Lakota chaNnuNpa

[Koontz John E]

On Mon, 26 Jun 2006, Rory M Larson wrote:
>          tobacco                pipe
> La.      chaNli                 chaNnuNpa
> Os.      noNni'                 noNni'oNba
> Om.      nini'                  nini'ba
>
> MVS      *RaNri'                *RaNri'-uNpa

This is the pattern Bob was talking about.

The tobacco set is totally irregular.

La c^haNli' => *yaNRi'
Os noNniN'  => *raNriN'
OP niNniN'  => *riNriN'

The OP form is probably a revision of something like the Osage form and
not a particular problem, but the Winnebago and Ioway forms don't match
either Lakota or Osage and it just gets worse further afield.  It turns
out that 'tobacco' and/or 'kinnikinnik' (however spelled) are "widely
borrowed" forms in North America, though I don't know of a reference for
that.  (If there isn't one maybe there should be.  Any willing students of
loan words?)  So these terms resemble each other as 'coffee' and 'tea'
terms resemble each other elsewhere and are not reconstructable for
Proto-Siouan, even though PS probably and PMVS certainly postdate use of
tobacco.

There are a few cases of *y in Dakotan matching *r in Dhegiha among the
body part terms, e.g., La c^haNte' 'heart' vs. OP naN'de.  My favorite
explanation is that the initial here in OP is an epenthetic *r, and that
Pre-Proto-Dakotan has *yaNte < *i-aNt-e, while Pre-Proto-Dhegiha preferred
*i-r-aNt-e.  Perhaps 'tobacco' was inalienably possessed, but that still
leaves the medial consonant and nasalization.

This is a set that you can keep coming back to over and over!  It doesn't
hurt to invite specialists from other language families, too.