Talk:Uyghur language/Archive 1

Turkish alphabet comparison
Is the chart meant to be comparing the Uyghur alphabet with the closest representable sounds in Turkish, diachronically determined correspondences, or just presenting a Turkish-based pan-Turkic system?

Some problems with the table, such as "ng" and the "?"s, could easily be solved once the answer to this question is decided.

—Firespeaker 20:18, 5 August 2005 (UTC)
 * I support this point: it should be clarified. Also the sentence another Roman script is used in Turkey and on the internet is not clear enough: does it say that another Roman script is used in Turkey to write Uyghur (and there are rules how to do that), or this is simply a comparison to the way another Turkic language&mdash;Turkish&mdash;uses Roman script.--Imz 18:06, 23 October 2005 (UTC)

Clarification required: "Arabic" vs. "Persian-Arabic" script
Is there a point in the following extract in using two different words (for the same thing, I assume? it was reintroduced, which means it is the same thing, doesn't it?)
 * The language traditionally used the Arabic script since the 10th century. The Chinese government introduced a Roman script in 1969, but the Persian-Arabic script was reintroduced in 1983, but with extra diacritics to distinguish all vowels of Uyghur.

(Also, two times but in a sentence is awkward, I think.) Could someone who understands the matter make it more clear please?--Imz 21:38, 23 October 2005 (UTC)

Native name
I've changed the "native name" in the language box from "ئۇيغۇرچه, 维吾尔语, Uyğurçe" to "ئۇيغۇرچه, Uyƣurqə", because "维吾尔语" is Chinese, not Uyghur; and "Uyğurçe" is a Turkish, not an Uyghur spelling. Today, Uyghur is officially written in an Arabic-Persian script and the Latin-based writing system that was used until the early 80s is supposedly still used for transcription purposes. —Babelfisch 01:04, 26 December 2005 (UTC)


 * Yes, I agress. However, I've re-added Uyğurçe. Even though this spelling is based on the Turkish alphabet, it is usual practice to add a transliteration of the native name in italics if that name is not in a Latin script. I feel that the presence of ƣ makes the official Latin script unusual enough for such a gloss. The Turkish-based transliteration will be familiar to anyone with any knowledge of Turkic languages, and will be easier for anyone with little knowledge of linguistics to interpret. --Gareth Hughes 13:35, 28 December 2005 (UTC)


 * I agree that the Uyghur Latin script is unusual, but using the Turkish alphabet to transliterate Uyghur may be imprudent because the Turkish alphabet doesn't distinguish between certain distinct phonemes in Uyghur, specifically ' & ', ' & ', and ' & '. This is, nonetheless, a useful tool for those familiar with the Latin scripts of other Turkic languages and should be included in the article, just perhaps not in the basic information box. That no further transliteration is provided in other cases of unusual Latin script, Azərbaycan dili for example, indicates that these transliterations are provided without the intention to facilitate pronunciation. —129.174.186.89 00:30, 8 January 2006 (UTC)


 * Yes, the Turkish orthography is lossy. The letter 'ƣ' is the difference: it is an unusual letter, and unusual enough to say that text containing it is not strictly in the Latin alphabet. The name of the language could be transliterated in IPA, but that would contain too much information, and be less readable to non-specialists. 'Azərbaycan' is easily understandable by most, even if 'ə' isn't understood as a mid-central vowel, and 'c' is /dʒ/ (we already have the fairly accurate English spelling 'Azerbaijan'). --Gareth Hughes 13:25, 9 January 2006 (UTC)


 * Some works that transliterate Uyghur do so with a modified Turkish orthography similar to that of Tatar, Crimean Tatar, and Azerbaijani. All the values are the same as Turkish except for the additional sounds not in Turkish.  These substitute ä or ə for Yengi Yezik̡ ə, x for h, q for k̡, ñ for ng, and w for v.  This is compatible with Turkish and elements of it are used on some Turkey-based Uyghur websites, especially the x and q. LuiKhuntek 07:38, 13 January 2006 (UTC)

Orthography confusion
I’m confused by the Latin orthographies of this language, or at least their presentation here. There seems to be at least three: the Chinese one, the Turkish one, and an Englishish-based random one used in the table of consonant phonemes, which also seems inconsistent with the IPA used in the presentation of the alphabets.

So does Uyghur have a glottal stop phoneme, and is it spelt q, implied by a word/syllable-initial vowel, or notated some other way? Does it it have a voiceless uvular stop phoneme, and is it spelt k, q or some other way? Does it have a voiceless palato-alveolar affricate, and is it spelt chinese-q & turkish-ç or some other way? Does it have an alveolar affricate, and is it spelt ch or some other way?

The orthography table does not present the letters h̡ and k̡, but by comparing the Turkish, Cyrillic and Chinese orthographies, I’m guessing they represent a glottal /h/ and a uvular stop /q/, in comparison with velar h /x/ and k /k/?

The text also appears to present ŋ as a letter of the alphabet, but the presentations always use ng. Which should it be?

Thanks!

—Felix the Cassowary 05:52, 23 September 2006 (UTC)


 * Hi Cassowary, this link http://www.xjyw.gov.cn/han/YWGZDT/wwesiyingwenzhuanxie.htm and this link http://ukij.org/teshwiq/UKY_Heqqide(KonaYeziq).htm may help you a little bit about that ...


 * Nice links that don't help the already clueless a bit. The transcription examples use the letter q in several scripts which does not appear in the table (of the article, and in one of the links). Turkish doesn't notate glottal stops, and according to the consonant table there are no uvular sounds. Much guesswork that shouldn't be there. --91.32.32.204 12:28, 3 August 2007 (UTC)
 * Taking a clue from other sites, the q letter stands for [tʃ] and is indeed in the table, although under the "Latin" label. This would mean it is used incorrectly in the examples. But I'll leave the editing to others who really know. --91.32.32.204 12:36, 3 August 2007 (UTC)

Uyghur Scouting
Can someone render Tayyar Bol (Be Prepared), the Scout Motto, into Uyghur script? Thanks! Chris 20:17, 11 August 2007 (UTC)

تاييار بول --68.239.70.248 18:01, 26 August 2007 (UTC)

Scripts
Is there any way we can improve the Arabic symbols? They seem really hard to read (to this non-Arabic reader). cf. http://www.uighurlanguage.com/logs/2005/01/alphabet_chart.php. Also it is my understanding that there are (at least) 3 scripts for Uyghur - Latin, Cyrillic and Arabic-based. Are the "Latin" letters in this chart actually the ones in their script? (again see that page, which has different symbols - I don't know which is right) And it would be cool if we could put the cyrillic characters in as well. pfctdayelise (translate?) 09:19, 28 May 2006 (UTC) I think old alphabet should also be included here, the one influenced Mongolian and Manchurian. And how they write Sogdian alphabet from up-to-bottom like ancient Chinese did. —Preceding unsigned comment added by 219.223.190.214 (talk) 14:44, 2 April 2008 (UTC)

Uyghur sound system
The consonant chart lists /f/ in its inventory. AFAIK, the majority of Uyghur dialects conflate /f/ > /p/, e.g. tarap (cf. tarāf) 'direction', pikir (cf. fikr) 'thought', wapa (cf. vafā) 'love'.

I've also noted that the rhoticism in Uyghur is more often than not regionally affected. In Qäshqär, conversations amongst native speakers were more often than not fully rhoticized, while those in Turpan tended to be non-rhoticized.

Furthermore, it distinguishes short and long vowels that respond differently to certain phonological processes, ...

Was this cited from Vaux?

AFAIK, al-maq /āl-maq/ 'to take' (which undergoes fronting) is also a "long vowel" environment, akin to ata /āta/ 'father'.

Also, kitap /kitāp/ 'book' undergoes vowel raising and the 'a' is a 'long a'.

