Status of "u" in Omaha-Ponca

[Koontz John E]

On Sun, 14 Aug 2005, ROOD DAVID S wrote:
> I am intrigued by this information, even though I probably knew it once,
> because the major difference between Wichita and Pawnee vowels is that
> Wichita merged [u] and [i] to /i/.  There are traces of the old *u in
> the morphophonemics.  Some rules that apply before /i/ don't apply
> before certain /i/s, namely, those that go back to *u.

This is sort of like *<barred i> in Eastern Eskimo.  The historic fourth
vowel that merges everywhere with *i, *a, or *u on the surface, but tends
to retain its own consistant pattern of morphophonemics.

> Wichita then had a three vowel system with no front/back distinction:
> i/e/a.  More recently they have evolved an /o:/ from VwV sequences, but
> I know of only one word with a short "o" in the whole language, and it's
> suspect, too, because it's /ho'os/, i.e. has a medial glottal stop and
> the two instances harmonize. I have no etymology for it, so it stands as
> a unique instance of that phoneme.

What does ho'os mean?

> I wonder about some sort of areal phenomenon now.  That sound change
> strikes me as kind of unusual. (I know about conditioned umlaut and
> unrounding, e.g. foot/feet, of course, but that's not an unconditioned
> merger.)

I think this sort of unmotivated rounding of *u is not entirely without
precedent.  For example, what about */u/ > /u"/ in Greek and French?  I
think in both cases this is essentially unmotivated and accompanied by a
(preceding? subsequent?) shift of */ou/ > /u/.  I also think, but am not
sure, that (Ancient) Greek /u"/ becomes Modern /i/, too.  However, these
are the only similar examples I know of.

Western American English does a sort of unmotivated unrounding and
centering of <lax u> (=> <barred u> or <barred i>), e.g., in good, book,
look, etc.  I don't know if this is similar enough to be considered a
parallel.