A Small Discussion of Aspirated Fricatives (Re: OP: coming and going)

[Koontz John E]

If the Dakotanists (especially those with a knowledge of Stoney or
Assiniboine) will stick with me (or skip to the bottom) I have a question
for them.  In fact, to give it here, out of context:  what is the second
person of 'to say'?

I'll also give the moral here.  No matter what the answer is to the
proceeding question, the moral is, "If any personal form of a Siouan verb
is irregular, elicit all personal forms for it, and in the case of a
dictionary, note all forms (either explicitly or with a paradigm code)
for all verbs."  This harks back to Jimm's sad experience with 'to wound',
of course!  Somebody, sooner or later, will regret any non-compliance in
this area.  Even the most innocent of verbs may have a single irregular
form, e.g., OP i'...bahaN 'to know' has idhappahaN 'I know' instead of
expected ippahaN.  Everything else is regular, though.

On Mon, 22 May 2006, Rory M Larson wrote:

Given forms like s^ + hi => s^i, ...

> The question of aspirating a fricative might deserve a small discussion
> in itself!  Here, it looks like it makes no difference to a fricative
> whether the leading vowel of the following morpheme is logically "rough"
> or "smooth".

Well, let's see ... Swanton represents Ofo as having something that he
represents as aspirated fricatives.

But in the contexts of Dhegiha, to some extent it seems to me that the
difference between "voiceless" and "voiced" fricatives in Dhegiha
(Omaha-Ponca, anyway), is more of a difference between "bright" or
"clear" fricatives and "dull" or "muffled" ones, which might be
characterized also as aspirated vs. unaspirated.  This observation
has always been tempered by a suspicion that I was wandering off the
beaten track in an area, phonetics, where I knew just enough to
suspect this sounded silly.

You've probably noticed that Dorsey specifically marks "s" and "s^" before
n as "turned," or "unaspirated."  I think he uses turned letters for the
voiced fricatives in Osage.  I wondered about this, but, sure enough, when
I heard Omaha spoken, words like snede sounded like znede.  (I have a
tendency to recall this word as snede', but I think this must be a case
where it is easy to hear pitch accent (sneHdeL) incorrectly as final
stress accent (sne 'de).)

To revert to the case of the second persons of h-stems, s^ there is
definitely one of the "bright" ones.  The same morpheme comes out z^
before ?-stems, e.g., az^iN 'you suspect' (a=...(?)iN 'to suspect, to
infer'), which I think behave as vowel-initial in the first and second
persons of Dhegiha.  So maybe it's *s^hi vs. *a=s^iN, leading to s^i vs.
az^iN in the usual spelling.

However, I am a bit reluctant to think of cases like xidha 'eagle' and
ghage 'to cry' as xhidha vs. xage.  I guess this would make waxe
'whiteman' out to be waxhe.  And nuxe 'ice' would be nuxhe.  But gaghe
'make' becomes gaxe.  And isn't it gaxa 'branch', so gaxha?

Here's something for the Dakotanists:  at one point I noticed that Dakota
has this inflectional paradigm for 'say', compared with the OP one:

          Da                  OP
A1        ephe                ehe (but egiphe in the dative)
A2        ehe                 es^e
A3        eye                 a=i

The third person stem might be *e=...(r)E (*r > y in Dakota), though it
seems to be just unepenthesized (e=)E in OP (with ablaut before =i).  If
there's no epenthetic glide between two vowels, the regular behavior is to
lose the first, so perhaps e=...E always comes out e or a in OP.  I think
cases of e are rare.  I don't recall one, and they are hard to search
for!

But, getting to the point, the first and second person stems are clearly
*e=...E.  What I noticed was that the Dakota second person form form has h
(with loss of s^), while the OP form has s^ (with loss of h).  Then it
occurred to me that the middle demonstrative in Dakotan is he, while in
Dhegiha it's s^e.  At that point I suggested that the middle demonstrative
was PS (or PMV) *s^he.

David and Bob, however, pointed out that Assiniboine and Stoney have s^e
instead of he.  That was disappointing, though interesting, too.
However, I think that at the time I forgot to ask what the second person
of 'to say' was in Assiniboine or Stoney!  I suppose it's still ehe?
This becomes an interesting question because eya 'to say' is the only
"functioning" h-stem in Dakotan.  The others have all lost their h's and
gone regular.