It seem that vowel raising/neutralization lies primarily in metricality, rather than length, even though length often correlates, e.g. bala /bāla/ 'child', wapa /vapā/ 'love'.

Pachooey (talk) 20:32, 9 April 2008 (UTC)

Script displays disjointed
Hi, the Uyghur Arabic script is appearing disjointed (not joined up) on my screen, even though I do have a Uyghur font installed. Does anyone know how to make it display correctly? thanks. —Preceding unsigned comment added by 81.158.73.89 (talk) 11:27, 13 May 2008 (UTC)

An Introduction to Latin-Script Uyghur
Quite well done article about the Uyghur Latin script: http://www.uyghurdictionary.org/excerpts/An%20Introduction%20to%20LSU.pdf —Preceding unsigned comment added by 82.66.145.250 (talk) 22:46, 9 November 2008 (UTC)

Box
I've created a template for Uighur and Chinese names: Template:Uighur-Chinese-box. —Gregor Kneussel (talk) 05:20, 31 December 2008 (UTC)

Uyghur template
If you are a native speaker of Uyghur, then you can use this template:

--Amazonien (talk) 03:15, 20 January 2009 (UTC)

English pronunciation
"Many English speakers pronounce it as /ˈwiː.ɡər/ but the pronunciation /ujˈɡur/ is closer to native [ʔʊɪˈʁʊː]."

The way this is worded makes it into a bit of a non sequitur. [uj] is not a syllable that exists in English, so we might as well say that [ʔʊɪˈgur] is closer to the native pronunciation. I have sometimes suggested that English-speakers could say [ɔɪˈgur], but, for one thing, that has a pretty awkward cadence by English standards and would tend to be simplified to something easier to pronounce; and, for another, it doesn't really sound that much like the native pronunciation, so what's the point.&mdash;Nat Krause(Talk!·What have I done?) 18:11, 11 July 2009 (UTC)
 * Do we need a "closer English pronunciation" at all? Why not just give the pronunciation English speakers tend to use, /ˈwiː.ɡər/, and the native pronunciation? r ʨ anaɢ talk/contribs 18:21, 11 July 2009 (UTC)


 * If there were two competing English pronunciations; or if there was a feasible English pronunciation that was nearly identical to the native pronunciation, then it might make sense to include a "closer English pronunciation". Since this is not the case, I agree with your suggestion.&mdash;Nat Krause(Talk!·What have I done?) 19:42, 11 July 2009 (UTC)

variant pronuncation of "i"
According to the spelling correspondence table in Uyghur language, "i" (aka Uyghur-Arabic ئى and Cyrillic и) can be pronounced either as [i] or as [ɨ] (close central unrounded vowel), corresponding to the distinction between dotted-i and dotless-ı in the Turkish alphabet. The table in Uyghur alphabet says the same thing. Is this accurate? Why don't any of the Uyghur writing systems other than Turkish-alphabet-style make a distinction between the two?&mdash;Nat Krause(Talk!·What have I done?) 20:02, 11 July 2009 (UTC)
 * Because they're not phonemically contrastive in Uyghur (at least, not in the standard dialect). ئى is usually pronouned [ɨ]; it may sometimes be pronounced [i], but that difference never signals a change in meaning. Thus, there's no need to write the difference. r ʨ anaɢ talk/contribs 20:35, 11 July 2009 (UTC)

Pronunciation of "gh"
... as in Uyghur. Is it  or ? Or either? Or both?! The sound doesn't appear in the table of consonants. --NigelG (or Ndsg) | Talk 10:11, 5 August 2008 (UTC)


 * Don't count on being able to find reliable info on this. Field descriptions of languages are notoriously lax on this detail. (E.g. right now I'm working on the article Central Morocco Tamazight, which has exactly the same problem.) I think the reasons for it are that:
 * Languages rarely contrast /ʁ/ and /ɣ/
 * There is no clear-cut boundary between velar and uvular points of articulation (c.f. all the confusion in lots of other Turkic languages between /k/ and /q/)
 * /ʁ/ and /ɣ/ are probably acoustically pretty similar
 * I haven't looked into Uyghur specifically, so I really have no idea if this has been examined, but I wouldn't be too optimistic at this point. Mo-Al (talk) 22:02, 11 August 2009 (UTC)
 * People argue nonstop over whether it's /ʁ/ or /ɣ/. Hahn (1991) actually goes as far as to claim that they're both allophones of an underlying representation /G/. In any case, we shouldn't be taking a strong stance either way in the article; all we need to do is summarize the phenomena (specifically, the alternation between of gh-q-k-g in suffixes, governed by backness harmony and voicing). r ʨ anaɢ talk/contribs 22:45, 11 August 2009 (UTC)
 * I'm inclined to agree, though in a sense uvulars are often in the eye of the beholder, so to speak. I mean, one might almost be able to argue that Japanese /ko/ is realized as [qo], so perhaps it's better to view coarticulation as an inevitable and dismiss it from a phonological perspective except in languages where uvulars are contrastive. But I admit that my personal rant doesn't have a place in the article... Mo-Al (talk) 23:54, 11 August 2009 (UTC)

Uyghur Latin Yeziqi spelling
I think it's appropriate to use this script throughout the article- along with the Ereb/Kona yeziqi; this follows Uyghur Wikipedia conventions. Anyways, although v is technically part of the script, Google search results show preference for w, and dictionaries list loans with w, not v. Also, adjacent vowels do imply a glottal stop in between w/o needing an apostrophe. Mar de Sin  Speak up!  21:54, 10 August 2009 (UTC)


 * That's reasonable. I wasn't aware of the apostrophe thing (I only officially learned Uyghur Arabic script, not Latin script), but it makes sense. There are a very small number of Chinese loanwords with adjacent vowels and no hamza (for example, xua from Chinese 華) but I guess they're so rare that they don't pose a problem in latin script. r ʨ anaɢ talk/contribs 22:31, 10 August 2009 (UTC)


 * ULY is a pretty good system: user-friendly for the educated Anglophone; basically very precise at representing Uyghur phonemes; and largely apolitical. Any interest in us developing a Manual of Style (Uyghur) which specifies the use of ULY wherever precision with Uyghur language is desired? One occasionally sees Uyghur Pinyin and ad hoc spellings used instead (for along time, Rebiya Kadeer said "Uyghur: رابىيه قادىر, Rabiyä Qadir", which is a perfectly reasonable spelling, but not correct according to any system I know of. (Preferably, MOS should specify the variant of ULY that uses "zh" instead of "j" when the underlying Ereb is spelled that way).&mdash;Nat Krause(Talk!·What have I done?) 00:12, 12 August 2009 (UTC)

Which script to use?
Presumably this article needs a convention for which of the numerous Uyghur scripts to use for examples and such. But each seems to have a problem associated with it:
 * Arabic: inconvenience
 * Latin: controversial? not as widely used?
 * Cyrillic: only used in the former USSR as far as I know
 * IPA: not a good choice for a language with established writing systems?
 * some mixture: maybe Arabic with Latin or IPA transliteration

Any thoughts? Mo-Al (talk) 00:58, 12 August 2009 (UTC)


 * I think the default option should be Arabic + Uyghur Latin Yéziqi + IPA, all three. In any given particular case, concern for brevity might motivate us to include only one or two of the above. Of the three, Uyghur Latin Yéziqi is likely to be the most legible to our readers, so we should tend to include that when whole words are being referred to; for individual phonemes, IPA is better.&mdash;Nat Krause(Talk!·What have I done?) 01:11, 12 August 2009 (UTC)


