Jump to content

Help talk:IPA/Oghuz languages/Archive 1

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 1

ɣ

This page is missing the IPA symbol [ ɣ ]. Can someone add it with an example? ForestAngel (talk) 20:40, 29 September 2009 (UTC)

It's listed as [ɰ], [j], and length, as Turkish does not have [ɣ]. Or do we need it for Azeri? kwami (talk) 00:18, 30 September 2009 (UTC)

we don't have "ɣ" in Turkish but we need it in Azerbaijani. It stands for "ğ" for example in "toprağ :[tɔpraɣ]" — Preceding unsigned comment added by 46.28.74.149 (talk) 07:45, 14 October 2015 (UTC)

Tuncay Şanlı

Can someone check the stress for Tuncay Şanlı? Can both syllables be stressed? Lfh (talk) 13:16, 6 December 2009 (UTC)

Split

I suggest that this page be split. Azeri phonology is quite different from Turkish. The page also has some inaccuracies.

  1. Azeri contrasts between velar stops (represented by q and k) and palatal stops (represented by g and k), while in Turkish the latter are simply palatal allophones of the former;
  2. Azeri contrasts between /e/ and /æ/, while Turkish does not;
  3. Azeri does not contrast between /l/ and /ɫ/, unlike Turkish, or between /a/ and /ɑ/
  4. Azeri has additional phonemes: /x/, /ɣ/ (represented by ğ and much stronger than in Turkish, so please do not remove it from transcriptions of Azeri names in articles) and /ç/ (an allophone of /c/, represented by k);
  5. Azeri does not have the voiceless uvular stop /q/. The grapheme q represents /ɡ/. Parishan (talk) 12:20, 14 January 2010 (UTC)

I tend to agree. Even more so with Turkish and Turkmen. Cyco130 (talk) 14:37, 11 April 2013 (UTC)

gelin (Azerbaijani), 'bride'

It's 'gəlin', not 'gelin'.--PPerviz (talk) 02:48, 16 December 2010 (UTC)

mid vowels

I've gone through and replaced ɛ œ ʊ ɪ with e ø u i, including in the Turkish language article. There didn't appear to be any rhyme or reason to the variation. Check my changes for this hour (22:00, 22 December 2010 UTC) to see if anything needs to be reverted.

I'm finding numerous long vowels which are not reflected in the orthog, so I'm deleting them too.

Scanning the transclusions for all letters not in our chart for Osmanli. Getting lots of mistakes for /ɾ/ and /a/.

Not sure what to do for palatal/velar between mixed vowels, as in Lale, so I'm choosing to assimilate to the following vowel: [ˈɫale]. ('Course, in that case it might really be ?Lâle, but there are others, like Yenikapı.)

Not sure what to do with Akhal-Teke. Where does the [j] come from, the Russian maybe? Other than that, the only likely problem I see would be if there was a historical ‹â›, still pronounced as such (front harmony or length) but no longer written, in which case I've miscorrected it; or if I've changed an [l] to [ɫ] or [k] to [c] in Azeri or Turkmen without noticing. — kwami (talk) 23:54, 22 December 2010 (UTC)

Missing sound

The current IPA for showing the "ey" sound is not accurate.

"ey" should be listed in the vowel section, for example:

  • IPA: ei
  • Example: Zeynep Bey Cüneyt
  • English equivalent: Hay Pay

Could someone add this sound?

Thanks & Regards. John Cengiz talk 01:50, 25 January 2011 (UTC)

Someone mistakenly put for Turkish beygir. That word is pronounced [bejˈgiɾ] as far as I know... It needs to be corrected. Also, since when is i in dil long? Dil is dil, not dîl. — Preceding unsigned comment added by 178.223.30.108 (talkcontribs)
You are right, it is wrong. I am removing it. --İnfoCan (talk) 16:55, 20 July 2011 (UTC)

I've reverted the last couple months' edits. They either were not improvements, or they would've been fine if they matched the articles and were consistent with each other, but they weren't. — kwami (talk) 17:40, 18 August 2011 (UTC)

