Oral ~ Nasal Correspondences

[Koontz John E]

On Mon, 26 Feb 2001 BARudes at aol.com wrote:
> The situation with respect to nasality becomes much more complicated when the
> Catawban languages are brought into the picture.  Unlike the Siouan
> languages, there is no evidence to suggest that /m/ and /n/ are secondary
> developments from *w and *r before a nasal vowel.  Rather, *m, *n,*w and *r
> must be reconstructed as independent phonemes for Proto-Catawban,along with a
> contrast between long and short oral vowels and (inherently long) nasal
> vowels.

There are a few Siouan sets that make *n seem likely, too, like *ni... 'to
be in pain'.  This is a sort of messy set and I don't recall the details
off the top of my head.  This one just never seems to act like r + VN
unless the language is quite faithful to the r + VN => nVN pattern.  There
might well be ways around this, e.g., taking the first syllable as
prepronominals (a preverb), but as neither I nor anyone else has worked
out the details, and as most comparative Siouanists seem somewhat
uncomfortable with a presumption of *m and *n conditioned strictly by
vowel nasality, I'm taking a cautious approach.

> At present, based on the Catawba data and at least some of the irregular
> sets within Siouan, I lean toward the reverse of the usual explanation
> for the development of the nasal/oral distribution within Siouan, that
> is, that pre-Proto-Siouan (Proto-Siouan-Catawban) distinguish phonemes
> *m,*n, *r, *w and oral and nasal vowels, and that the nasalization of *r
> and *w to*n and *m before nasal vowels, and the denasalization of *m and
> *n to *w and *r before oral vowels was a Proto-Siouan innovation
> [something that got rendered as gibberish in my mailer JEK] an
> innovation that was not implemented in the Catawban languages until the
> mid-1800s.

This wouldn't be the only case of similar developments in different
Siouan(-Caddoan) languages separated by long periods of time.  For
example, the loss of *s^ 'second person agent' in *r-stems in Dakotan
(before contact) and in Omaha-Ponca (from the 1870s).

JEK