 * Does the Arabic Uyghur script mark vowel length? It might be undesirable to have everything be in a defective orthography. If it does, then I think what you are suggesting is optimal as it reflects what is used in Xinjiang. (Though I'm unclear on whether ULY is actually official, or in use at all.) Mo-Al (talk) 01:34, 12 August 2009 (UTC)
 * No script for Uyghur marks vowel length. I think using the current orthographies is fine, since long vowels can basically be avoided in all sections but Sounds/Phonology. They're very limited, only in loans and developments from r-deletion.  And then IPA can be used when appropriate.  As an analogy, I don't think the Turkish language-related pages bother w/length (Turkish orthography got rid of the ^ diacritic to distinguish homographs I think.  Mar de Sin   Speak up!  01:58, 12 August 2009 (UTC)
 * Ah. I didn't realize vowel length was that unproductive. In that case, I guess both Arabic and Latin are equally acceptable. I would imagine that using the Latin script only would be controversial, though the article doesn't make it very clear how well ULY has been received among the community. Without any actual statistics I would imagine that Arabic is the most widely-used anyway, since the old Latin script wasn't popular and ULY is a recent introduction. Mo-Al (talk) 02:26, 12 August 2009 (UTC)
 * It would be nice to also give Arabic script when possible, as it is the most widely used in the Uyghur community. Latin script is good for readers; IPA is only necessary, I think, in the phoneme charts, or where a particular sound is being discussed in detail. As for vowel length, it is never contrastive in Uyghur so there's no reason to write it. r ʨ anaɢ talk/contribs 02:30, 12 August 2009 (UTC)


 * Mo-Al, I think Arabic is by far the most commonly-used script for Uyghur, at least among Uyghurs in China (which is most of them). So, Arabic is the native script, but Uyghur Latin Yéziqi is likely to be more useful to many of our readers. I agree with Rjanag as far as that. On the other hand, I think that it can sometimes be good to have IPA as well even for words: some readers will know IPA but not ULY, but readers who don't know IPA will have an easier time with ULY. As for acceptance of ULY among Uyghurs, the only information I have is from here, but that is presumbly written by a ULY booster. It seems like the most English-reader-friendly of the Romanisation systems, and it also might be the most popular in Xinjiang.&mdash;Nat Krause(Talk!·What have I done?) 03:40, 12 August 2009 (UTC)


 * The other major latin script is a Pinyin-based one. I imagine it's probably more common in Xinjiang than anywhere else (for example, all the people I know who use it are either from Xinjiang or learned Uyghur there). ULY is probably a better choice, both for the political reasons and because it's more intuitive (the other one, I think, only makes much sense to people who already know Pinyin). r ʨ anaɢ talk/contribs 03:50, 12 August 2009 (UTC)


 * My opinion is that, given that it's relatively faithful to Uyghur phonology, it would be better to use ULY than IPA. Given that the article contains a correspondence chart between the scripts and IPA, technically informed readers will be able to work it out. I think IPA really only makes sense for languages without faithful writing systems, or when discussing phonetics. (Also, the article is currently unclear on whether the older Chinese Latin Uyghur alphabet was modeled after Pinyin, the Soviet Latin Uyghur alphabet, or both to some degree. If anyone understands this better than me, I encourage them to clarify it in the article.) I think the Turkish language article (currently a featured article) should be our model in most respects, and note that they use the Latinate orthography despite it's failure to mark vowel length or distinguish between velar and palatal plosives. Mo-Al (talk) 04:02, 12 August 2009 (UTC)

Phonetic vowels
Can anyone find a source detailing what the Uyghur vowels are phonetically? I'd be inclined to believe that they are similar to Äynu, in which case the inventory would be best described as /a o u ɵ ʉ ɛ e i/. Mo-Al (talk) 00:10, 12 August 2009 (UTC)

Also, given that the pronunciation given for the language is [ʔʊɪˈʁʊrtʃɛ] and not [ʔuɪˈʁurtʃɛ], it would be nice to know more about the peripherality of short vowels, though this again may be one of those things which is unlikely to be documented in detail. Mo-Al (talk) 02:31, 12 August 2009 (UTC)


 * I'm going to try to work on the vowel chart soon; it has some issues that need addressed anyway. For example, [u] is usually more like [ʊ] and [i] like [ɨ]; also, [ɨ] and [e], while considered front vowels in usual IPA, do not act as front vowels in Uyghur (with respect to vowel harmony, they are "neutral". r ʨ anaɢ talk/contribs 02:30, 12 August 2009 (UTC)


 * When I added the IPA to the vowel chart, I basically used Turkish phonology as a precedent. If you want to use a more narrow transcription, then that's fine too, but I don't think it matters too much. Mar de Sin   Speak up!  03:08, 12 August 2009 (UTC)


 * Phonologically it doesn't matter, but phoneticians would be very interested in knowing whether Uyghur has phonetic rounded central vowels. Mo-Al (talk) 04:08, 12 August 2009 (UTC)

UDHR issue
The version of the UDHR given in this article doesn't quite match these two (e.g. xaş in the article vs. xas here). I understand the orthography may be in flux, or these may omit certain details, but at least these are sourced. Are there objections to replacing the two in the article with these? Mo-Al (talk) 22:47, 11 August 2009 (UTC)
 * The version you present is more official, so I'd prefer it. Mar de Sin   Speak up!  01:04, 12 August 2009 (UTC)
 * Hm, the omniglot version of the Latin UDHR conflicts with the one on unicode.org... Mo-Al (talk) 07:35, 12 August 2009 (UTC)

Consistency
The table in Uyghur language now gives IPA for Russian, Arabic, and Persian entries, Pinyin for Chinese words, and ULY for Uyghur words. All entries should have, in addition to the Arabic/Cyrillic/Hanzi native text, either r ʨ anaɢ talk/contribs 12:06, 12 August 2009 (UTC)
 * 1) Both IPA and that language's Latin orthography
 * 2) Only IPA
 * 3) only Latin orthography

Vowel issues
This edit has some errors. First of all, "defective" is not really a great way to describe the orthography, it's both linguistically questionable and slightly POV (although probably not the editor's POV, probably just Hahn's). Secondly, this is not a "defect" because the two vowels mentioned are not contrastive, and the notion that they 'play a role in vowel harmony' appears to just be Hahn's theory; there are plenty of vowel harmony accounts that never mention a /ɯ/. Unless some more context can be given and more explanation of what the deal is here, I think that whole aside should just be removed.

The bit "e only occurs in loanwords and as a result of vowel raising" is inaccurate. For pretty much the same reason as I gave in the discussion of [x] at Basawala's page; it occurs in many native words that may have originally been borrowed from Arabic or Persian hundreds of years ago, but by now it's fully assimilated into the phonology. Examples of words that have /e/ but are not recent loans and are not a result of vowel neutralization are /egiz/ "tall", /ejtmæk/ "to say", /jerɨm/ "half", and others. <b class="Unicode">r ʨ anaɢ</b> talk/contribs 03:28, 12 August 2009 (UTC)


 * The thing about /e/ must be mistaken then, though I would urge you to find a source that says that /e/ has been incorporated thoroughly into Uyghur phonology so that we can rectify this error in the article with a citation. Given that [i] and [ɯ] are apparently in complementary distribution, the "defective" comment shouldn't apply to it. However, I believe "defective" is the technically correct term for the orthographies' failure to mark vowel length, which apparently is phonemic (if never contrastive) in some loanwords. However if you think it sounds misleading then feel free to remove that word. What this article could really use would be examples of loanwords with long vowels that don't normally alternate with /r, l, j/. Mo-Al (talk) 04:07, 12 August 2009 (UTC)


