5

Most languages with vowel harmony seem to have two main divisions of vowels plus a neutral division.

The main division is along one vowel dimension such as frontness/backness or +/- ATR (advanced tongue root). These are the divisions I'm referring to with "strong" in scare quotes.

The neutral category is a different type because neutral vowels can mix with vowels of any "strong" category, so I'm not concerned with these.

I'm interested to know if there exist any languages where each non-neutral vowel in a word must be a member of one and the same category out of three.

I'd also be interested to hear about languages with more than three "strong categories".

4

A language with a triple vowel harmony distinction is Manchu. In this language there are three "strong" vowels, a, e, o and two "weak" ones, i, u. The vowel harmony in Manchu is a bit tricky, in a word a "strong" vowel can be followed by the same "strong" one or by any of the "weak" ones, but a "weak" vowel can be followed by a "weak" vowel or by a, but never by o. That is, three sets of vowels are possible within a word:

  1. a, i, u
  2. e, i, u
  3. o, i, u

The 3rd row can also be like that:

o, o, ..., i/u ..., a, i, u.

That is, the 3rd row can change into the 1st one, so we can say the vowel harmony in Manchu has a stair-step character.

According to that vowel harmony, suffixes in Manchu are of the two kinds:

  • some have a "weak" vowel and they never change, since they can be added to roots that have any vowels: -mbi - present tense, -fi - participle: "arafi" (having written), "genefi" (having gone), "songofi" (having wept), "arabumbi" (make write), "genebumbi" (make go);
  • others have three variants, with each of the strong vowels: -sa/-se/-so - plural noun suffix, -ha/-he/-ho - past participle: "araha" (wrote), "genehe" (went), "songoho" (wept).

The change of the 3rd row into the 1st one can be illustrated by "songobuha" (made weep), where the u in the causative suffix -bu changes the 3rd o row into the 1st a row, so -ha, not -ho is added.

В. Аврорин, Грамматика маньчжурского письменного языка, СПб, "Наука", 2000

6
  • I thought there might also be "weird" cases as I was writing the question but didn't mention them since I expected answers would come in about them if there any (-: – hippietrail Dec 10 '13 at 10:51
  • @hippietrail - So do you consider Manchu a "weird" case? As for me, I do. ;) – Yellow Sky Dec 10 '13 at 10:53
  • 2
    I do indeed, and I'm very glad we have them! (-: – hippietrail Dec 10 '13 at 10:53
  • 2
    @hippietrail - Great! I adore all the "weird" cases. – Yellow Sky Dec 10 '13 at 10:55
  • I would've gotten bored with linguistics after my third language if not for them. I'll pop into the chatroom now if you like ... – hippietrail Dec 10 '13 at 10:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.