Theiling Online    Sitemap    Conlang Mailing List HQ   

TERMS: dominant/recessive harmony and umlaut

From:dirk elzinga <dirk.elzinga@...>
Date:Wednesday, November 17, 1999, 17:00
On Tue, 16 Nov 1999, FFlores wrote:

> Raymond Brown <ray.brown@...> wrote: > > > > No, not at all. It is the vowel in the _root_ word which conditions all > > the vowels in the affixes (usually suffixes) in langs with vowel harmony. > > > > Umlaut is almost the opposite; the vowel of a suffix has modified the root > > vowel. The suffix then, of course, usually thinks it's done its job & > > disappears. > > I thought you could call them simply forward or backward vowel harmony, > or something like that (like there is forward and backward assimilation). > But I guess the main difference is that vowel harmony involves the root > (while umlaut may modify non-root vowels too, can't it?).
In dominant/recessive harmony systems, the harmonic trigger may be either a root or an affix; what matters is its ranking in a hierarchy of dominant and recessive morphemes. Dominant morphemes impose harmonic requirments on recessive morphemes regardless of root/affix status. Nez Perce is such a language (as is Chukchi; but I know the facts better from Nez Perce). In Nez Perce the harmonic sets are: recessive: {i,e,u} dominant: {i,a,o} (Notice that [i] is contained in both sets; a nice wrinkle in Nez Perce that I got a qualifying paper out of in grad school.) root-controlled harmony (1s poss prefix: ne?-, vocative -e?) mex 'paternal uncle' ne?mex 'my paternal uncle' mexe? 'Uncle!' to:t 'father' na?to:t 'my father' to:ta? 'Father!' In these examples, the root governs the quality of vowel in the affix. Affix-controlled harmony ce:qet 'raspberry' ca:qat'ayn 'for the raspberry' tisqe? 'skunk' tisqa?laykin 'near the skunk' In these examples, the affix governs the quality of vowel in the root. In a dominant/recessive harmony system, the direction of harmony is controlled by the dominant/recessive hierarchy and not by directionality. Dirk -- Dirk Elzinga