I am going to return these essential sounds; the iː (as in "see") and eː (as in "pay").
I have spent some time adding them with no help and you remove them, it is not possible to do an IPA for many words without them. John Cengiz talk 18:02, 18 August 2011 (UTC)
I'll just remove them again, unless they're actually used in the articles. This isn't a phonology article, it's a key to how the IPA is used. If those transcriptions aren't used, they don't belong here. — kwami (talk) 18:06, 18 August 2011 (UTC)
Also, how is ü the vowel in "soon"? — kwami (talk) 18:33, 18 August 2011 (UTC)
Of course the ü is longer and such, "soon" is a far better example than "German ü or French u".
Ceyhun, Cüneyt, Eyüp, Feyyaz, and most of the other articles that link to IPA for Turkish, Azerbaijani and Turkmen are IPAs added by me. John Cengiz talk 18:40, 18 August 2011 (UTC)
But you transcribe them differently in IPA than you explain them in English. Something's wrong here. If it's like "soon", then it should be [u], not [y], which is nothing like English "soon". — kwami (talk) 18:50, 18 August 2011 (UTC)
This is ridiculous. I just protected the article rather than having you blocked. We have an ongoing discussion over how to handle the vowels; you don't get to come in here and unilaterally decide to use a different system than the 500 existing articles, and convert the key to match your preferences rather than those 500 articles, without some agreement from the other people here. Meanwhile, the sources contradict each other, but they also say you're wrong. — kwami (talk) 19:11, 18 August 2011 (UTC)

conflicting transcriptions

We've been arguing over the vowels, so here are two conflicting accounts. We might consider the consonants, as well, esp. if we say they occur in Turkmen or Azeri but not in Turkish, as with /h/.

  • Turkish Phonology, Morphology and Syntax
Zülal Balpınar
Anadolu Universitesi, 2000
/ɪ/ › [i] before ğ
/ʊ/ › [u] / ğ
/ɛ/ › [e] / ğ, [æ] some speakers, some environ. is [ej].
/ɐ/ › [ə] first syllable, [a] some speakers, not common
/ɾ/ › [ɾ̝] init, [ɾ̝̊] final
/ʋ/ › [v] after vcls C, silent before labials
/t d n s z l ɾ/ are alveolar
  • Turkish: A Comprehensive Grammar
Aslı Göksel and Celia Kerslake
Routledge, 2005
/ɑ/ › [a] at c ɟ and l in loans
/e/ › [ɛ] finally, › [æ] before m., n., l., r. ; değil is [dejil] or [diːl]
/i/ › [ɪ] final
/ɯ/ maybe › [ɯ̞] final
/u/ › [ʊ] final
/y/ › [ʏ] final
/aː uː iː eː/ – this last is in teessüf, temin, iyi, not teğ-
/h/ › [ç x ː ] assim.
/n̪/ › [ɲ ŋ] assim.
/ɾ/ › [ɾ̥] final
/ɣ/ › [ ː j β̞ ], [ɣ] dialect.
/v/ › [β β̞] at rounded vowel (approx. intervocalic)
/f/ › [ɸ] before rounded vowel
/pʰ t̪ʰ tʃʰ cʰ kʰ/
/ l̠ʲ /, / l̪ˠ /
/t d n s z/, dark /l/ are denti-alveolar

kwami (talk) 19:03, 18 August 2011 (UTC)

