obviation in Siouan languages

[Koontz John E]

On Tue, 12 Jun 2007, Jan Ullrich wrote:
> I do not know what the etymology of the word is, but it is probably 
> relevant to mention here that it occurs in two variants: with aspirated 
> kh (chaNkhe) and with unaspirated k (chaNke). ... (distribution of 
> forms)

Bob Rankin has pointed out in various other contexts that alternations 
between -ke/a and -khe/a in verb finals usually have two different 
sources.

One pattern is *CV-ka => CVka vs. *CVh-ka => CVkha.  This explains the 
doublet suffixes of similar meaning in Da -ka ~ -kha, OP -ga ~ -kka, IO 
-ge ~ -khe, Wi -k ~ -ke.  A slightly different pattern occurs with 
*CVC-ka, e.g., Da -ka, OP -ka, IO -ke, Wi -ke.

The other pattern, which I think applies here is that many positional 
verbs have an inflected auxiliary *he attached to them.  In Dhegiha this 
appears in the first person as *he and in the second person as *s^e.  It 
is usually absent in the third person.  Elsewhere it tends to lead to 
doublets in -kA vs. -khA, etc.