Talk:Abkhaz phonology

Geminates in the Bzyp Sadz dialect
Are there any consonants which occur only ungeminated in the Bzyp Sadz dialect of Abkhazian? --88.76.251.11 14:10, 17 April 2007 (UTC)
 * Sadz, you mean (Bzyp does not have phonemic gemination). Vaux and Pəsiypa (1997) do not give a list of the consonants which have geminated forms in Sadz, leading me to think that all consonants may be capable of appearing either short or long. The consonants attested in Vaux and Pəsiypa's paper as having geminated forms are the following: []. As you can see, there appears to be no phonological reason why all consonants shouldn't have a corresponding geminate. Thefamouseccles 02:10, 20 April 2007 (UTC)

Are there any consonants which occur only geminated in the Bzyp Sadz dialect of Abkhazian? --88.76.248.25 09:39, 22 April 2007 (UTC)


 * No, as far as I know all consonants may appear as singletons, and geminate consonants are also reduced to singletons in word-initial position. Thefamouseccles 02:18, 23 April 2007 (UTC)

Abkhaz and Kwakwala
The phonology of these languages seems to me very similar —Preceding unsigned comment added by Segnor Bugatti (talk • contribs) 17:23, 19 March 2008 (UTC)
 * The phonemic systems of Amharic and Mingrelian are quite similar, too. That doesn't necessarily mean anything. Thefamouseccles (talk) 14:20, 24 May 2008 (UTC)

Colourful phonemes
The text beneath the table refers to red and green letters, but I don't see any coloured entries in the table. Did they get lost in an edit somewhere? -- pne (talk) 11:07, 24 July 2008 (UTC)
 * The alveolo-palatal and uvular columns should have consonants that are dark red and dark green. — Æµ§œš¹  [aɪm ˈfɻɛ̃ⁿdˡi]  16:16, 24 July 2008 (UTC)
 * this edit summary is confusing. How is switching to blue and green beneficial to the colorblind? — Æµ§œš¹  [aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]  08:27, 25 February 2010 (UTC)

I'm vision impaired and I still can't differentiate the dark green and blue from black! I'll add asterisk (*) and dagger (†) as per the description within the article itself, but keep the colours unchanged for those who use them. 121.45.172.20 (talk) 03:43, 5 August 2019 (UTC)

Quality of the high close vowel 'ы'
So there's seems to be quite a lot of uncertainty in regards to this vowel, the description here being: >and a close vowel /ɨ ~ ə/. The page only cites one source, and there it's referred simply as a "close vowel". Chirikba refers to it as a high close vowel. Another paper by Bert Vaux, first mentions that "The phonemic status of ә is disputed" and "will be assumed here for ease of exposition", and then he generalises it saying it's [ɨ] for simplification. I want to argue that the vowel 'ә' isn't as high as [ɨ], and [ɘ] would be a better description for the sound in general. In the book "Грамматика абхазского языка: фонетика и морфология", they make it very clear that the ы in Abkhaz, and the ы in Russian are different sounds. The first indication that 'ы' isn't [ɨ] comes from this passage:

>гласный ы по сравнению с а является более закрытым, что зависит от более высокого положению языка. По сравнению с абхазским ы, русское ы по положению языка более задний и более закрытый гласный.

Translation by google translate

> The vowel ы is more closed than a, which depends on the higher position of the tongue. Compared to Abkhaz ы, the Russian ы is a more back and close vowel by the position of the tongue.

This part alone saying that the Russian 'ы' is more closed, should be enough to say that the Abkhaz 'ы' is probably [ɘ] or [ɘ̝]. The next relevant passage I'll mention is this:

> During articulations of ы, the tip of the tongue touches the crowns of the lower incisors so that the anterior part of the tongue at the level of the canines rises somewhat above the level of the lower incisors. the cavity of the pharynx(?) is a resonator, tapering slightly in its upper part. The oral cavity is a tube, the wide part of which is turned backward, and the narrow part is forward.

