Ablaut et al.

[Rankin, Robert L]

> Yes, I think we have ended up with this sort of disagreement before.  From my point of view, you tend to confuse phonology with orthography.  Your approach is more letter-oriented, which leads to EITHER...OR dichotomies like the one that started this discussion.  

I don't think phonologists "confuse" phonology with spelling, but, you're right, what you seem to be talking about has been known as "the phonemic principle" for over a century.  Precise definitions differ, but it IS segment-based for the most part, and a given segment is either present or not in the surface string.  There's no "in between".  There is a general agreement that a phoneme, whether underlying or superficial, is composed of a set of distinctive features that can, themselves, change, thus altering entire sets of phonemes.  And a given feature or phoneme doesn’t have to contrast with all the other phonemes of the language in every single environment.  

> I am more inclined to pronounce the words to myself, paying attention to the mechanics of how they are made and how these mechanics would evolve from one stage to another.  This leads me to consider intermediate possibilities between the EITHER and the OR.  We both are interested in the evolution of these languages, but it seems to me that your view of phonological evolution is somewhat more punctuational than mine.

I can see that, although I can’t sympathize with the technique as a primary way of learning in a literate society where there are centuries of genuine past scholarship.  You must have a very low opinion of past generations of linguists.  While I commend the attempt by a highly intelligent person such as yourself to invent linguistics/ phonology ex-novo, I've found that it's generally more useful to take the literature seriously and see what other intelligent scholars have found first.  I've heard good things about the recent Linguistic Institute at CU from Justin, and I attended one back in '97 myself to get "caught up" with certain aspects of the discipline.  I highly recommend the experience.  

> I believe it is active verbs of CVCv type that we have been discussing here, as a tangent from your post on ablaut that raised the question of possible CVC roots in that context.  In a previous message, I forgot to include "active verb" in the specification list, and was kicking myself shortly after pressing the "Send" button.  

Gee, that's never happened to me.  :-D    LOL

> My question is: Do we have monomorphemic active verb roots of CVCv type in Proto-Siouan where v is a vowel other than -e?  If so, is there a significant number of them, and what are some examples?

OK, But you realize that you are evidently positing morphemic status for this -e of active verb stems.  In effect it would mean "I am an active verb stem." just as -r in Spanish is a morpheme that says "I am the nominalized form of an active verb stem."  (i.e., an "infinitive").  

As we've agreed before, -e might be a morpheme.  JEK has argued for it; I've argued against it.  I'm still not satisfied with either argument from a historical point of view.  The matter can never be settled from a synchronic point of view.  One counter argument is that a morpheme -e should also mark monosyllabic active verbs, but it doesn’t.  There IS a different –e that is found affixed to verbs, and that’s the causative morpheme.  If the “ablauting” –e is really a morpheme, I’d expect it to be suffixed with an epenthetic glide, -r-, like the causative, but that doesn’t happen.  

> We’re talking about phonology primarily, not morphology.  The syllable structure ought to be uniform.  And, in fact, it demonstrably is.  

That’s really all I’m saying.  Languages have syllable canons.  The cognate sets show that proto-Siouan had an open syllable structure and that all your CVC words in fact had the final vowel.  Any differences between this surface structure generalization and underlying structure have to be motivated.  You can’t just say there are underlying CVC syllables without proper motivation, and you don’t have that except in Dakota and maybe Winnebago.  How many times and ways do I have to say it?  The cognate sets militate against it and the overall syllable canon militates against it.  

> So if the idea that mąąðį means 'earth-move' is comical, exactly how do you relate it to Catawba mą ‘go’?  

One of the salient characteristics of Siouan languages is the compounding of motion verbs.  Best treatment is Allan Taylor’s IJAL paper about ’73 or ’74.  

>  . . . how do you know that the original etymology was not, in fact, 'earth-move', which was used for 'walk' or 'go' in both languages?  It is a bimorphemic word, after all, because the second part of it conjugates separately.  Or was the original Siouan-Catawban word for 'go' mąą?  In that case the Siouan word apparently works out to 'go-move' instead of 'earth-move'.  (Reading your commentary in the CSD, I gather you take the latter view.)