I'm completely on "Turkish: A Comprehensive Grammar"s side. — amateur (talk) 19:35, 18 August 2011 (UTC)
Yeah, Routledge is usually pretty good. But something we should discuss: which allophones do we indicate? in which environments? do we adjust consonants too? Because whatever we decide, we'll need to update 500 articles to match, so it would be nice to do this in one go, and then have convention stable for a while.
One thing, though: while I'd expect those /f, v/ allophones by front rounded vowels, I wouldn't expect them for back rounded vowels. That's the kind of thing you get with Japanese, where /u/ isn't really [u] at all. — kwami (talk) 19:58, 18 August 2011 (UTC)
Could I just ask about the meaning of the notation above? For example, what is usually (everywhere else) e, before m, n, l, r it is æ, word-finally it is ɛ and the sequence -eği- is either [eji] or even [eː], even in standard speech, taught e.g. in Turkish drama schools?! What do Karl Zimmer and Orhan Orgun say?
And what does
/pʰ t̪ʰ tʃʰ kʰ/
/ l̠ʲ /, / l̪ˠ /
/t d n s z/, dark /l/ denti-alveolar
translate to? p, t, tʃ, c, k are always aspirated? l is always palatalized and velarized? Last line: l is 'dark' before t, d, n, s, z? How does that go along with the previous line?
You can look it up. The text explains it. — kwami (talk) 13:41, 19 August 2011 (UTC)
Then perhaps you should leave it as it is, perhaps only chaning Turkish a to ɑ, as per the Routledge grammar, and write a new section about Turkish allophones in Turkish phonology article and dividing it by sections which would correspond to the grammars listed by you above.

Can s.o. verify the stress is irregular? And V length while you're at it. — kwami (talk) 18:41, 11 April 2012 (UTC)

We have an anon. IP who is placing or moving stress to final syllable in many personal names, such as Roxelana, Şevval Sam, Selma Ergeç, and Tuba Büyüküstün, as well as adding vowel length not reflected in the orthography. Is this correct? I thought (ante)penultimate stress was pretty much the norm for Turkish personal names. (The user generally seems to know what they're doing, but has made elementary errors in English transcription.) — kwami (talk) 20:52, 11 April 2012 (UTC)

The "ə" in Azərbaycan

Is pronounced as "e" in "pet". This is the letter they use for the "e" ("pet") sound. It is far from always the quick "uh" sound like shown in the English IPA. The IPA key shows Azerbaijani not having a "e" ("pet") sound. John Cengiz talk 17:46, 25 May 2012 (UTC)

The table says it's closest to the vowel of cat (which is close to the vowel of pet, especially for speakers of languages that don't distinguish the two vowels). Is there something that indicates otherwise? — Ƶ§œš¹ [ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ] 20:24, 25 May 2012 (UTC)
I don't see the vowel sounds in "cat" and "pet" as close when describing a languages sounds properly. The 2 vowel sounds are distinguished between in Azerbaijani. Much like how our IPA for the quick "a" sound (e.g. the "e" in runner) uses the "ə" letter, while we use the "ɛ" letter for the sound in "fell".
The key was probably made by a non-speaker of Azerbaijani, and such confused the English IPA letter "ə" sound for every "ə" letter in Azerbaijani being the same. Developing on my point, the only Azerbaijani word in the whole key is "Azərbaycan", which wrongly shows it pronounced as the "English" way. John Cengiz talk 00:39, 28 May 2012 (UTC)
The key was made based off of information found at Azerbaijani language and Azerbaijani alphabet. There are other clues that lend credence to ә/ə indicating the pronunciation [æ] (the vowel of cat in English). For example, the article on the alphabet indicates that this vowel was written ä for a short period of time, which is traditionally used to indicate [æ]. Also, as our article Schwa (Cyrillic) indicates, utilization of this letter for [æ] also occurs in Bashkir, Kalmyk, Kazakh, and Tatar; romanization of these languages also utilizes the letter ä.
Of course, all of these could be false. As I said, [æ] and [ɛ] are very similar and speakers of languages that don't distinguish them have trouble with the distinction when learning English. But nobody was confused about the distinction between the Azeri letter and the IPA transcription of English sounds. — Ƶ§œš¹ [ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ] 15:50, 28 May 2012 (UTC)
Here is a link to a site where 2 people from Azerbaijan pronounce their countries' name, it corroborates my view. John Cengiz talk 19:23, 29 May 2012 (UTC)
I speak a dialect of English that doesn't distinguish between [æ] and [ɛ] before /r/. So I can't hear the difference. — Ƶ§œš¹ [ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ] 20:38, 29 May 2012 (UTC)
Try listening carefully, for example in Spanish, by default a word with "er" it it (assuming the letter "e" isn't being used for the sound in "play") isn't like our English sound in "runner" (ər). It is a definite "ehr" (ɛr), it is the same with Turkish, who have influenced Azerbaijan, and such pronounce words very similarly or the same. Try using Forvo, here's a few words (Spanish: interceptor, eruditos, santander, roberto). John Cengiz talk 02:45, 1 June 2012 (UTC)
Nobody is saying that the Azeri letter ⟨ә⟩ represents [ә]. — Ƶ§œš¹ [ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ] 06:13, 1 June 2012 (UTC)

