Hochunk HO

[Koontz John E]

On Wed, 18 Feb 2004, Koontz John E wrote:

> On Wed, 18 Feb 2004, Rankin, Robert L wrote:
> > As a monosyllabic noun, /ho:/ 'voice, fish' is always long.  I don't
> > know its status in the compound.  As I recall from Miner, some of these
> > automatically lengthened monosyllables lose their length in larger
> > constructs.  Others may not.
>
> Actually, I had looked it up in Miner, if that helps.

One way this could be analyze this is that the historical compound is
*ho(o)'=thaNk, cf. OP hu(u)'=ttaNga. which, presumably is a calque or
adapted loan, not an inherited cognate, of course.  Assuming that initial
accented syllable lengthen or are long, and at least other formerly
accented initial syllables in Winnebago are regularly seen to be long,
this would yield (with length) *hoo'c^haNk and then (accent shift)
hooc^aN'k.  This would parallel forms with outer instrumentals, boo...'
forms, for example.  Of course, it is important to realize that this model
is pretty much conditioning independent.  Perhaps *hoo'thaNk is so
accented because long; perhaps it is long because so accented.  The best
way to determine this is to turn up forms with the same morphosyntax but
CV=CV'... accent (CV=CVCV'... in modern Winnebago).  There are actually
some candidates in OP among the animal + body part compounds.