Ablaut et al

[Rankin, Robert L]

Bruce, Paul, et al.

I believe the question whether or not Proto-Siouan had consonant-final stems is answered by looking at those same stems in the rest of the Siouan languages.  In each instance, virtually all of the other Siouan languages (besides Dakotan and Winnebago) have a stem-final -e, that is, where Dakota has ablauting -A, Winnebago has zero and Pat and various other Dakotanists posit CVC stems.  To me, this means that EITHER:

a) Proto-Siouan did have a final vowel, namely -e, and it was lost in Winnebago and replaced by -A in Dakota, OR:

b) Proto-Siouan had consonant-final, CVC, stems, and all of the other Siouan languages innovated a final -e in these stems INDEPENDENTLY.  

Given the pretty much unanimously agreed upon subgrouping of the Siouan language family, it seems to me that (b) is very unlikely.  I included lots of evidence for this in that paper I sent to several of you/us.  For example (I hope formatting holds here):

       make marks   ripe   	shallow 
PS 	*ká:xe      *aRú:te       *xé:pe     
CR 	-ka:xi    	  ó:ši   	xé:pi   
HI 	-ka:xe    	  ó:te   	xé:pi      
MA 	-kaáx                    		
LA  	 káγA	         lútA   	xépA    
CH  	 gá:γe   	dú:je   	xé:we    
WI  	 gá:x    	tú:č    	γé:p    
OP  	 gá:γe   	ní:de    	xébe      
KS  	 gá:γe   	ǰü:ǰe           	
OS  	 ká:γe   	cü:ce    	xé:pe    
QU	 ká:γe  	títte                          
BI	               atutí    	xépi    
OF	               atúti                           

You can see that these 3 verbs that "ablaut" in Dakota all show evidence of a stem-final -e (or a reflex of -e in CR and HI).  I personally don't see any way around reconstructing the *-e, given the subgrouping and generalized distribution of virtually identical vowels.  I might add that final short -e is vulnerable even today and tends to devoice in languages like Omaha following certain voiceless consonants, as in:  /mikhe/ 'I am the one who', which is often pronounced [mikhE] with a whispered e.  (I think John Koontz mentioned this to me; correct me if I'm wrong).  Anyway, the prevalence of an underlying final -e is evident in all these stems.

Now, whether there was a period during which Dakotan dialects lost final -e and actually had CVC stems, I don't know.  It may be possible, but whether or not this happened, Dakotan generalized an [-a] in these stems, and the impetus for this analogical change seems to have been suffixes that had a suffix-initial a-, such as -ape/-api 'plural', -as^ 'negative' and others.  As far as I can tell, Dakotan is the only language in the family that generalized the vowel /-a/ to this degree.  So, while "ablaut" is very real in Dakotan, it is, for the most part, phonologically conditioned in the rest of Siouan.  

There are isolated examples that muddy the waters like "ablauting" nouns, e.g., s^uNka 'dog' in Dakota, but this sort of thing is true of all analogical change.  Then there a few cases like ablauting nasal vowels, aN ~ iN, etc. where analogy REALLY went to town.  These don't exist outside of Dakota as far as I know.  

Anyhow, that's my story, and I'm sticking to it.  :-)  If this presentation is confusing, to anyone, I can send a copy of the original paper, as I already have to several.

Sorry, I can't contribute anything intelligent at all to the Semitic part of the discussion.  

Bob

________________________________________
From: Siouan Linguistics [SIOUAN at listserv.unl.edu] on behalf of Rankin, Robert L [rankin at KU.EDU]
Sent: Tuesday, August 30, 2011 12:42 PM
To: SIOUAN at listserv.unl.edu
Subject: Re: Ablaut et al

Bruce,

This is indeed an interesting topic.  There is a close correlation between Shaw's (and Carter's, etc.) Dakotan "consonant-final stems" and stems where the other Siouan languages have long vowels.  The rule seems to have been:  If the 1st syllable is long, it is accented; if it is short, accent the 2nd syllable.  Or, it could be phrased in terms of morae.  This begs the question whether or not Dakota had final vowels in the initial accent words.  I'm off this afternoon on a short trip up to Omaha and Council Bluffs and will return to this issue when I get back.

Best,

Bob

________________________________________
From: Siouan Linguistics [SIOUAN at listserv.unl.edu] on behalf of shokooh Ingham [shokoohbanou at YAHOO.CO.UK]
Sent: Tuesday, August 30, 2011 7:23 AM
To: SIOUAN at listserv.unl.edu
Subject: Ablaut et al

--- On Mon, 29/8/11, rankin at KU.EDU <rankin at KU.EDU> wrote:

Bob,
This is in reply to your earlier message , where you sent your article on Mississipi Valley Siouan "Ablaut".  Thanks very much for that.   I found it very interesting and noticed that you mentioned Shaw's work on Dakota phonology, which I read many years ago, there being a copy of it in the SOAS library.  One thing which interested me in Shaw was her explanation of the exceptional initial stress in certain disyllabic stems, káǧa  'to make' being one I think. She posits an earlier monosyllabic, final consonantal form for these stems such as kaǧ- . I have never seen this discussed much and wondered what other Siouanists thought about it.  It seems like a very neat analysis and parallels the argument of Greenberg about Semitic lexical stems which are now disyllabic in the majority such as katab 'to write'.  He suggests that Semitic stems were originally monosyllabic (in fact bisonsonantal) and that the second syllable (or the third consonant depending on
 how you look at it) is a later addition allowing for lexical expansion, an initial qat- 'cut' giving later qata', qataf, qatam, qasar and others all realtable to the idea of 'cutting'.  The other advantage is that it makes Semitic stems look more like Indo-European ones, which is attractive.
Without wishing to appear to be talking Nostratic, I do like the idea of original monosyllabic stems, but of course it does get into difficult ground as to how far back you think you can go.  I wonder whether it holds up in other Siouan languages.
Bruce