Sounds like [ɛ] to me. And Turkish words like sen, which Turkish speakers claim have [æ], are clearly [ɛ] in the Forvo recordings they refer people to. I wonder if there isn't just a tradition of using ⟨æ⟩ for this sound in Turkish lexicography, so that's what most Turks think the letter means. I've been all over Turkey, usually speaking Turkish, and I don't ever remember hearing anything like an English [æ] from anyone. — kwami (talk) 22:10, 2 August 2012 (UTC)

Well, Azeri məna is clearly [mænɑ] at Forvo, but the same speaker varies as high as [e] in other words.

Turkish alem is [ɛ], but eldan (diff speaker) is [ɛldæn]. But per our sources, that's only some speakers. — kwami (talk) 22:23, 2 August 2012 (UTC)

There is considerable variation in [e ɛ æ] in Turkish. As far as I can tell it's not phonemic (anymore?) and it's not reflected in ortography. Cyco130 (talk) 14:41, 11 April 2013 (UTC)

loot/cue

I've replaced 'loot' with 'put' so we can keep /u/ (in 'cue') for Turkish/A/T /y/. /u/ is central in most places outside of NA. — Lfdder (talk) 18:07, 27 April 2013 (UTC)

ğ

I am confused about footnote 2. What if it's not between rounded vowels, or between unrounded vowels, but rather between a rounded vowel and an unrounded vowel, to wit: Erdoğan? I see the IPA on the Erdoğan article, so I know the answer for that particular name, but what is the overall rule? I think fn2 is either unclear or incomplete. Can anyone help fix it? -- Y not? 05:42, 20 June 2013 (UTC)

I second that plea! PJTraill PJTraill (talk) 20:57, 10 August 2014 (UTC)

Stress incorrect

I added a note that stress is generally wrong in our articles. Can someone review them? — kwami (talk) 01:20, 31 October 2013 (UTC)

Should we split this guide?

User:Cfsenel keeps reverting my redirect from Help:IPA for Turkish to this page. The last reason for his revert was "Exactly, we need more than two people to call it a consensus, so merging the articles is hardly the consensus." The problem is, you split Turkish from this guide without discussing it first! If anyone's actions are against the consensus, they are yours. Either way, the current state of things is unacceptable - Turkish can't have two pages. We should decide what to do - should Help:IPA for Turkish redirect to Help:IPA for Turkish, Azerbaijani and Turkmen? I think it should - there's no valid reason to split Turkish from this page. — Peter238 (v̥ɪˑzɪʔ mɑˑɪ̯ tˢʰoˑk̚ pʰɛˑɪ̯d̥ʒ̊) 22:15, 8 December 2014 (UTC)

