random phonological observations.

[Rankin, Robert L]

> Re: Kathy's long vowels:
> I have gotten niNde for 'butt' and niide for 'ripe,' but I'll try
listening again.

For what it is worth, niide 'ripe' isn't historically nasal.  It's a
"funny *R" word, corresponding to Osage cu'ce and Lakota luta 'red'.  I
think Winnebago has duuc^.  The "funny R" words have a nasal reflex of
the consonant "funny *R" in OP, but nowhere else.

Yep, Kaw jü:je 'cooked, ripe' (Initial syll. accent).

What does come to me is that *pr also behaves as *R in nouns in OP,
e.g., in nu 'man', and nu 'potato' compare Lakota bloka', blo, and ne
'lake', compare Dakota ble.  And the *pr can precede a nasal vowel, as in
ni(N?)'fluid, water, major river' compare Dakota mni.

For the record, my analysis of these clusters is a little different from
John's. I identify the /b/ of ble, blo, bloka, etc. and the /m/ of mni as
reflexes of what we sometimes (probably inaccurately) call the "absolutive
prefix". For me that would have been *wa- with inanimate nouns and *wi- with
animate nouns. Therefore, for me, there is no such cluster as *pr. /bl/ and
/mn/ both go back to somewhat earlier *w-r (after vowel syncope), one in a
nasal and the other in a non-nasal environment.

In fact, I derive ALL [b]'s in Dakotan through earlier *w. Even the syllable
codas in b~m found in causatives, reduplications, etc. go through the /w/
stage as Dakotan allows only sonorants in syllable codas. This accounts for
the analogous reduplications in l~d, g~ng from ultimately underlying -t and
-k also.

P.S.  Sorry about typing vowel as vowl in the subject earlier.  See what
I mean about me and vowel length?

That would be the "zero grade", wouldn't it?

bob