Catawba data do suggest that *wąą was an older motion verb.  Why you would have to say ‘earth-move’ for ‘walk’ if there is no other kind of movement for contrast?  There’s no ‘water move’, people can’t fly and there were no horses or wheels.

In Omaha, ttą́ąðį conjugates as a unit as it does in Kaw.  In the CSD, you reconstruct the Proto-Siouan form as *wa-htą́he, and suggest that the Dhegihan form reanalyzed the final -he as ðį, 'be in motion', by analogy with mąąðį.  So this word too, though an active CVCv form (unless -he is a separate morpheme), turns out to end in -e in Proto-Siouan.

But not in Dhegiha languages.  Ðį is unaccented and is an exception to your CVCv rule, such as it is.  But you bring up a genuinely interesting question, or actually set of questions.  There are several proto-Siouan verb suffixes that need a lot of work.  And they show that several of your “CVC” roots are actually CV.  The second C belongs to a proto suffix or enclitic when it’s an /r/ or /h/.

If you go through all the verbs in the CSD you find a lot with the suffixes *-re, *-he, and *-ų or –ą.  These occur with such frequency that they must have been morphemes.  In an OV language like Siouan they must have had some sort of auxiliary status.  Nobody has attempted to explain these, but someday it will surely be profitable to do so, and it could help explain some of the phonological structures we’re discussing.

-Re is one common proto-Siouan verb suffix.  (Dick Carter believes that there’s an element of epenthesis in *-re.  He wrote this up for Mandan at one of the Siouan Conferences.)  I believe it is definitely a morpheme related to one of the notions we translate as ‘be’.  It’s also commonly found in Catawba.  No one has tried to determine if it is ‘be of existence’, ‘be of class membership’, ‘locative be’, etc.  But it must have had a function.  

The suffix or enclitic *-he found at the end of many verb reconstructions may or may not be related to the auxiliary –he found with virtually all the Dhegiha positional auxiliaries:  k-hé, ðįk-hé, athą́-he, t-hé, aðį-he, etc.,( the defective verb conjugated only in the 2nd person).  It is certainly prominent in the CSD verb reconstructions.  In Dhegiha it’s meaning seems to be related to ‘locative be’.

The other apparent proto-Siouan auxiliary is apparently derived from *-ʔųˑ ‘be, do’.  There is clearly a conjugated auxiliary: m-ų́, ž-ų́, ʔų, seemingly ‘did’ or ‘was’, but there is also an invariant version with the shape –ų or an apparent reduced *-ą that is just the final phoneme of many verb stems.  There is also an invariant enclitic with the usual epenthetic –r- to separate vowels:  *rą = [ną] or [na].  It functions to mark ‘anterior mode’ in most languages and is post-verbal.  It’s found throughout Dhegiha, in Biloxi and, I think, in Mandan and Chiwere/Hochunk.

Many of you will have noticed by now that at least two of these affixes or enclitics, -re and –he also turn up prominently as demonstrative particles, ‘this’ or ‘that’.  This may be coincidence, but it may not be.  Somebody, sometime needs to attack these data and see what shakes out.  

> At this point, we still have no Proto-Siouan monomorphemic active verb roots of CVCv form, where v is other than -e, on the table to serve as evidence that the -e at the end of CVC(e) active verb roots was contrastive, and therefore phonemic.

We don’t need “monomorphemic active verb roots of CVCv form…”  That’s another mistaken view of phonology.  Many phonemes have a low functional load in particular contexts.  They need only contrast in other environments or in other parts of speech.  There are lots of nouns, particles, adverbs and active verbs that end in all kinds of vowels.  You just want to ignore them.  You seem to want to think that different parts of speech (grammatical categories) (or even SUB-categories!) have different phonological inventories.  That may possibly be true in some creoles, but it isn’t true in Siouan.