 * The /e/ thing is not so much a mistake as a miscommunication over what is meant by "loanwords". The sources seem to use this a lot to refer to words with an Arabic or Persian etymology (the equivalent of a French or Latin etymology for tons of English words). But it's poor word choice, and repeating it here makes it sound like it's referring to 'new' words that were borrowed recently because Uyghur has no word for them (televisor, compyuter, mashina, doxtur, etc.). When you look past that, the fact that /e/ is an assimilated part of Uyghur phonology is quite uncontroversial.
 * As for the defective bit...well, no, to the best of my knowledge vowel length is not phonemic, it never signals a change in meaning. The only evidence for vowel length that I see given in the text right now is some words that don't undergo vowel neutralization if the vowel is "long"&mdash;but this is not a phoneme issue, it's a specific feature, or a specific environment, or what have you (keeping in mind that "long" vowels arise in particular environments, and thus are predictable). Same issue with the [i] and [ɯ] thing: these are not separate phonemes, except under Hahn's theoretical analysis. Hahn gives examples such as "til" and "bil" (/tɯl/ and /bil/, in his view), but these only show that the two are allophones; the sound represented by "i" in ULY in these words is pronounced differently in each word, but importantly there is no */til/ or */bɯl elsewhere in Uyghur, so these two sounds don't signal a meaning difference, they are predictable. Another telling point is that, at least in some speakers, [tɯl] "language" becomes [til] in some agglutinated forms, such as [tilʃʊnas] "linguist", further suggesting that [ɯ] and [i] are allophones.
 * When using sources like the ones you guys are using now, you need to remember that a lot of books like this, especially in matters of theoretical phonology and theoretical syntax, are speculative by nature&mdash;they're not merely summarizing observed phenomena, they're also proposing theories to account for the data. This is par for the course in linguistic texts, and they all need to be taken with a grain of salt, and when there's any doubt you have to attribute these theories to the author rather than reporting them as fact. Especially in cases like some of the stuff Hahn is doing, proposing whole extra phonemes (notice that his vowel chart has one vowel more than other vowel charts do) and abstract representations (/G/) it is important to recognize that these are just theories. <b class="Unicode">r ʨ anaɢ</b> talk/contribs 04:18, 12 August 2009 (UTC)


 * I defer to your knowledge on the issue of "loanwords" -- I was not aware of the issues in the use of that term in Uyghur. On the other hand, just because there is no minimal pair for two sounds doesn't mean that there isn't a phonemic difference. The biggest issue is whether the difference is predictable. If the occurence of [i] and [ɯ] can always be predicted independently of knowing what other words exist in the lexicon, then the difference is allophonic. In this case though, unless there is some good rule operating in the background which explains why you get [til] and [bɯl], there must be a phonemic difference, if a marginal one. C.f. Hindi-Urdu phonology: /æ/ and /ɛ/ are considered seperate phonemes, even though as far as I know there is no minimal pair distinguishing them. Thus from your point of view you could consider them "predictable", but it seems that even though they don't contrast, they could, in a sense, contrast. You make a good argument for /ɯ/ being marginal, but to be an allophone everywhere there must be a conditioning environment. Mo-Al (talk) 05:17, 12 August 2009 (UTC)


 * Similarly, if there are words with underlying long vowels that don't result from predictable morphological processes or vocalic assimilation, then they should also be considered phonemic differences. And I don't understand you statement that "this is not a phoneme issue, it's a specific feature, or a specific environment". 1. Vowel length is a perfectly acceptable "specific feature". 2. There must be a clear environment conditioning the allophony. As far as I can tell this is a lexical feature in such words as /pɛːr/ 'father'. 3. Uyghur hypercorrective r-insertion provides further evidence for the phonemicity of long vowels. Mo-Al (talk) 05:23, 12 August 2009 (UTC)
 * About features -- sounds can have different distinctive features without being separate phonemes. For example, the /t/ in English "stop" and "top": in traditional generative phonology, one is [-spread glottis] and the other is [+spread glottis], but they are the same phoneme.
 * About environments: there is no proof that any of these vowels are "underlying long", it's just a speculation that some writers created to account for a small portion of the data. In the example /pɛːr/ given here, it's equally likely that the UR is /pɛr/ and the vowel is pronounced a bit more long-ly because of the /r/ following it. Really, the "underlying long" thing, as far as I can tell, was introduced to account for the fact that some /a/s do not undergo reduction when they should: words like "ayal, dunya, bina, hava" with suffixes yield "ayali, dunyasi, binasi, havasi" instead of the expected "ayili, dunyisi, binisi, havisi". But there is no strong proof yet that this failure to alternate is due to "underlying longness" rather than the more common assumption (especially among native speakers) that there are just lexically-marked exceptions to rules such as vowel reduction and vowel harmony&mdash;in other words, the failure to alternate in environments like this isn't evidence in favor of underlying long vowels, it's just one of the reasons the vowel length hypothesis was presented. (Not to mention there are other issues behind vowel reduction that are ignored here; for example, in certain tenses /a/ never undergoes neutralization to /e/, but /æ/ does.) Again this article is presenting theories as fact, and in this case it's even misrepresenting the very source it was taken from: the Johansen book only says "there are arguments for a vowel length distinction" (p. 381), it doesn't say it's 100% convinced of them. In any case, like I said above, our article should only be reporting the facts, not repeating abstract theories as if they are fact. If you talk to a native speaker (even a linguistically-trained one) about long vs. short vowels in Uyghur, s/he will probably just look at you funny&mdash;long and short vowels are not separate phonemes in Uyghur, if anything they are just ancillary features that have some effect on the application of other rules. <b class="Unicode">r ʨ anaɢ</b> talk/contribs 05:35, 12 August 2009 (UTC)


 * (To clarify, I was confused about your comment on features because I don't think featural geometry is relevant here -- either vowel length is conditioned in all cases, or it has at least marginal phonemicity.)
 * First and foremost, we must find sources to back up your claims. If all of our sources state that vowel length is phonemic, then it would be OR to ignore it. I'd imagine Johansen is just collecting data from other sources since his book isn't Uyghur-specific, so the contents of his bibliography which deal with Uyghur would be extremely useful.
 * Also, to play devil's advocate, here are some arguments which I could see for phonemic vowel length: Whether native speakers notice it isn't that useful if it's marginal -- for instance, I am under the impression that Arabic speakers don't notice that /lˤ/ has marginal phonemic status in their language. (A more relevant example would be /k/ vs. /c/ for Turkish speakers -- I don't know whether they notice that one!) In addition, I suspect the loan words all originally had long vowels in their native languages, though I'd have to check. Also, this still hasn't fully explained hypercorrective r-insertion. Mo-Al (talk) 05:59, 12 August 2009 (UTC)

I understand your concerns, Rjanag, and I agree that we should write parts of this article more as phonological theory than fact. In my opinion, it's best to look at the consensus of scholarly sources on Uyghur phonology. The section in Johansen's book was written by Hahn, and he believes /i/ and /I/ are different phonemes. I also have another source (by Dwyer, Arienne) that I haven't cited or used yet that states the same thing. If a consensus of sources agrees that /i/ and /I/ are separate phonemes, then I think we should accept it, since I'm sure these scholars have more authority than us. Mar de Sin  Speak up!  06:48, 12 August 2009 (UTC)
 * A consensus of theoretical phonology sources might agree on that, but you won't find it in Uyghur textbooks or dictionaries, to the best of my knowledge. It's still best to present this as a theory (ie, "linguists have proposed that...") rather than an undisputed fact.
 * I don't have Dr. Dwyer's book, but I know the one you're talking about. I'll probably be seeing her sometime within the next two weeks, so I can talk about this some more. <b class="Unicode">r ʨ anaɢ</b> talk/contribs 12:01, 12 August 2009 (UTC)
 * I'm referring to the "Uyghur" chapter in Fact's about the world's languages . How will you be seeing her?  Mar de Sin   Speak up!  13:58, 12 August 2009 (UTC)

