Indian perfume set.

[Rory M Larson]

John wrote:
> Another point, which the Dhegihanists are aware of,
> but maybe not others, is that OP *ppi 'good' is
> completely obsolete.  In fact, I think *ppi is
> obsolete throughout Dhegiha.  It exists in a few fossil
> forms, e.g., ppi'=az^i ~ ppez^i' 'bad < good + not'.
> The usual OP term for 'good' is u(u)daN.

This is a good point, and it might help to explain
the outcome in OP of the first syllable.  For a
Dhegihan speaker, the first syllable *hpi=u- would
be unanalyzable because they no longer had the word
*ppi.  But since the second syllable *braN was clear,
there would be some motivation to reanalyze the first
syllable into a familiar morpheme.


>  The Osage cognate of this - o(o)taN - means essentially
> 'be a coup'.

"Be a coup"??  Do you mean as in 'counting coup'?
I don't understand this translation.


>  Anyway, if nubdhaN < *p(p)y=obdhaN then it isn't
> likely to be a recent form.  This is also evident
> in the necessity that any such form has undergone
> the *py > *pr > *R changes, with *R subsequently
> becoming n in OP by regular sound changes.
> Corresponding Osage, Kaw, and Quapaw forms would
> be *tobraN, *doblaN, and, I think, *topdaN or *dobdaN -
> I forget how the *R and *pr sounds come out in Quapaw
> at the moment!  I don't believe there are any cases
> of n ~ bdh alternations in OP at present.  That is,
> there is no perceptible contemporary connection
> between the two.

If nubdhaN is coming from MVS (Hypothesis 1), then
it certainly wouldn't be recent.  If it is coming
as a loan from a prehistoric IOM *pyobraN or *pyubraN
(Hypothesis 2), then the most likely time for it
to happen would be at the *pr stage of the above
sequence (assuming OP nu < *pro, but not < *pyo).
Since all the Dhegihan languages have single
consonant sounds where the *pr should be, I would
suppose that *pr had gone to *R before Dhegihan
diverged.  That would favor putting the borrowing
after proto-MVS but before proto-Dhegihan (Hypothesis 2a).
On the other hand, the circumstantial considerations that
this set apparently exists only in IOM and OP, and
that there are traditions of OP and IO associating
with each other after the divergence of OP, and that the
IO template pyu- is a significantly modified contraction
of the hypothetical proto-MVS *hpi=o-, make the
idea of a later transference tempting (Hypothesis 2b).

For Hypothesis 2, these considerations might be
resolved if we imagine proto-Dhegihan existing as
a dialect field over a wide area for a period of
several centuries.  During this period, *pr > *R,
and thence toward its reflexes in the various
Dhegihan daughter languages, which are still
contiguous dialects.  During the same period,
proto-IOM lives in a neighborhood adjacent to the
pre-proto-OP part of Dhegihan, and maintains
especially close relations with these people.
Early in the proto-Dhegihan phase, pre-proto-OP
adopts and readapts the proto-IOM word for mint,
IOM *pyubraN > pre-proto-OP *prubraN, where
*pru is reinterpreted as 'potato'.  Later
in the phase, this evolves to *RubraN and still
later to *nubdhaN.  Finally, in the course of
some crisis, the proto-OP people move away from
the other Dhegihan groups, accompanied by their
IO associates.

Otherwise, we always have Hypothesis 3, which
holds that the first syllables of IOM pyubraN
and OP nubdhaN are only coincidentally similar!

Rory