Siouan root constraints

[Koontz John E]

On Thu, 3 Feb 2005, Koontz John E wrote:
> Apart from this there are certain verbs with initial *p, *t, *k.  The
> standard Dhegiha set is something like:
>
> - *p-instrumental verbs
> - *taNpe 'see, look at'
> - *kaghe 'to make' (originally 'to make marks')
> - *kaNdha 'to donate' (not sure I have this form right by memory)
> - *kaNze 'to immitate, demonstrate'
> - *kaNyiNka 'not to want to'
> - *kaN=ra 'to want'

These were intended as sort of Omaha-Ponca influenced Proto-Dhegiha, but
Bob's post reminds me that at least *koNze, *koNyiNka, and *koN=ra are
probably not aN forms.  I suspect the first syllables of the last two are
etymologically connected.  *koNyiNka 'not to want to' looks like 'to
desire little to'.

> Nouns with initial T are also rare.  The one I remember easily is
>
> - Da paha 'hill'

At one point the extreme rarity of *T-initial nouns and demonstatives and
"outer preverbs" - things that precede all pronouns, not just
non-inclusive pronouns in verb inflection - led me to suggest that the *R
and *W phonemes of Proto-Mississippi Valley (and maybe Proto-Siouan) are
the "strict word initial" or maybe "proto-Siouan phrase initial" reflexes
of *t and *p.  There isn't a potential corresponding velar alternation.
At least *R and *W do occur initially in all of the forms listed and not
usually in other contexts.

One "within-Dhegiha" exception would be OP ama : Osage apa for 'the
(mostly animate plural proximate)', also used for what Dorsey called
'purposefully moving singular', though Ardis Eschenberg and Carolyn
Quintero have been arguing for more sophisticated analyses that amount to
"degrees of proximity."

The main thing to note with *T in Siouan is that, like voiced fricatives,
it is very limited in pre-tonic distribution.  Also, liked sonorants, but
unlike voiced fricatives, it is generally associated historically with
anomalous "lost vowel" patterns of behavior in regularly occurring
preceding elements, e.g., syncopated inflectional prefixes and
*ki-prefixes.

And the flip side of this is that *hT and *Tr- and voiceless fricatives
are generally limited to pretonic positions and associated with either
loss of a preceding syllabic element (often there is evidence of former
*wV- in various capacities) or with regular occurrence of preceding
elements ending in a retained (or perhaps epenthetic) vowel, e.g., regular
inflectional prefixes and *ki-prefixes.

One possibility is that *hT is a conditioned allophone of *T in an
environment like *#(CV)_V', which seems to be how aspiration works in Ofo.
However, it's not clear how that would interact with the presence/absence
of vowels in prefix patterns.  Perhaps the inflectional pattern was
created after the *hT : *T opposition was no longer productive.

Cases of *Th seem to result from contractions and cases of *C? might
involve realignment of glottalization, perhaps epenthetic glottal
transitions following some stems.