Other minor things
Things that might be nice to mention if possible, although I don't know if sources cover them:
 * About the hamza: while usually described as being a glottal stop, in word-medial positions in actual speech it generally isn't. In some words it's just a marker that two vowels are keeping to their own syllables rather than becoming a diphthong (compare mu'allim "teacher" to zhong hua "Chinese", the first is written with a hamza and the vowels stay separate&mdash;although there is not a glottal stop in between, at least in the speakers I know&mdash;where the first has no hamza and is pronounced as a diphthong, pretty much the same as the Chinese word it came from; another example is sa'et "clock", although I don't have a corresponding non-hamza example for it). The other thing hamza is used for that isn't necessary a glottal stop is marking unusual syllabification: for example, in sün'i "artificial", the main thing is that it's syllabified sün.i rather than the normal sü.ni. Of course, there's a good likelihood that the reason for this funny syllabification, and the reason for lack of diphthongization in the above examples, is that there used to be a glottal stop there and people simply don't pronounce it anymore.
 * Another example of [i]~[ɯ] alternation/complementary distribution: the suffix -lik/-liq/-luq/-lük (meaning "with" or "having", for example tem+lik = "with taste" i.e. "tasty"). "-liq" is really pronounced [lɯq], whereas "-lik" is pronounced [lɨk] (or [lɪk], if you prefer); the alternation is similar to in many other suffixes, "-liq" appears after back words (those with back vowels and/or consonants) and "-lik" before front words. (-luq and -lük are the versions of these that appear after round vowels, as in küchlük [kyʧlyk] "strength", but they're not really important to this discussion.) <b class="Unicode">r ʨ anaɢ</b> talk/contribs 16:22, 12 August 2009 (UTC)
 * I explained the glottal stop and usage of apostrophe in the 3 non-Arabic scripts under the chart in orthography. In words like muallim the glottal stop phoneme has to exist since Uyghur technically doesn't allow syllables without an onset (the CV(C)(C) structure) (according to Hahn and Ethnologue too I think), although this might be elided in casual/fast speech (I'm not sure how it works, but if you have a source that explains it more clearly please provide it).  Chinese loans are an exception to the orthography, and maybe we should explain it too, although I think they are minor as they are pretty much non-nativized.   Mar de Sin   Speak up!  17:26, 12 August 2009 (UTC)


 * Here is my best approximation right now of what they sound like (the first two of each word is the actual pronunciation, the second two is what it would be if it were really pronounced as a glottal stop). Note, again, that this does not mean glottal stop is irrelevant&mdash;as you suggested, it's possible that there really is underlyingly a glottal stop in these words, and it is just deleted through some phonological operations in everyday speech.
 * Also, that reminds me: if anyone is interested in having some recordings of Uyghur speech in this article, let me know, there are some native speakers that I see on a near-daily basis and can ask them to do a favor. I am a bit busy right now (and this article was only on my long-term to-do list, until things suddenly got a lot more active ;) ) but after 2 weeks or so I will probably have time to add some stuff. <b class="Unicode">r ʨ anaɢ</b> talk/contribs 17:52, 12 August 2009 (UTC)


 * Questions and comments:
 * What articulatory correlate is there of alternate syllabification in the word sün.i?
 * Do we have any idea how to explain [tɯl] and [bil]?
 * If you can get recordings of native speakers, it would tremendously improve the article. Mo-Al (talk) 18:37, 12 August 2009 (UTC)

I agree, it would be great to have recordings! I think two weeks is perfect, since we'll probably have more of an idea of which examples would need it the most, other than Uyghurche of course. And the bil vs. tïl definitely needs explanation; according to Hahn's description, bil and til fit the same condition and would be pronounced with [ɨ]. So there's no surface phonemic contrast, but an underlying difference? This really needs to be sorted out. Mar de Sin  Speak up!  21:01, 12 August 2009 (UTC)
 * Since I haven't read the books, I think I'm a bit behind: what evidence is there that shows that [i] and [ɯ] shouldn't be treated as in Turkish, where one could say they are usually in complementary distribution due to vowel harmony but are still separate phonemes? Also, are people arguing that vowel length exists on the surface, or is only underlying? Mo-Al (talk) 22:31, 12 August 2009 (UTC)
 * About the vowel length: no one disputes that it exists on the surface, because syllable-final /r/ and /l/, for example, often lengthens the preceding vowel (and the /r/, at least, often gets dropped after that). For example, تارتىپ tartip is often pronounced with a long /a/ and little or no /r/. This, however, is an allophonic distinction, not an underlying one. The claim of underlying vowel length contrasts is a speculative one. <b class="Unicode">r ʨ anaɢ</b> talk/contribs 02:09, 13 August 2009 (UTC)
 * I was thinking more about surface long vowels in words without /r/ in careful speech. Mo-Al (talk) 07:52, 14 August 2009 (UTC)
 * Don't think any such words exist, to the best of my knowledge. The only argument I've seen so far for long vowels is as an abstract explanation for words that don't undergo vowel neutralization (ie, dunya, ayal, and weqe, for example, become dunyasi, ayali, weqesi, rather than dunyisi, ayili, weqisi), which people could just learn lexically as exceptions. <b class="Unicode">r ʨ anaɢ</b> talk/contribs 12:42, 14 August 2009 (UTC)
 * Vaux makes an argument, which I haven't read fully but find pretty convincing. She argues that /i/ is either neutral or marked as [-back]; I'm not sure if that would make them separate phonemes.  Hahn says VL is definitely underlying, but only pronounced long optionally/in certain conditions, but he admits lack of research into these conditions.  Dwyer says that VL occurs phonemically only in colloquial speech and in loans; I think there is agreement that VL is in fact phonemic.   Mar de Sin   Speak up!  01:41, 13 August 2009 (UTC)
 * I don't think two sources are enough to claim "widespread agreement" about vowel length, especially if one of them is specifically saying that vowel length is not phonemic in Standard Uyghur.
 * Well can loanwords have surface long vowels? If there are words with surface long vowels that don't occur in a predictable phonetic environment and don't undergo r-insertion in non-hypercorrective formal speech, then phonemic vowel length is very likely. And regardless, we can't just pick and chose sources. Us filtering our sources to fit whatever theory we prefer is a form of OR in and of itself. Mo-Al (talk) 04:22, 13 August 2009 (UTC)
 * As for [ɯ] and [ɪ/ɨ], it's not really my place here to try to prove or disprove any theories; I am just trying to stress that these things are still just theories and should not be presented as fact. Obviously the phenomenon warrants mention in the article. I am just concerned about the way it is presented, and want to make sure none of these sources are eaten up without the usual healthy dose of skepticism. In short, we can verify through reliable sources that people have proposed theories about these sounds; what we cannot verify through reliable sources is whether any of these theories are right or wrong. <b class="Unicode">r ʨ anaɢ</b> talk/contribs 02:15, 13 August 2009 (UTC)
 * Well my issue is that given that Uyghur has vowel harmony we'd expect them to be in almost complementary distribution anyway. But given that we have what appears to be a near-minimal pair, I think the burden of proof is on the allophony side now. Besides, any phonological inventory of a language is a theory -- so either we accept the theory of lumping them together or the theory of having them both be phonemes. Mo-Al (talk)
 * I'll try to get some time for a longer response later. The thing about Uyghur vowel harmony, which I've been meaning to mention for a while but have kept forgetting, is that it's frequently broken (could be as many as 50% of words, not sure, don't have a corpus). Old borrowed words, such as sa'et, often break vowel harmony; this makes it impossible to determine phoneme or allophone status on the basis of harmony alone. For example, within the ke/ge/qa/gha suffix alternation, g and gh are in perfect complementary distribution, and they often are in words as well (ie, q appears in words that have a's, o's, and u's, whereas k appears in words with the front vowels). But once you start looking at exceptions (mostly old borrowings, such as heqiqet, kalivet, and pakiz), you see that there really must be a phonemic distinction between these in spite of frequent alternation (for example, in near-minimal pairs such as paqa/pakar, and actual minimal pairs such as qoy/koy, although koy is a recent borrowing). This isn't really a point in favor of one claim or the other as far as [ɯ] and [ɪ/ɨ] are concerned, it's just something to keep in mind. <b class="Unicode">r ʨ anaɢ</b> talk/contribs 11:57, 13 August 2009 (UTC)
 * By the way: the Vaux article you linked looks interesting. I've only had time to skim it right now, but it looks like it's not so much about URs of ى, but more about whether ا and ە neutralized to ى retain their frontness/backness or become neutral. So I'm not sure how useful it will be for this particular discussion, although as a source of information on some closely related issues it should be great. <b class="Unicode">r ʨ anaɢ</b> talk/contribs 02:24, 13 August 2009 (UTC)

