Loading ...
Sorry, an error occurred while loading the content.

Re: ablaut plus height harmony

Expand Messages
  • neo gu
    ... Hmm. Actually, I have no idea how the ablaut was originally conditioned. Or the vowel harmony. But such things occur. Speaking of derivation, I guess those
    Message 1 of 5 , Jun 14, 2013
      On Fri, 14 Jun 2013 21:43:38 -0400, Alex Fink <000024@...> wrote:

      >On Fri, 14 Jun 2013 14:08:57 -0400, neo gu <qiihoskeh@...> wrote:
      >>In my latest sketch, which is supposed to be naturalistic for a change, the protolanguage has 4 vowels, call them I, U, E, and A (or O). Most roots are CVCVC with each V specified only as being front or back. There may be a suffixed V, also front or back. This leaves 2 ablaut grades, high and low, which affect both the root and suffix V due to the height harmony. E.g. *xIwIm-U vs *xEwEm-A. For transitive verbs, the low grade is used for active forms and the high grade for passive ones. I'm not sure of the distribution for nouns, prepositions, and intransitive verbs. My question is, how naturalistic is this scheme? Also, is it too limiting?
      >I reckon it's naturalistic so far, but it would be easy to go overboard by making the height harmony too systematic, especially in other parts of speech.
      >If intransitive verbs aren't a totally different part of speech to transitive verbs, then which height class they have may well have systematic meaning, in line with the transitives. But for classes like nouns unrelated to the verb, the expected answer would be "some nouns have high vowels; others have low; there is no systematic meaning to which one a given noun has." There could of course be any number of further uses of the height contrast that make up small systems of their own -- inflectional, derivational of either high or low productivity, etc. -- one for each morphologically relevant place the pre-proto-language phoneme which conditioned the height change occurred.
      Hmm. Actually, I have no idea how the ablaut was originally conditioned. Or the vowel harmony. But such things occur.

      Speaking of derivation, I guess those suffixes should be affected by the height harmony. E.g. *xIwIm-tU vs *xEwEm-tA.

      >I think the most likely interpretation of the fact that there is a _single_ meaning that vowel height has for _all_ verbs is that the pre-proto-language had a prominent but less global pattern, and then a sweeping analogical change rewrote every form of every verb which didn't fit. If e.g. harmony also showed a single meaning everywhere in nouns, that would demand another such analogical change, and in semantics an essentially separate one, unless you can construe some way that nouns can be active or passive. And I don't think one such analogical perestroika in verbs makes a separate one any likelier in nouns.
      OK, then the distribution of high V words and low V nouns and prepositions should be more-or-less unpatterned.

      >On Fri, 14 Jun 2013 21:10:35 -0400, neo gu <qiihoskeh@...> wrote:
      >>On Fri, 14 Jun 2013 20:32:55 -0400, Anthony Miles <mamercus88@...> wrote:
      >>> How large is your consonant inventory? A low consonant inventory limits the name of roots. I like small
      >>> inventories and that's one of the things I have to live with. Depending on the syntax and the possibility of
      >>> compounding, this may not be a problem.
      >Eh, a small inventory is also okay if you accept root homonymy, which if you're being naturalistic you should. There are plenty of ways to work around inconveniences that root harmony may cause. One example is differing morphological behaviour in any other category: e.g. a low-vowel intransitive and a high-vowel intransitive that are phonological counterparts of each other could coexist with no problem, as potentially could two phonologically-identical roots which took aspectual inflection differently, or so forth. Or there could be other lexical material, e.g. adverbs or various classifier-like stuff or whatnot, which is (more or less) obligatory with one root or the other that tells them apart, or of course oneĀ of the roots could be restricted to fixed collocations. Etc.
      Root homonymy should appear eventually.

      >>Another issue I forgot to mention is whether the height harmony has to apply to case and aspect suffixes; currently it doesn't.
      >Well, if it doesn't, then either the harmony isn't purely phonological but rather morphophonological, or else the case and affix morphs aren't suffixes but rather some kind of enclitics or particles. Neither of which is actually a problem.
      The case markers at least may still be enclitic at the protolanguage stage; mostly the same markers are used for both singular and plural and across declensions. Possibly the noun plurals are enclitic as well. I'll have to take a close look at these and the aspects.

    Your message has been successfully submitted and would be delivered to recipients shortly.