BL accent patterns. The real results.

[Rankin, Robert L.]

> There is no "schwa" there phonologically.  That's confusing phonology with phonetics.  So of course the GL or BL syllable gets accent if there's another prefix.  But what I'm saying is that the CL syllable gets the accent anyway.


> First, I think the *BL cluster is underlyingly a syncopation of the first syllable of *wv-rv in Proto-Siouan, correct?  So *wv-rv => *w-rv => *BLv ?

Yes.  That's standard Siouan historical phonology that everybody agrees on.

> Next, I assume Proto-Siouan had second syllable accent.  Then *wv-rv-cv is accented *wv-rV-cv.  This syncopates to *w-rV-cv => *BLV-cv, with accent on the vowel immediately following L, aka *r, where it always was.  Here, I think we’re on the same page, and you’ve shown good evidence from Dhegiha to support this.

That's unless the initial syllable V is long, in which case it gets the accent.  But that's not applicable in these instances because if the initial syllable had been long it wouldn't have undergone syncope.

> Now what happens when we take the same sequence and add a preceding syllable: *cv-wv-rv-cv?  By second syllable accent, the vowel between *w and *r gets the accent:
                *cv-wV-rv-cv

No, the *wa- or animate *wi- prefixes only really occurred word-initially as far as I know,(or as far as any of us has analyzed, I think).  So your scenario maybe didn't really occur.

> If the second syllable with the accent is syncopated out: *cv-wV-rv-cv => *cv-w-rv-cv => *cv-BLv-cv, I don’t see how the “BL syllable” can get the accent, because there is no vowel there to accent.  I can think of just three possibilities here:

> 1.       Accent shifts to the morphophonological third syllable: *cv-wV-rv-cv => *cv-w-rV-cv => *cv-BLV-cv

Can only happen under the Winnebago accent shift.  Other Siouan languages don't seem to allow it with the possible exception of Dorsey's really screwed up accentual patterns in Biloxi.  In any event, there are no know cases.



> 2.       Accent shifts to the first syllable: *cv-wV-rv-cv => *cV-w-rv-cv => *cV-BLv-cv

If the 1st syllable was long it always had the accent (without having to posit a shift).



> 3.       This can’t happen, because if the intervening syllable is accented, syncopation does not occur.

Yes, I think that's it.  You're right out on the cutting edge of reconstructing proto-Siouan phonology here, and I don't think we have any cases of your vulnerable syllable structure to try to explain.  As far as anyone has gotten there weren't instances of derivational *wa/wi that had prefixes.  Locative prefixes, *ii-/aa-/o- were originally postpositions on preceding NPs that were later reanalyzed as verb prefixes (as I recall Randy has instances of them still functioning that way in Crow.)

1st singular *wa + /r/ clusters are different in that they are supported by active morphophonemic alternations all along, but they behave the same way until the patient and locative enclitics became prefixes creating sequences such as you describe above.  By that time the BL 1st sing. clusters were established and didn't revert to /wa-r/ sequences.

The GL clusters have a more complex history since they interact with all the damned KI morphemes and generally (always??) arise through inflection rather than derivation.  I don't have time to even think about than can of worms.

> Do you have a sense of which of these three possibilities should happen in the *cv-wV-rv-cv case?  Or if the outcome is something else that I haven’t been able to think of, can you describe what that is?

If you can't think of it, I'm sure I haven't.  I think the only thing you have left out of the discussion of this is the complexity of inflection vs. derivation, and that's better left for the next generation of historical Siouanists, if there is one.  Suffice it to say that there ARE additional problems with initial syllable unaccented short vowel syncope. It only seems to have happened with *wV- syllables.

Bob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/siouan/attachments/20130911/59294c26/attachment.html>