Mo-Al, I assumed that Dwyer implied vowel length was indeed phonemic: the loans she says they occur in are nativized. So these loans would be where the vowel length in "careful" speech occurs, if allophones from phonological processes (r-deletion, etc) are discounted. Rjanag, this is where the k/q = 1 phoneme clashes with the i/ɯ = 1 phoneme assumption: the coexistence of native ki and qi: kir- "enter", and qir- "scrape, shave". If I'm mistaken and one of them isn't native, then kim and qiz are. Mar de Sin  Speak up!  06:10, 14 August 2009 (UTC)
 * Like I said above, I think, k/q are not 1 phoneme&mdash;they seem that way because of sets such as ke/ge/ka/gha, but because of the huge number of exceptions they are often contrastive nowadays. As for kim and qiz, kir and qir, they're not really damning either way because if you assume that ى is a "neutral" vowel that surfaces as [ɯ] in some places and as [ɪ/ɨ] in others, words like kim and qiz are consistent with that. The question of whether the contrast lies in /k/-/q/ or in unwritten /ɯ/-/ɪ/ (and then shifts contrast to the other sound as well) is what's interesting, and it brings up questions of which sounds are native, etc. As far as I can tell further research is still required, there's been no totally satisfactory answer yet. Personally, for me the most powerful support for the "i is two phonemes" argument is til and bil, although that could be explained by learned exceptions or by some other features we haven't considered yet (laxness of the tongue or something, I don't know), but in all this hasn't been studied much. <b class="Unicode">r ʨ anaɢ</b> talk/contribs 12:42, 14 August 2009 (UTC)
 * Well the whole thing about "features" and "learned exceptions" has no bearing on phonemicity here. Either the environment conditions it or there is something phonemic which we haven't accounted for.Mo-Al (talk) 18:26, 14 August 2009 (UTC)


 * (out) Another interesting thing: I just spoke with a Uyghur friend (grew up in Kashgar and Urumchi) who said she pronounces til and bil the same. On the other hand, I've heard them pronounced differently (more [ɯ] and [ɪ]) from some Uyghurs with different dialect backgrounds. This, of course, is not evidence for or against anything (since there are sill other word pairs that can be used both in support of or against the /ɯ/-/ɪ/) argument; it's just an observation. <b class="Unicode">r ʨ anaɢ</b> talk/contribs 02:41, 17 August 2009 (UTC)

/w/?
Duval (2006:10) implies that what is transcribed in this article as /w/ would be more appropriately /β̞ ~ ʋ/. Is this true? Mo-Al (talk) 23:58, 12 August 2009 (UTC)
 * Hahn says that the <w> which represents /b/ allophonically (kawap, xewer, aptotowuz: intervocalic, although I think there are exceptions) is in fact [β̞ ~ ʋ], but /w/ itself is [w ~ ʋ]. Mar de Sin   Speak up!  01:51, 13 August 2009 (UTC)
 * So in some cases there is underspecification/neutralization of the difference between /b/ and /w/? Mo-Al (talk) 04:41, 13 August 2009 (UTC)
 * How would underspecification apply in this case? I'm not sure about neutralization; I think (based on Hahn) for some speakers lenited /b/ is different from /w/ in all conditions.  Mar de Sin   Speak up!  06:15, 14 August 2009 (UTC)
 * Ok, I don't think underspecification was the right word. But yeah, the question was whether some speakers pronounce /w/ and /b/ similarly in certain environments. Mo-Al (talk) 07:46, 14 August 2009 (UTC)
 * Some do, and this is sometimes reflected in the orthography as well (spelling words with v/ۋ instead of b/ب). <b class="Unicode">r ʨ anaɢ</b> talk/contribs 03:47, 26 August 2009 (UTC)

Uyghur grammar
I see that the expansion of Uyghur grammar is on the todo-list. I have created a new article named Uyghur grammar, where I think all detailed information on Uyghur grammar should be found. I have already added some information on demonstrative pronouns in Uyghur, but there is still a long way to go. I will keep adding new information when time allows me to do so - and it would be great if others would do the same. The section on grammar in the main article, Uyghur language, should still be expanded though. -- Llonydd (talk) 20:34, 7 February 2010 (UTC)
 * I have been planning on expanding the grammar session for a while- now that you've brought it up, I'll try to add content in my free time.  Bʌsʌwʌʟʌ   Speak up!  00:33, 8 February 2010 (UTC)

CVCC syllable structure
My rationale for this edit: I have lost my copy of Hahn (1991) so I can't check what it says, but I don't think it's true that CVCC occurs only when the third element is sonorant; some counterexamples are meshq (مەشق, "exercise/practice"), herp (ھەرپ, "letter/character"), and sirt (سىرت, "outside"). The reason I added "for some speakers" is that this varies by dialect (at least for the words I am thinking of), with many Ürümchi speakers I think adding an epenthetic vowel in the middle. For example, meshq is pronounced as [mæʃq] by some speakers and [mæʃəq] by others; likewise for gherb (غەرب, "west"), which is [ɣæɹb] for some people and [ɣæɹəb] or others. <b class="IPA">r ʨ anaɢ</b> (talk) 14:12, 3 February 2011 (UTC)
 * Wouldn't /r/ be considered a sonorant in Uyghur? Anyways, I checked Hahn, and he doesn't actually stipulate anything for syllable coda CC's.  However, he does say that something is almost always epenthesized or deleted when CC precedes another consonant or nothing.  Bʌsʌwʌʟʌ   Speak up!  19:10, 3 February 2011 (UTC)
 * Ah, you are right, I miscounted (was thinking "third element" referred to the third consonant, or last element). meshq, though, is still a counterexample (as far as I can tell). I guess we all agree that epenthesis/deletion happens, and the possibly controversial point is whether it rules out CVCC structure (e.g., if CVCC is underlying and the epenthetic vowel is added by rule to make a surface CVCVC, does that mean the language does or doesn't allow CVCC)? <b class="IPA">r ʨ anaɢ</b> (talk) 20:30, 3 February 2011 (UTC)

Relationship with Russia and Russians
Some statements about relationships of Uyghur with Russia, the Russian language and and the Russian people. Some of them are unreferenced or dubious:
 * Today the Uyghur language is also used as a lingua franca among non-Uyghurs, such as the Xibes, Wakhis, Tajiks of Xinjiang and Daurs, and even some Russians - this urgently needs referencing and should be removed soon if no source is found.
 * Russian sources cite the central dialect of Ghulja as the pronunciation norm for modern Standard Uyghur. - which ones?
 * The inclusion of the article in the Languages of Russia category - it's probably wrong. I removed it. --Amir E. Aharoni (talk) 18:44, 11 February 2011 (UTC)
 * The statement has been sourced. That sentence and the next had the same source, which was cited at the end of the two.   ʙʌsʌwʌʟʌ   spik ʌp!  23:43, 3 March 2011 (UTC)

Bad title
The "l" of language in the title should be a capitalised "L". As it is this page is inconsistent with all other language pages. 187.59.242.122 (talk) 13:58, 8 December 2011 (UTC)

Map "Geographical extent"
I have reverted two recent additions of a map "Geographical extent of Uyghur in China and Kazakhstan" for the following reasons:

1) The map does not say which criterias were considered to define the "Geogaphical extent of Uyghur" language. Does it represent areas with a significant percentage of Uyghur speakers (e.g. more than 10%), or an area where Uyghur people are more represented than other ethnic groups, or an area where Uyghur people do represent a majority? I would understand "geographical extent" rather in the first sense, meaning areas populated by a significant percentage of Uyghur speakers, but the wording is ambiguous.

