Wiktionary talk:Kazakh transliteration

RFC discussion: December 2014

The following discussion has been moved from Wiktionary:Requests for cleanup (permalink).

This discussion is no longer live and is left here as an archive. Please do not modify this conversation, but feel free to discuss its conclusions.

---

Contributions of User:Amateur55

This person seems to be fairly knowledgeable about several Cyrillic-based languages, but they've been changing transliteration modules and transliteration pages right and left. I would appreciate it if someone who knows something about said languages (@Atitarev? @Stephen G. Brown? @Borovi4ok?) checked to see if I was correct in reverting some of their edits and checked to see if I missed any that should have been reverted. Chuck Entz (talk) 05:12, 24 December 2014 (UTC)

I think you were right in reverting the changes. Any changes can be discussed here. They are numerous and affect various languages. With some languages it's a matter of choice and convention, like e.g. Bashkir. --Anatoli T. ^{(обсудить}/^вклад) 07:06, 24 December 2014 (UTC)

Let me explain the changes I made before your rollbacks:

• The Kazakh transliteration module is horribly wrongly implemented. Among other things, it romanizes both "Ы" and "І" as "I" while the latter should have been "İ" as can be seen in the Wikipedia article about the Kazakh alphabets and the Latin version of the Kazakh news agency website itself. That module also lacks a transliteration of the Kazakh letter Һ. Similarly, all the other changes I made on that module follow the national KazInform system, and thus I'm restoring my changes concerning Kazakh.
• The Kyrgyz transliteration module itself is fine, but in this case, it's the page WT:KY TR that lacks critical information, namely the fact that the Cyrillic letter И should be romanized İ and not I. I don't see anything wrong with restoring this simple change either.
• My other changes can be subject of some discussion so I'm not restoring them until there's a consensus, but here's my reasoning for them:
  • The Tuvan alphabet is identical to the one of Kyrgyz, and the only difference in the sound values of the letters is the fact that Ж represents a fricative rather than an affricate, hence the use of J instead of C. Being related languages, it only makes sense that they are romanized using similar systems.
  • The Bashkir language is closely related to the Tatar language to the point they are considered dialects of one language by some linguists. Again, it would make more sense to romanize these two languages using similar systems rather than one following a Turkic and the other following a Slavic convention. – Amateur55 (talk) 22:50, 24 December 2014 (UTC)

Such big changes, even if 'correct' (transliteration is an art not a science after all) must be discussed on WT:BP before implementing. These are straightforward rollbacks as there is no other option. Renard Migrant (talk) 23:54, 24 December 2014 (UTC)

@Amateur55 The changes to Kazakh and Kyrgyz transliterations are OK, still you're doing it unilaterally. Bashkir and Tatar transliterations need to be discussed. --Anatoli T. ^{(обсудить}/^вклад) 00:00, 25 December 2014 (UTC)

@Atitarev It would be nice if you (or someone else familiar with the procedure) start a discussion at the appropriate place since I'm not sure where/how to do it. – Amateur55 (talk) 01:27, 25 December 2014 (UTC)

@Amateur55 Your Kazakh and Kyrgyz changes seem like minor fixes, which make sense to me (not sure about others). Yes, Bashkir is close to Tatar and they are mutually comprehensible but Bashkir has specific letters and sounds, such as ҙ /ð/ and ҫ /θ/. It doesn't make sense to use ź and ś to transliterate them. Also, Tatar only has /к/ (k or q) but Bashkir has two letters к (k) and ҡ (q) and cognates don't match in spellings. Transliterations of these languages have been around for many years, so please respect the editors involved. You can suggest changes in Wiktionary:Beer_parlour/2014/December (or a page with a corresponding month). --Anatoli T. ^{(обсудить}/^вклад) 04:24, 25 December 2014 (UTC)

@Atitarev Most Bashkir words with /ð/ and /θ/ are cognates with words having /s/ and /z/ in the same position in Tatar and other Turkic languages. Transliterating these phonemes as ś and ź is a common convention among Turkologists.

I don't see how the spelling of cognates regarding ⟨q⟩ and ⟨k⟩ don't match. They definitely do. – Amateur55 (talk) 04:57, 25 December 2014 (UTC)

I see some problem with viewing all Turkic languages from the Turkish perspective. E.g. Tatar verb lemmas are represented differently from Russian Tatar linguists, Tatar (also Kazakh and Kyrgyz) appear in Roman letters in Turkish dictionaries, when speakers of these languages officially and mainly use Cyrillic. ś and ź represent /ɕ/ and /ʑ/ in Polish and may represent /sʲ/ and /zʲ/ in some romanised Belarusian (łacinka). Bashkir usually use ҡ when Tatar uses к, that's where a mismatch is. --Anatoli T. ^{(обсудить}/^вклад) 05:09, 25 December 2014 (UTC)

@Amateur55, are your transliteration schemes documented somewhere? In my opinion, we should use the Turkological convention, if it is documented in some serious published source. --Vahag (talk) 07:41, 25 December 2014 (UTC)

@Vahagn Petrosyan I don't know whether these count as serious published sources but, though both differ slightly in some other letters from the version I posted, both of the sources below use Ś and Ź for transliterating Ҫ and Ҙ.

http://gaspirali.kultur.gov.tr/default.asp?lehce_ID=10&strDil=English

http://www.gov.uk/government/uploads/system/uploads/attachment_data/file/320084/Bashkir_Romanization_11corr.pdf

The second table references the BGN/PCGN 2007 Agreement. It is a serious source. And it uses scientific transliteration, unusual for BGN/PCGN. I think we should switch to it. --Vahag (talk) 08:30, 25 December 2014 (UTC)

That system is generally fine but it has some of its own peculiarities, like the transliteration of ы as i (I think this might be an error, actually) and ә as ә. – Amateur55 (talk) 08:49, 25 December 2014 (UTC)

So you want a Wiktionary-specific mixture of established systems? I don't like that, but we have precedents: WT:RU TR. --Vahag (talk) 09:43, 25 December 2014 (UTC)

New official Latin alphabet

Kazakhstan is officially moving over to Latin to replace the Cyrillic alphabet. What are the steps we plan to undertake to move over to this new system? (A good argument to wait for some time is the possible volatility of the new alphabet: originally unveiled in 2017 and just earlier this month it got a major revision). I currently have a modified version of Module:kk-translit that uses the new alphabet under my user pages. SURJECTION _{·talk·contr·log·} 18:26, 28 February 2018 (UTC)

@Surjection: Wiktionary:Beer_parlour/2017/October#Kazakh_orthography --Per utramque cavernam (talk) 18:29, 28 February 2018 (UTC)

Thank you, I haven't noticed that thread. Some of it is outdated now, as there is a new version of the alphabet (that discontinues the use of apostrophes). However, the policy based on WT:ATTEST seems reasonable. SURJECTION _{·talk·contr·log·} 18:33, 28 February 2018 (UTC)

Module talk:kk-translit

I am proposing a change in the transliteration: see Module talk:kk-translit. --Anatoli T. ^{(обсудить}/^вклад) 00:45, 26 June 2019 (UTC)