I am restoring the merge. Irrespective of whether it should be split or combined, Cfsenel is making a mess of things by the way they have handled this. Peter is also right. Discussion should come first before making the split. Cfsenel, please cease the disruption. — Ƶ§œš¹ [lɛts b̥iː pʰəˈlaɪˀt] 23:54, 21 December 2014 (UTC)
Calling it disruption is heavy-handed. Also see #Split, which is the discussion they were referring to. 62.228.28.188 (talk) 00:20, 22 December 2014 (UTC)
If it were just a matter of article space, it would simply be a bit contentious. But the help page is linked to by many other articles and the way it was done was sloppy. If we are to do a split, this page needs to be renamed to IPA for Turkish and a new IPA for Azerbaijani and Turkmen should be created.
Anyway, quick question, how many Azerbaijani and Turkmen terms are transcribed at Wikipedia? I notice that there's no IPA-tk or IPA-az templates, which means that someone would have to go through the hundreds of articles using {{IPA-tr}} to see if a split would even be wortwhile. — Ƶ§œš¹ [lɛts b̥iː pʰəˈlaɪˀt] 02:43, 22 December 2014 (UTC)
IPA-tr has parameters for Azeri and Turkmen we could use a hidden cat on, if we really cared. I suggest we simply do away with the 3 sub-columns, 'cause this page really is meant to help people find the pronunciation of a letter, not give them a rundown of the differences in these languages' phonologies; and keep the redirect. 62.228.28.188 (talk) 03:24, 22 December 2014 (UTC)
It's important to keep the three columns because the letter pronunciations differ from language to language. That's what the separate columns show. — Ƶ§œš¹ [lɛts b̥iː pʰəˈlaɪˀt] 19:56, 22 December 2014 (UTC)
What's important is to show the pronunciation an IPA letter corresponds to. We'd split adjacent IPA letters into new rows. For example:
IPA Orthography Approximation
tr az tk
f ɸ f food; food but made only with lips
j y (Turkish/Azeri), ý (Turkmen) yet
ɣ ʁ ğ (Turkish/Azeri)[1], g (Turkmen) [1]
j
β

IPA Orthography Approximation
tr az tk
f f f food
ɸ f like food, but with both lips
j y
ğ[1]
y ý yet
ɣ ğ Spanish "weak" g
ʁ g French r
93.109.28.109 (talk) 20:04, 22 December 2014 (UTC)

With the system on the right, editors would not have the information to tell them how to transcribe the specific languages. — Ƶ§œš¹ [lɛts b̥iː pʰəˈlaɪˀt] 21:02, 22 December 2014 (UTC)

