BL accent patterns. The real results.

[Rankin, Robert L.]

> From the Omaha side, I could probably help with the GL reflexes, as they stay GL in Omaha and Ponka.

Actually, it's not a problem in Osage and Kaw.  Initial /l/ simply subs for original /gl/.  The real problem with initial /gl/ is that it is mostly inflectional, primarily possessive, and, as such, does not undergo the phonological restructuring that /bl/ does.  In other words, /gl/ is supported by active morphophonemic alternations in all the languages.  That doesn't happen with /bl/ or its nasal counterparts except in inflected 1st person sg. verb forms.  I'm interested primarily in the cases of real restructuring where there is no support from phonological alternation.

I’m not sure I totally understand what we’re looking for here, though.

I’m saying that, because initial BL (along with nasalized bn-, mn-) is all that is left of an initial disyllable, *wvlv́ (where v is any vowel), we can’t expect to find many modern lexemes with the accentual pattern blvCv́.  This is because, in the older disyllable, accent would have become stranded on the initial syllable vowel after the first vowel in the word underwent syncope, i.e., dropped out.  To find the pattern blvCv́ would imply a proto-Siouan accent pattern cvcvcv́ unless we posit massive accent shift.

> I thought the idea was that *bl and *gl represent primordial syllables, so that if they are word-initial the accent should be on the vowel immediately following them, i.e. the underlying second syllable rather than the third.

Primordial DIsyllables.  And, yes, you're exactly right about the "underlying second syllable rather than the third."

> Anyway, here’s a list of *gl  (gr-) initial words from the dictionary I’ve been working on.  Most have the accent on the following vowel, but two of them, ‘across’ and ‘hawk’ have it on the “third” syllable.  Discounting variants of the same root and two that have no further syllables, I’d say there are about 9 or 10 that take the accent immediately following initial *gl.

 Thanks for looking at the *GL set.  I concentrated on the BL set in 4 languages and found almost exactly what I said last evening.  If you have a chance to look at the BDH pattern in Omaha, it would be very welcome.  Here are the results from the other languages:

Kansa  Of approx. 22 lexemes in initial bl- all accent the initial syllable except for reduplicanda, which always accent the 2nd duplicated syllable.

Osage  Of 10 lexemes with initial br- all 10 accent the initial syllable.  (Quintero).

Quapaw approx. 18 lexemes in bd-/bn-.  Only one lexeme, bdasé ‘shout, cry out’, plus a few reduplicanda, accent the 2nd syllable.

Dakota  Of approx. 71 lexemes in bl-/mn- all but 16 accent the initial syllable.  Of those 16, 6 are reduplicanda, leaving only 10 out of 71 with 2nd syllable accent.  5 are examples of incorporated mni ‘water’.  Apparently incorporanda are unaccented or, at least, accent the second syllable, as in Willem’s study. My data are from Buechel because it's the only computerized Dakota dictionary I have.  I'm sure Jan's dictionary would be an improvement.

So I stand by my original statement, and it works for the most part except for some restructuring (but not much) in Lakota.  Data provided on request.

Bob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/siouan/attachments/20130910/4ce1f2ae/attachment.html>