'mila, and so in this case
does not follow Riggs; Ingham agrees, giving kimi'mila.  The -na and -la
are, of course, diminutive, and we expect -na in Santee after a nasal
vowel, and we expect -A > e before =la in Teton, so presumably the
underlying form is something like original *kimima with considerable local
revision, though original kimimi or kimama seem about as likely!  In any
event, the core part seems to be mVNmVN.

JEK