Yonagu by Nancy - vowel systems

[finlay]

I've always heard that Ubykh (and as Mecislau said, Abkhaz) has only two vowels, but I've also heard that those vowels sound like an ordinary full vowel system because of allophony with the palatalized and labialized consonants. In that case, why do we even describe it as having only two vowels?

Because what you just wrote there is pretty much the very definition of allophony?

It's not that stupid a question, actually, especially for Eddy. Having only two contrastive levels doesn't necessarily mean that its speakers consider themselves to have 2 vowels, although their writing system might cloud their judgement. Or something. Basically it's just questioning phonemic theory, which has been done quite a lot by linguists too...

[jmcd]

I thought of one nitpick: You should probably specify that T7 only fits Bengali for the oral vowels. It has seven nasal vowels on top of that.

Also I would say this is more like Scottish English:

Code: Select all

i     ʉ
   ɪ
  e   ə   o
   ɛ    ʌ ɔ
      a

Also, as

Typo?

System S4, our first square system, is found in Proto-Slavic and in the short vowels of Latvian and Lithuanian, and also in Akkadian and Magalasy.

(Actually, now I'm finding usages of both spellings. Just kidding!)

, that should be Malagasy.

[Cedh]

Also I would say this is more like Scottish English:

Code: Select all

i     ʉ
   ɪ
  e   ə   o
   ɛ    ʌ ɔ
      a

Which is basically S9 with slightly fronted /ɨ u/ and an additional open-mid central vowel.

[finlay]

/ɨ/???

[Torco]

I recommend making the phrasing much more fun, verbose, and friendly, putting in a few jokes and giving it a more introductory theme.

Also, this is GUD

[Cedh]

/ɨ/???

S9 has /i ɨ u/ as its high vowels, and the system given by jmcd has /i ɪ ʉ/ instead, so one could say that /ɪ ʉ/ in Scottish English correspond, slightly fronted in comparison, to /ɨ u/ in S9.

[Mecislau]

I've always heard that Ubykh (and as Mecislau said, Abkhaz) has only two vowels, but I've also heard that those vowels sound like an ordinary full vowel system because of allophony with the palatalized and labialized consonants. In that case, why do we even describe it as having only two vowels?

Because what you just wrote there is pretty much the very definition of allophony?

It's not that stupid a question, actually, especially for Eddy. Having only two contrastive levels doesn't necessarily mean that its speakers consider themselves to have 2 vowels, although their writing system might cloud their judgement. Or something. Basically it's just questioning phonemic theory, which has been done quite a lot by linguists too...

Well, it's typically the linguists that came up with the two vowel system. If you look at actual written Abkhaz, for instance, it indicates many additional vowels that linguists would typically call allophonic variants.

And in general, I've found that "native speaker perceptions" tend to be a pretty awful yardstick, at least as far as abstract concepts such as phonemes are concerned. Well, after all, they're abstract.

[Niedokonany]

It would be also interesting to know how, say, Russian loanwords - I suppose there are some - conform to the two-vowel analysis of Abkhaz, whether they break some of the nice phonological patterns or not, and e.g. whether the 'allophonic' variants rhyme in their poetry or not (provided they use rhymes ).

[ná'oolkiłí]

It would be also interesting to know how, say, Russian loanwords - I suppose there are some - conform to the two-vowel analysis of Abkhaz, whether they break some of the nice phonological patterns or not, and e.g. whether the 'allophonic' variants rhyme in their poetry or not (provided they use rhymes ).

The materials I have access to at the moment unfortunately aren't stellar, but I can find a few examples for Abxaz.
Georgian maneti → Abxaz /amaːt/ "ruble" (the initial /a/ is an article)
Georgian kalaki → Abxaz /akalakʲ/ "town"
??? → Abxaz /abajk'əwn/ "master" (This looks a lot like a loan, but I couldn't say from where. Phonetically I think the root'd be [bɛːk'uːn] or [baik'uːn]. Does anyone recognize it?)

However, I can find some better examples for Kabardian (only some with specifically given etymologies, though, so I'm digging around a little).
Arabic → /sahbəj/ [saːbiː] "baby"
Russian naučnɨj → /nawətʃna/ "scientific"
Turkish bin (I'm guessing) → /mjən/ [miːn] "thousand"
Turkish top → /tawp/ [tʰoːp] "cannon"
Persian din (via Turkish) → /dəjn/ [diːn] "religion, faith"
Turkish (?) → /bwərʃ/ "black pepper"
Something old and IE ? → /k'rəw/ "crane" (cf Latin grūs, Armenian krunk)
/wərəs/ "Russian" (the first syllable is prosthetic, as except for recent loans, apparently no word begins with /r/)

Recent loans to Kabardian can be more transparent:
Arabic sˤūra (via Georgian surati?) → /swərat/ "picture"
Russian respublika → /rajspwəbljəka/ "republic"
English computer (via Russian) → /kawmpjawtajr/

So it appears that phonetically they have no problem borrowing noncentral vowels, at least recently; phonemically, however, they can be boiled down to /a/ or /ə/ + a glide like native vowel allophones.

[Tropylium⁺]

I have no time to get into this deeper right now but I'd like to point out that "Lappish" and "Berber" aren't languages…

OK, 2nd thing: some Australian language might be useful as a "pure" 3-vowel example. Arabic, Aleut, Inuit all have a length contrast (as does Sanskrit for which you do remark on this), and Quechua has noticable allophonic variants near uvulars, so they're really kinda 3+3 vowel systems.

