Re: THEORY: vowel harmony [was CHAT: Another NatLang i like]

At 2:27 pm +0200 28/6/99, Kristian Jensen wrote:
>Matt Pearson wrote:
>-----<snip>-----
>>
>>When I originally invoked the notion of nasal spread, I meant the
>>spread of nasalisation from a nasal consonant to the surrounding
>>vowels.  In Mixtec, for example, when a word begins with a nasal
>>consonant, all of the following vowels in the word will be nasalised,
>>unless there's a non-nasal, non-glottal consonant somewhere in the
>>word, in which case that consonant will block further spreading.
>>(We have a less interesting kind of nasal spread in English, where
>>a vowel becomes nasalised if followed by a syllable-final nasal
>>consonant.)
>>
>>What I was asking for (with a certain dubious tone in my voice) was
>>a language with a *phonemic* contrast between oral and nasal vowels,
>>in which all of the vowels in a word must be either oral or nasal.
>>In such a language, /katima/ and /ka~ti~ma~/ would be possible words,
>>but /kati~ma/, /ka~tima~/, /katima~/, etc., would not be possible
>>words.  That, as I see it, would be an example of nasal vowel harmony,
>>and would be quite a different phenomenon from nasal spread, which
>>involves the interaction of a nasal consonant and an adjacent vowel
>>(or vowels).
>
>OK... ummm... (here we go again), wouldn't Desano be just such a
>language? You probably missed the samples I gave of Desano words, but
>_all_ the vowels of a Desano word must be either oral or nasal - you
>can't mix both. (I can repost the samples if need be). OTOH,
>nasalization also nasalizes voiced consonants, while voiceless ones
>are unaffected.
Yes, indeed, the Desano example is more akin to vowel harmony than to the
nasal spread as described by Matt since IIRC even tho the voiceless
consonants are not affected they do not block the nasalizalation of
following vowels and, where affected, consonants.

I had come across something like the Desano examples before (but can't lay
my hands on the notes at the moment) so thought that that was what Matt was
referring to when he mentioned nasal spread.  I now see I was mistaken.

Thanks, Matt, for clearing this up & sorry, Kristian, for muddying the
waters.  It seems that the Desano examples are vowel harmony but with a
plus, since certains consonants must harmonize also.  Wonder if there's a
term for that  :)
-------------------------------------------------------------
At 12:22 am -0500 28/6/99, Nik Taylor wrote:
>"Raymond A. Brown" wrote:
>> But a thought occurred to me somewhen round about 5:00 this morning: the
>> modifications of the consonants in "k=F8begyn" are clearly all allophonic
>> modifications - there is no substitution of one phoneme by another or a
>> modification so great that a consonant merges with that of another
>> consonant phoneme .......
>
>Perhaps I'm missing something, but in a language with, say, front-back
>harmony, what would make /u/ and /y/ separate phonemes, at least any
>more so that the modifications you stated.
I think you must be.

In Turkish, e.g. many suffixes have a vowel harmony whereby the vowel in
the suffix may be /u/, /y/, /|/ or /u/ (where /|/ is the high, back
unrounded vowel written as an undotted i in Turkish).  But the vowels
certainly have separate phonemic status in the language.

Ray.