So to close my case, I say we should present [ɘ] as one of the main realisations of 'ы' Lugë Ushqimi (talk) 02:34, 5 August 2021 (UTC)

Vowels
The article now says:

Abkhaz has only two distinctive vowels: an open vowel /a/ and a close vowel /ɨ ~ ə/. These basic vowels have a wide range of allophones in different consonantal environments, with allophones [e] and [i] respectively next to palatals, [o] and [u] next to labials, and [ø] and [y] next to labiopalatals

Unfortunately there is no source for it. Chirikba 2003 says that "[o] is the phonetic realization of the sequences aw, wa and awa, as well as the allophone of /a/ in contact with the labialized consonants. [e] is the allophone of /a/ in combinations /aj/ = [ej], /ja/ = [je] ..., or in combinations "palatalized consonant+a" ... . [i] is the allophone of /ə/ in sequences /jə/, /əj/, or in combinations of /ə/ with the palatalized consonants. Finally, [u] appears as the realization of phonemic sequences wə, əw, and of combinations of labialized consonants with schwa." It's close enough but he doesn't mention [ø] and [y]. Hewitt 1979 adds that /a/ can be realised as Close_central_unrounded_vowel ([ɨ]) (not Close_front_rounded_vowel ([y]) and /ə/ can be realised as Close-mid_central_rounded_vowel ([ɵ]) (not Close-mid_front_rounded_vowel [ø]). Should we follow Hewitt 1979 and update the article?, I'm pinging you as you have edited the article recently. Alaexis¿question? 06:55, 26 August 2021 (UTC)

e and o
Unlike Chirikba and Hewitt, folks at northeuralex (University of Tubingen project) use [ɛ] and [ɔ] in their IPA transcriptions instead of [e] and [o]. Is there a good reason for that? Should we note it in the article? Alaexis¿question? 07:10, 26 August 2021 (UTC)


 * Yeah the article now isn't in a good state, very few information about everything and lots of these claims are without any sources, and most likely wrong.
 * > Should we follow Hewitt 1979 and update the article?
 * It's kinda wacky honestly, he's the only person who has made these claims, I don't think it should be the only one being described in the article. I think we should present the different analysis by all the linguists here, they don't really agree with each other at all, and lots of these are old. At least Hewitt 2010, he doesn't seem to stand by his old findings, he doesn't mention the [ɵ] realisation for ы, he rather only analyses it as [ə] there, and instead of [a] he uses [ɑ] instead, which is the analysis of some Russian linguists that I cited in the main article.
 * > Unlike Chirikba and Hewitt, folks at University of Tubingen project use [ɛ] and [ɔ] in their IPA transcriptions instead of [e] and [o]. Is there a good reason for that?
 * Kinda, they cite a Russian source for it . Hewitt 2010 seems to agree with the [ɔ] only when analysing the dynamic suffix '-уа-'. I'd mention this one made by Hewitt instead.
 * Also there's a problem, I've spent some time analysing, reading and listening to Abkhaz, and not all realisations of /aj/, /ja/ are actually a long [e], same for [aw] and [wa] being a long [o]. The new orthography, laid be Kaslandzia 2005, seems to be phonetic. That would explain a lot of stuff. For example in the old orthography they wrote the word for 'all' as 'заигьы', in the new orthography they write it as 'зегьы'. We can't know for sure, since the only updated orthography is in Abkhaz , but that would explain the discrepancy of why in some places 'е' is used instead of 'aи', specially in nouns. The orthography changes the vowels and consonants only in some contexts, here are some I've noticed:


 * 1) With verb suffixes the vowel in the verb stem changes. As an example, with the stem -ца- meaning 'to go', if we add the dynamic suffix -уа- it becomes -цо-, then we can add the finite suffix -ит and the personal marker for the first person singular с(ы) we get сцоит 'I am going'. If we get a verb with a 0 stem like -ԥс(ы)- 'to die', when we add the dynamic suffix and the finite stem we get -ԥсуеит-.
 * 2) With some nouns receiving some suffixes, the vowel or the consonant may change. аҵыс -> ацар
 * So yeah there's a lot of inconclusive stuff. Lugë Ushqimi (talk) 14:19, 26 August 2021 (UTC)
 * Thanks for the answer! I think that for the the Abkhaz alphabet article we can use vowel transcriptions from Hewitt 1979 to be consistent, noting that there are alternative transcriptions and pointing the reader to this article. There the goal is just to give the general idea of the correspondence between the letters and the sounds, while the details can be dealt with in this article. I don't think I'm qualified to improve this article, if you could do it it would be awesome! Alaexis¿question? 08:52, 29 August 2021 (UTC)