Then I suggest that we shift the subcols to the orthography (see amended example). This way, language use will be implied. Bottom line is, the table needs to be made easier to navigate by our readers -- which means one row per IPA letter. 93.109.28.109 (talk) 23:16, 22 December 2014 (UTC)
I see what you're saying. That does have merit for readability to users, but there are also issues of consistency. The help pages for Catalan, Irish, Dutch, Portuguese, and Scottish Gaelic have similar layouts to this. I wouldn't recommend changing one and not the others. — Ƶ§œš¹ [lɛts b̥iː pʰəˈlaɪˀt] 23:56, 22 December 2014 (UTC)
Fair enough, but I'm not touching Portuguese with a sixty-foot pole. Catalan and Dutch are passable because there's only two columns and (nearly) all the letters appear in the first column. I don't see anything that needs to be done to the Gaelics. 93.109.28.109 (talk) 00:06, 23 December 2014 (UTC)
OK, well let's wait for the determination regarding a split before we implement any changes. — Ƶ§œš¹ [lɛts b̥iː pʰəˈlaɪˀt] 00:35, 23 December 2014 (UTC)
Ok, I've made some small changes to the approximations in the meantime. 93.109.28.109 (talk) 03:20, 23 December 2014 (UTC)
As far as I know, consensus does not mean the way things have been. It is something debated, and agreed on. I see no consensus, and indeed, no reason, to have an IPA guide for these three languages in the same page. (By the way, I still have not seen any reason above to have a combined page. Well, maybe lethargy.) I suppose Peter238 was especially uneasy with Turkish having two pages. I did consider removing Turkish from this page, and renaming it as Help:IPA for Azerbaijani and Turkmen (until someone takes the time to create separate IPA guides for these languages as well), but I did not bother, thinking leaving it would not hurt. I imagine you would be more comfortable with that if I removed it?
I think 'messy' would be the word I would use for the current state of the page, when you would expect the opposite, because Turkish has almost one-to-one correspondence between letters and pronunciations. Right now, it is as if the article is written deliberately to confuse readers (I am confused anyway, even though I am fluent in one of these languages). I don't think my solution was all that messy. As Ƶ§œš¹ rightly put, there are no templates, and indeed probably not many terms transcribed, for Azerbaijani and Turkmen. Making the users interested in Azerbaijani and Turkmen go through one more click is not ideal, but it is a step towards the ideal; eventually someone should create a separate template for these languages.
Maybe other solutions can be found to alleviate the problem, but 93.109.28.109 and Ƶ§œš¹ seem to agree that these are not without issues. Could someone tell me, why Help:IPA for Turkish should not exist, which solves these issues, and is much cleaner?--Cfsenel (talk) 10:11, 24 December 2014 (UTC)
We generally combine languages that are philogenetically and phonetically similar (Dutch and Africans, Portuguese and Galician, Thai and Lao). That said, I'd actually like to see it demonstrated that we even need to have help guides for Azerbaijani and Turkmen in the first place. There are a few of these IPA for X help pages that aren't used at all. If that's the case, then I'd say remove their columns and rename this to IPA for Turkish. — Ƶ§œš¹ [lɛts b̥iː pʰəˈlaɪˀt] 20:12, 24 December 2014 (UTC)
In whose benefit would that be? If there's at least one transcription of Azeri, why would we wanna remove the Azeri column and phonemes? 62.228.118.129 (talk) 22:30, 24 December 2014 (UTC)
Is there even one? — Ƶ§œš¹ [lɛts b̥iː pʰəˈlaɪˀt] 03:42, 25 December 2014 (UTC)
[1]. 62.228.118.129 (talk) 10:00, 25 December 2014 (UTC)
OK, six. Not much, but certainly more than Turkmen. There are no articles that transcribe Turkmen pronunciations. I'd say remove Turkmen and rename accordingly. — Ƶ§œš¹ [lɛts b̥iː pʰəˈlaɪˀt] 19:28, 25 December 2014 (UTC)
I've implemented the changes I suggested above, which should alleviate your concerns. If you still think it's messy, I wouldn't be strongly opposed to a split. 62.228.118.129 (talk) 11:41, 24 December 2014 (UTC)
Thank you 62.228.118.129, I would not call this messy, nice job. But I would still argue that a split would serve better for a number of reasons. First of all, especially in the case of Turkmen, the languages are vastly different, as a result of almost a millennium of geographic and cultural separation between them, which is incomparable to, say, Dutch and Africans. In the case of Portuguese and Galician, one could even say that they are variations of the same language. Actually, this is more comparable to the case if you just had a page for Turkish: In different regions of Turkey, pronunciation of Turkish varies just as much as between Portuguese and Galician, which is partly the second reason a split would be beneficial: A separate page for Turkish would allow contributors to note the differences between different accents of Turkish more easily; in the page's current state, no one would dare make the article even more complicated. Actually, having a united page for Turkish, Azeri and Turkmen makes even less sense than having a united page for Portuguese, Spanish and Italian (the languages are at least as different), and if the latter was the case, no one would attempt to add the different pronunciations in Galician or Brazilian Portuguese, as having three different languages (Portuguese, Spanish and Italian) is complicated enough. Same is true for the purposes of adding example words to the article. IPA for Turkish, which Peter238 redirected to this page does involve example words, as in, say Help:IPA for French, which aids the readers. This is much harder to do for a united article, because common words are not that prevalent between the languages, and giving examples from each language makes it far too messy. For Portuguese and Galician one can do this, but again this is more comparable to the case of different accents of Turkish: Karadeniz region or Ege region has exactly the same words, it's just that some letters are pronounced differently, which can be managed in a single article.
So, if there are no objections, I would like to revert Peter238's redirect. Then, if you like, one can remove Turkish from this page and rename it IPA for Azerbaijani and Turkmen. I suppose Ƶ§œš¹ disputes the necessity of having an IPA key Turkmen, so he can make the case for removing it to make it Azerbaijani only, although I would say that it does not hurt to have it on Wikipedia even if it is rarely used. But this is a different discussion. May I say that the consensus is to have a page for Turkish by reverting the redirect?--Cfsenel (talk) 22:51, 27 December 2014 (UTC)
As I said before, the way you did it was improper. If we agree to a split, the thing to do is rename this page to IPA for Turkish and create an IPA for Azerbaijani or IPA for Azerbaijani and Turkmen page.
I have two questions for you, Cfsenel:
1) How would an IPA for Turkish "allow contributors to note the differences between different accents"? I would advise against a free-for-all, especially if it would require original research. At IPA for Spanish, for example, the dialectal differences noted basically have to do with the appearance or merging of some consonants depending on region, but other differences (the realization of vowels, the incidence of certain rhotics, etc) are not addressed. Meanwhile, at IPA for Russian, there is no dialectal variation represented.
2) What are the irreconcilable differences between Turkish and Azerbaijani?
You've made a compelling case for splitting Turkish from Turkmen, but I don't think we should have Turkmen here at all. It's not just used "rarely." It's not used at all. — Ƶ§œš¹ [lɛts b̥iː pʰəˈlaɪˀt] 23:05, 27 December 2014 (UTC)
I've just added Turkmen pronunciations to Turkmenistan and Ashgabat. — Peter238 (v̥ɪˑzɪʔ mɑˑɪ̯ tˢʰoˑk̚ pʰɛˑɪ̯d̥ʒ̊) 08:32, 29 December 2014 (UTC)
There's little harm in having Turkmen here or creating a separate guide for it. It may actually encourage certain editors to start incorporating Turkmen IPA into articles. Mr KEBAB (talk) 04:24, 27 April 2017 (UTC)

