Our wish

[Koontz John E]

On Thu, 31 Jul 2003, Rory M Larson wrote:
> I'd like to clarify some of what you said:
>
> > whereas 'to think' is *DEMONSTRATIVE=ye.
>
> Did you mean =ye here, or =re?  I thought *r => Dh. [dh],
> [y], etc., while *y => Dh. [zh].

Yes, that's correct for *y in most contexts, but not intervocalically,
when epenthetic, or when in clusters like *py or *ky.  I went into this in
great detail - actually, in excruciating detail - in

http://listserv.linguistlist.org/cgi-bin/wa?A2=ind0306&L=siouan&P=R4795
http://listserv.linguistlist.org/cgi-bin/wa?A2=ind0306&L=siouan&P=R4852

> So across all of Siouan, this *ra stem is known only in
> Dhegihan *kuN=ra ?

I think it also occurs in Ioway-Otoe reflexes of the stem.  I'll check.

> How many distinct *k- stems do we have?  Could it be that when
> consonant clusters are reduced, *pku- => *kku-, while *pka- => *ppa- ?
> I.e., high, back vowels like velar stops, while other vowels prefer
> labial stops?

Well, appart from gaNdha < *kuNra and gaghe < *kaghe there's gaNze 'to
show' < *kuNze, which has A1 ppaNze.  Also, gaNziNga 'not to know how to'
has ppaNz^iNga.  I don't know of cognates outside Dhegiha, but I suspect
the intial element there might be *kuN as in gaNdha, in which case the
construction is perhaps literally 'to little desire to'.  So it looks like
the vowel and maybe even the identity of the morpheme are irrelevant.  I
have pondered this a bit, but I haven't come up with anything.

Other k-stems:

gi (< *ku) 'to come back'

(Teton may have phu < *hpu (?) for the first person of this in an isolated
and obsolete idiom involving someone seeing someone they haven't seen for
a while and saying s^ku, whereupon that person replies with phu.  I think
this is the only attested s^ in the second person in Dakotan.  Allan
Taylor pointed this out to me.  It's in Buechel's dictionary.)

gadha (gadhae?) 'to donate'

This might be the lot.  I don't recall any more examples off hand.  Of
course, essentially the only other g-initial verbs are those in ga- 'by
striking, by action of current', which has its own unique paradigm.
Verbs in b/d/g are rare, if you count the cases of stem-initial
instrumentals b/d/g as a single example.  That is, count all examples of
ba-stems as one form ba-, all examples of bi-stems as one form bi-, and
all examples of ga-stems as one example.  All cases of instrumentals in
b/d/g reduce to three examples, because there are only three instrumentals
that do this, if I haven't forgotten one.  This statement is phrased for
OP, but applies with appropriate modifications in other Siouan languages,
too.

While you're at it, consider Mandan pke 'turtle' vs. Dakotan khe(ya),
Dhegiha kke, Winnebago kee(=ra).  This might be a *pk > *hk initial set,
too, and behaves like the *kuN in *kuNra.

JEK