2)The map is incomplete. A lot of Uyghur speaking areas are missing from the map, as I already mentioned during my first revert. Even with the most restrictive definition of "geographical extent of Uyghur language", some predominantly Uyghur speaking areas can be found as far east as in the Kumul/Hami Prefecture.

3) Unsourced. No references are given, for both China and Kazakhstan.--Pseudois (talk) 03:35, 10 February 2012 (UTC)


 * As was mentioned in the summary, the first version of the map was based on the map for UCLA's Language Materials Project, which lists their references. Since you wanted mention of the "Uyghur speaking oasis around the Taklamakan", I supplemented the map with information from a 1990 CIA source. Note that most language maps used on Wikipedia don't even list a source! I don't know the exact methodology of these maps, but we can infer that the area colored in is where Uyghur people and language predominate. By the census figures (and historically, too) Uyghur-speakers are concentrated in southwestern Xinjiang. Since the Chinese economic reform, Uyghur migrant workers have flocked to Urumqi, a historically Mongolian area developed by Han Chinese, but they are both recent migrants and only about 10% of the population, like in Kumul/Hami Prefecture. I'm sure we could mark Beijing, Shanghai, and other major cities as areas where Uyghur-speaking people have migrated, as with most other languages, but then that's not really a useful map then, is it? Since your second revert, against my request, was based on a misunderstanding ("unsourced"), I hope you won't rush to revert if I restore it again. Shrigley (talk) 04:12, 10 February 2012 (UTC)
 * Thanks for providing these references, point 3) is now solved. Points 1 and 2 remain open. You will find more information on the same UCLA page you gave as reference: for example example the Eastern dialects around Lopnor are not included in the map you submitted. I hope you won't rush to restore an incomplete map. Regards,--Pseudois (talk) 04:55, 10 February 2012 (UTC)
 * Both the UCLA and CIA maps mark Lop Nur, part of a Mongol Autonomous Prefecture by the way, not as predominantly Uyghur-speaking but as Mandarin-speaking. Just because some kind of dialect developed in that area does not mean that it's numerically significant. I could possibly mark peripheral non-native areas like those with some lighter shade, but I would need either a map source or a text source that describes the geographical area well.


 * But again, marking all areas with some Uyghur speakers is not the purpose of this map. Maps of the extent of Mandarin always look like this, rather than covering the whole of China, although there are certainly Mandarin speakers in every single city of China. This is about where the language predominates, not where the people migrate. As it stands, the map is useful, and not incomplete, as it matches comparable academic and governmental maps. It might not be perfect, but nothing on Wikipedia is, and we can't be paralyzed by vague objections. If you want specific areas marked, then make clear, actionable requests with good reasoning and sources, and they can be added on-the-fly. Shrigley (talk) 20:11, 10 February 2012 (UTC)
 * My remarks were not "vague", but very clear and precise. First you should make it clear which criteria you wish to consider in order to define the "Geogaphical extent of Uyghur" (as the map was labelled). Maps can vary a lot, have a look for example at this one. It sounds a bit confusing when you write "this is about where the language predominates, not where the people migrate". It can often happen that a language predominates precisely because of migration. A clear defintion is needed. And, as I wrote, regardless of which criteria you use, your map is icomplete: just to give an example, Tufan has a large Uyghur majority (at least as per official statistics), but it is not included in both versions of your map. I know that WP is far from being perfect, but my position is that it is better to remove wrong information rather than keeping it. Regards,--Pseudois (talk) 10:30, 11 February 2012 (UTC)


 * The CIA map suggested by user:Jiang (https://www.cia.gov/library/publications/cia-maps-publications/maps/791363.jpg) looks reasonably accurate and includes the various areas I have mentioned in my previous comments. It might be reasonable to include the northern areas labelled "sparsely populated" in your geographical extent of Uyghur, as the western area is fully and the eastern area almost fully surrounded by Uyghur areas. The "sparsely populated areas to the south should not be added, as they are predominantly Tibetan (at least the southern part of it). I think the issue is now solved.--Pseudois (talk) 09:51, 12 February 2012 (UTC)


 * The northeastern sparse area is the Hui prefecture, and is surrounded by Kazakhs. The eastern area seems to be the Mongol grasslands. There are no problems with the western sparse area, but I've made it a slightly lighter shade of red to indicate the sparse population. We should be careful about conflating Uyghur people and Uyghur language. There are significant numbers of Uyghur, especially in eastern Xinjiang, who choose to be educated and to speak in Chinese for career advancement. Anyway, the map is now updated and redrawn after the 2010 CIA map. Cheers! Shrigley (talk) 15:29, 12 February 2012 (UTC)

Minority education for the uyghur
The chinese government allows ethnic minorities like the uyghur to attend either uyghur or chinese schools. At the uyghur school, thry are allowed to observe islamic religious practices such as ramadan while forbidding it at chinese schools.

http://www.globalpost.com/dispatch/china/100928/uighurs-border-culture-islam

Kathleen E. McLaughlinOctober 26, 2010 13:52Updated October 31, 2010 06:33

Borderland: All's not quiet on China's western front

"They attend Chinese high school rather than Uighur school, hoping for an advantage in future studies and jobs. In making that choice, they lost the right to wear their head coverings and celebrate their religious holidays at school. Though they could ask for these things, their teachers don’t stop the clock on studies when the five-day Eid holiday rolls around"

http://books.google.com/books?id=DMU8Ue0HECcC&pg=PA127#v=onepage&q&f=false

http://books.google.com/books?id=4q53niG5MLIC&pg=PA107#v=onepage&q&f=false

http://books.google.com/books?id=Jl_Zw9QzvxEC&pg=PA135#v=onepage&q&f=false

http://books.google.com/books?id=qlZQWWlPGssC&pg=PA262#v=onepage&q&f=false

A class of assimilated uyghurs who speak chinese are known as "chinese uighur".

http://books.google.com/books?id=yT8Yc994CuUC&pg=PA5#v=onepage&q&f=false

http://books.google.com/books?id=yT8Yc994CuUC&pg=PA52#v=onepage&q&f=false

The chinese government encourages uyghurs to worship and practice islam.

http://books.google.com/books?id=yT8Yc994CuUC&pg=PA54#v=onepage&q&f=false

Rajmaan (talk) 02:50, 27 January 2013 (UTC)

Chinese loanwords in Turki (uyghur)
please put these primary sources into wikisource and link them to this article.

http://books.google.com/books?id=_Is4AAAAYAAJ&pg=PA577#v=onepage&q&f=false

http://books.google.com/books?id=uZ8LAAAAIAAJ&pg=PA577#v=onepage&q&f=false

Rajmaan (talk) 02:55, 3 February 2013 (UTC)

Chagatai?
Besides just calling their language Turki, some called it Chagatai

http://books.google.com/books?id=9qJmAFvg-80C&pg=PA4663#v=onepage&q&f=false

http://za.ebid.net/for-sale/c1920-turkistan-mother-child-yarkand-river-kashgar-lady-35528374.htm

The Qing government's 5 language polygot dictionaries have been translated as either chinese, manchu, mongol, tibetan, and turki, or chinese, manchu, mongol, tibetan, and chagatai.