This page needs splitting, it's so clumsy/confusing right now, that there's been an error for 5 years and no one's noticed

Title says it all, this page absolutely needs splitting, one should have a separate page for each language. If they're different enough to deman an entire different column for each language, they're different enough to not be merged into a single Help page in Wikipedia.

Proof that things are surreally clumsy/confusing and that it disrupts people from actually understanding anything is that there's been an important mistake in the page ever since 2010, and apparently no one even managed to notice it, because simply trying to understand what refers to the language you're interested in (and what doesn't concern you at all) in a sentence like "[c]~[k] (Turkish/Azeri) / [k]~[q] (Turkmen), [ɟ]~[ɡ] (Turkish) / [ɡ]~[ʁ] (Turkmen), [l]~[ɫ] only contrast in loan words before <â, û> vs. <a, u>; in native words, [c/k, ɟ/ɡ, l] occur before the front vowels (/e/, /i/, /ø/, /y/), while [k/q, ɡ, ɫ] occur before the back vowels (/a/, /o/, /u/, /ɯ/)." can be quite a challenge.

But the thing is, the italicized phrase above contains wrong information and lack of logic (and has been wrong for half a decade, unnoticed). I had to waste precious minutes decyphering it, trying to understand Turkmen pronunciation (the only language I cared about at the moment, so, really, I shouldn't be obligated to have to try so hard to dodge the other two languages I don't wanna know about).

And, after finally decyphering said phrase, I understand it clearly says that, in Turkmen, "[ɡ]~[ʁ] ... only constrast in loan words. In native words, ɡ occurs before front vowels, while g occurs before back vowels". Sic. That's what this phrase has been saying for half a decade. There clearly is a mistake there; either the first "g", or the second "g", should have been a "ʁ". Which one? I have no idea, that's what I wanted to find out, but the way the article currently is written, I've lost one whole our to only find out that the article is wrong... MissionFix (talk) 02:33, 9 March 2015 (UTC)

Good find. You know, it would have been pretty easy for you to have fixed the problem while also making the above post. I also recommend previewing your edits before you save them; you actually ended up using the wrong template. — Ƶ§œš¹ [lɛts b̥iː pʰəˈlaɪˀt] 04:00, 9 March 2015 (UTC)


Still, it contradicts what is said here https://en.wikibooks.org/wiki/Turkmen/Alphabet , according to which the difference, in Turkmen, between usage of g and ʁ has nothing to do with the preceding vowels, but with whether the letter is at the beginning or middle of a word. Maybe we should just admit that we know nothing about the language, and that it is an entirely different language from Turkish, and separate the articles accordingly? And ask REAL Turkmen people who actually know how to speak their language to rewrite the whole pronunciation section for their language? MissionFix (talk) 18:51, 9 March 2015 (UTC)