[alice]

Deity, that's a lot to take in. I've made another round of changes, which you can see on the same page; Tropylium*'s post above appeared just after they went up. In particular I've highlighted the languages for ease of recognition.

I don't think I'll add any further vowel systems, since there are exactly 42 there now, so I'll only fix typos and other obvious mistakes from now on.

If anyone wants to rewrite any parts of the text to make it clearer to the uninitiated, I won't be at all offended; I don't think there's much point in copyrighting this kind of thing, and I'm happy for it to be credited as a group effort as long as I'm identified as the originator. Unless, of course, there are good reasons for doing otherwise.

[tatapyranga]

You should be a bit more careful with specifying varieties— the Chinese you cite is presumably Mandarin, but I don't know if you're doing Brazilian or European Portuguese. It'd also be better to say Southern Quechua.

Brazilian Portuguese fits more T7 (or possibly an 8-vowel variant of it where 'a' is split into 'a' and 'ɐ').

[Trailsend]

Also, as

Yeeeeah still trying to figure out how I did that...it was very late at night.

[Thomas Winwood]

some Australian language might be useful as a "pure" 3-vowel example. Arabic, Aleut, Inuit all have a length contrast (as does Sanskrit for which you do remark on this), and Quechua has noticable allophonic variants near uvulars, so they're really kinda 3+3 vowel systems.

Auslangs generally have a length contrast too, and considerable allophony. Using Martuthunira as an example because I have the PDF hanging around:

• /i/ is before a palatal consonant, [e] before /rr n l/ and [ɪ] elsewhere;
• /i:/ is [ɪ:] in stressed (= morpheme-initial) syllables, [e:] elsewhere;
• /u/ is in stressed syllables, [ʊ] in unstressed ones, [ʉ] following a lamino-dental consonant and [ɨ] before /r/;
• /u:/ is [ɔ:] following lamino-dentals, [ʊ:] elsewhere;
• /a/ is [ɛ] following lamino-dentals and lamino-palatals, [ɒ] between a back rounded segment (e.g. /w/) and a velar, [ə] when unstressed and [ɐ] elsewhere;
• /a:/ is [ɐ:] everywhere.

I seem to recall Jaqaru is a better example of a three-vowel system with minimal allophony, though it still has a length distinction.

[Niedokonany]

So it appears that phonetically they have no problem borrowing noncentral vowels, at least recently; phonemically, however, they can be boiled down to /a/ or /ə/ + a glide like native vowel allophones.

I wonder though if they really have a good justification to analyze the vowels as phonemically complex in those loanwords. I think the bivocalic analysis must be based on some regular alternations found in the language's inflectional and derivational operations, right?

[ná'oolkiłí]

Yep. Here are some Kabardian examples—

/tɬ'ə-ha-r/ → [tɬ'əhar]
man-PL-ABS

/tɬ'ə-w/ → [tɬ'uː]
man-ADJ "manly"

/tɬ'ə-j/ → [tɬ'iː]
man-and

/psa-m/ → [psam]
life-OBL

/psa-w/ → [psoː]
life-ADJ "alive"

/psa-j/ → [pseː]
life-and

Yes, I suppose it's really not possible to tell if /kawmpjawtajr/ is a more correct analysis than /koːmpjoːteːr/, however, since there are no infixes or any other way to alternate the approximants. It's very common crosslinguistically, of course, for phonological rules to be bent for lone words (so that second analysis is defensible), but certainly in native words there are only two or three phonemic vowels.

[tezcatlip0ca]

And your Liotan isn't so bad, I would call the two monsters /@ O/ and the vowel system becomes S12RC /i e E { y 2 9 @ u o O A/ but with /I/ instead of /9/.

[alice]

And your Liotan isn't so bad, I would call the two monsters /@ O/ and the vowel system becomes S12RC /i e E { y 2 9 @ u o O A/ but with /I/ instead of /9/.

Fair enough, but that's not how it actually worked

[Bedelato]

Interesting and helpful webpage, by the way, although I recall seeing it somewhere before.

I think I know what you're talking about.

It's not exactly the same, but there's obviously a relationship.

[makvas]

Interesting and helpful webpage, by the way, although I recall seeing it somewhere before.

I think I know what you're talking about.

It's not exactly the same, but there's obviously a relationship.

There's obviously some "borrowing" going on, which is fine and great, but Nancy should probably include a cite or acknowledgement.

[Radius Solis]

Interesting and helpful webpage, by the way, although I recall seeing it somewhere before.

I think I know what you're talking about.

It's not exactly the same, but there's obviously a relationship.

There's obviously some "borrowing" going on, which is fine and great, but Nancy should probably include a cite or acknowledgement.

That page was his own work. "Nancy Blackett" = "Bricka".

[makvas]

...

That page was his own work. "Nancy Blackett" = "Bricka".

And Bricka == Geoff?

[Nesescosac]

Yes, bricka == Geoff.

[alice]

Very minor edit: there are no longer any references to universals. I'm not sure any are necessary anyway.

[Rory]

I would like to see a discussion of vertical vowel systems, as they're pretty rad. That could tie in with discussion of two-vowel systems too.
I was halfway through reading when I understood what the T6 and C8R stuff meant - perhaps an overview at the beginning would be helpful?