That's already been pretty much agreed on. But the conversation regarding a split unraveled when I asked about Azerbaijani. If we're already talking about restructuring, I'd like to figure out if Azerbaijani also doesn't belong here. — Ƶ§œš¹ [lɛts b̥iː pʰəˈlaɪˀt] 19:20, 9 March 2015 (UTC)
The [g]/[ʁ] difference in Turkmen is indeed only dependent on the position of the phoneme in the word. – amateur (talk) 12:23, 13 March 2015 (UTC)

Notes

'cute' for [c], 'argue' for [ɟ]

These are really, really bad approximations. Like trying to compare an orange to a football. — Preceding unsigned comment added by 81.183.6.63 (talk) 18:38, 16 July 2016 (UTC)

No, that's clearly not true. Before /j/ and the front vowels, English /k, ɡ/ are fronted to (post-)palatal place of articulation, so they're either the same or very śimilar (source: Gimson's Pronunciation of English (8th edition, 2014)). Mr KEBAB (talk) 02:12, 22 July 2016 (UTC)
On second thought, I think I understand what you mean. The [j] that follows the fronted [k, ɡ] in those English words should be absent or at least very slight when you pronounce Turkish [c, ɟ]. That is true. Mr KEBAB (talk) 04:17, 27 April 2017 (UTC)

To adding Khorasani

Khorasani Turkic should have adding too. Would I do it but if new row cannot add, may be help it. Turkistanding (talk) 04:01, 26 September 2016 (UTC)

Actually, we should separate this guide into different guides for each language. — Ƶ§œš¹ [lɛts b̥iː pʰəˈlaɪˀt] 04:14, 26 September 2016 (UTC)
Agreed! I'm thinking about creating a guide for Uzbek. Hopefully nobody will want to merge it with this one... Mr KEBAB (talk) 04:14, 27 April 2017 (UTC)

Error report: [d͡ʒ] is missing

Since the reorganization to include Gagauz and South Azeri, the phoneme [d͡ʒ] seems to have been omitted (Turkish and Azeri c, I don't know enough about the other three to know if it's a phoneme in them). 86.137.167.165 (talk) 04:14, 1 February 2017 (UTC)

I've been bold and added rows for [d͡ʒ] and its South Azeri equivalent [d͡z]. Feel free to alter them if needed. 86.137.167.165 (talk) 04:24, 1 February 2017 (UTC)

Move discussion in progress

There is a move discussion in progress on Help talk:IPA which affects this page. Please participate on that page and not in this talk page section. Thank you. —RMCD bot 16:18, 15 July 2017 (UTC)

Problems with the Azerbaijani IPA columns

Within the past couple of days, there's been a minor edit war going on over how best to represent both forms of the Azerbaijani language, where a certain user has been reverting some edits to provide erroneous information not only about the language (particularly as spoken in Iran) but also spreading it into other languages represented here, such as Turkish and especially Gagauz (where <ê> represents /ə/, not /ɯ/ since AFAICT it's an areal borrowing from Romanian or at least a common sound found in Balkan sprachbund languages, and vice versa for <ı> in both Turkish and Gagauz). In an effort to resolve this, I have managed to obtain a copy of The Turkic Languages and will be going through the Azerbaijani chapter to provide proper reference citation, at least for the standard form of Azerbaijani. This will also expand to finally include citations for Turkish and also, more importantly, for Turkmen and thus provide a settlement to the controversy from a couple of years ago (and which remains lodged in the edit area) over the allophonic distribution of the sounds represented by the letters <k> and <g>. In the meantime, could someone please bring this page back to its state, or at least replace the interpolated Latin script back to the Perso-Arabic script, at least as shown in Azerbaijani alphabet? Thanks in advance. Daniel Blanchette 18:07, 25 September 2017 (UTC)

ɰ missing!

The IPA on Erdoğan has ɰ and links to this article, but it's missing here! --Espoo (talk) 05:36, 5 July 2018 (UTC)

  1. ^ a b c In Turkish, between unrounded front vowels (/e/, /i/), ğ is similar to English y. Between unrounded back vowels (/a/, /ɯ/), it is silent. Between rounded vowels (/o/, /ø/, /u/, /y/) it is either silent or may have a very light [β] sound. Word-finally or before a consonant, it lengthens the preceding vowel.