http://books.google.com/books?id=CaoXrIqxsJ4C&pg=PA39#v=onepage&q&f=false

The correct name of the dictionary is 五體清文鑑. It was "mispelled" as 五體清文監 in the above book. The chinese wikipedia article is at 御製五體清文鑑

http://books.google.com/books?id=NYFBtTUZFxEC&pg=PA396#v=onepage&q&f=false

http://books.google.com/books?id=NYFBtTUZFxEC&pg=PA397#v=onepage&q&f=false

http://books.google.com/books?id=NYFBtTUZFxEC&pg=PA398#v=onepage&q&f=false

http://www.chinaknowledge.de/Literature/Science/qingwenjian.html

http://www.dictall.com/indu/339/33892396291.htm

http://riccilibrary.usfca.edu/view.aspx?catalogID=1593

http://nccur.lib.nccu.edu.tw/bitstream/140.119/33598/10/53502110.pdf

Images here

http://abkai.net/blog/御制五体清文鉴/

http://abkai.net/blog/wp-content/uploads/2012/01/yzwtqwj.png

http://www.konan-u.ac.jp/~bun/rekibun/zemi/2007/02/index.html

http://www.konan-u.ac.jp/~bun/rekibun/zemi/2007/02/data01b.jpg

http://fansiren.blogspot.com/2010/02/80-8012-1957200-300-122007-unixunix-20.html?m=1

http://203.68.243.199/cpedia/Content.asp?ID=16427&Image=%BBy%A8%A5%A1E%A4%E5%A6r&Query=5

http://talkiyanhoninjai.com/2010/10/25/故宮博物院編『同文之盛――清宮蔵民族語文辞典/

http://tc.wangchao.net.cn/baike/detail_654150.html

http://abkai.net/blog/御制五体清文鉴/

http://auction.artxun.com/paimai-67494-337466482.shtml

http://fubito.blogspot.com/2010/04/az-otnyelvu-mandzsu-tibeti-mongol-ujgur.html?m=1

http://www.cnrepair.com/html/c275/200908/07/316715.htm

Manchu chinese version only

http://catalog.digitalarchives.tw/item/00/12/7c/e1.html

http://www.library.pref.osaka.jp/nakato/shotenji/74-11.html

http://www.library.pref.osaka.jp/nakato/shotenji/74_gai.html

This is a forum and not a source, however, the chinese characters posted are liftec from the dictionary so it can be used to search other sources for the text

http://www.manjusa.com/thread-985-1-1.html

I am looking for a full length online edition of the book, to see the exact name the Qing government used for the uyghur language.

Chinese wikipedia mentions it as the chagatai language or muslim language.

Some Qing dynasty texts mention chagatai language, these are to be put into wikisource

http://books.google.com/books?id=8CsrAAAAYAAJ&pg=PT57&dq=察合台&hl=en&sa=X&ei=6OINUdCoDcex0QGM_YC4DQ&ved=0CDAQ6AEwAA

http://books.google.com/books?id=S1MrAAAAYAAJ&pg=PA153&dq=察合台&hl=en&sa=X&ei=6OINUdCoDcex0QGM_YC4DQ&ved=0CDYQ6AEwAQ

http://books.google.com/books?id=ejUrAAAAYAAJ&pg=PT1092&dq=察合台&hl=en&sa=X&ei=6OINUdCoDcex0QGM_YC4DQ&ved=0CDwQ6AEwAg

http://books.google.com/books?id=k6kpAAAAYAAJ&pg=RA3-PA78&dq=察合台&hl=en&sa=X&ei=6OINUdCoDcex0QGM_YC4DQ&ved=0CEMQ6AEwAw#v=onepage&q=察合台&f=false

http://books.google.com/books?id=rEwrAAAAYAAJ&pg=RA2-PA9&dq=察合台&hl=en&sa=X&ei=6OINUdCoDcex0QGM_YC4DQ&ved=0CEkQ6AEwBA#v=onepage&q=察合台&f=false

http://books.google.com/books?id=ojorAAAAYAAJ&pg=RA1-PA8&dq=察合台&hl=en&sa=X&ei=6OINUdCoDcex0QGM_YC4DQ&ved=0CE8Q6AEwBQ#v=onepage&q=察合台&f=false

http://books.google.com/books?id=mlAsAAAAYAAJ&pg=RA10-PA68&dq=察合台&hl=en&sa=X&ei=6OINUdCoDcex0QGM_YC4DQ&ved=0CFsQ6AEwBw#v=onepage&q=察合台&f=false

A japanese text

http://books.google.com/books?id=csEtAAAAYAAJ&pg=PA94&dq=察合台&hl=en&sa=X&ei=6OINUdCoDcex0QGM_YC4DQ&ved=0CFUQ6AEwBg#v=onepage&q=察合台&f=false

More sources on the polgygot dictionary

Official languages of qing china, chinese, manchu, mongolian, tibetan, and turki (modern uyghur)

http://books.google.com/books?id=BMftUQCm5_IC&pg=RA1-PA97&dq=paper+manchu+turki+chinese+tibetan+mongol&hl=en&sa=X&ei=3727UP1Z6brQAYGJgJgP&ved=0CC8Q6AEwAA#v=onepage&q=paper%20manchu%20turki%20chinese%20tibetan%20mongol&f=false

http://books.google.com/books?id=mCS1940T5MkC&pg=PA480&dq=paper+manchu+turki+chinese+tibetan+mongol&hl=en&sa=X&ei=4L27UIrNCIf00gHY_YCwAg&ved=0CDUQ6AEwAQ

Page ix

"Si Kiang Basin — Other Rivers People and Languages — People Historical — Languages, Chineso Mongol — Manchu. . Turki — Tibetan — Aboriginal Languages CHAPTER II— CLIMATE OF CHINA Air Masses acting over China Typhoons ... The Army— Mongols and Manchou- kuo — Soviet-Outer Mongolian Protocol Chinese Protest — Soviet Reply Second ... Sugar— Chemicals and Pharmaceuticals Petroleum Products Paper — Timber Exports Silk Cotton 22 30 31 32 33 3H 57 ..."

http://books.google.com/books?ei=4L27UIrNCIf00gHY_YCwAg&id=JwwfAQAAMAAJ&dq=paper+manchu+turki+chinese+tibetan+mongol&q=paper+manchu+turki+tibetan+mongol#search_anchor

Page 10

"few gaps) Kasgari's Diwan Lugdti'l-Turk to compile a list of the basic words in Turkish about thousand years ago. ... Pentaglott", a dictionary of Manchu with translations into Tibetan, Mongolian, the Turki of Chinese Turkestan and Chinese . ... It is unnecessary to include in this paper more than 10 SIR GERARD CLAUSON."

http://books.google.com/books?ei=4L27UIrNCIf00gHY_YCwAg&id=APkXAAAAIAAJ&dq=paper+manchu+turki+chinese+tibetan+mongol&q=paper+turki+tibetan+mongol#search_anchor

Page 32

"essays & notes on Tibet & the neighbouring countries Nirmal Chandra Sinha ... Thus the Manchu, the Mongol and even the Turki ( Uighur ) had to accept Chinese language and script for varying periods to varying degrees and the ... The Tibetan book, though made of paper, did not 32 How Chinese was China's Tibet Region."

http://books.google.com/books?ei=4L27UIrNCIf00gHY_YCwAg&id=VdBwAAAAMAAJ&dq=paper+manchu+turki+chinese+tibetan+mongol&q=paper+manchu+turki+tibetan+mongol#search_anchor

Page 83

"InTibjt particular works are sometimes printed, for luxury or ritual use, in gold or silver ink on special paper with ... obtained from Central Asia, with the text in five or six languages, Sanskrit, Tibetan, Chinese, Mongol, Manchu, and perhaps Turki."

http://books.google.com/books?id=WKMtAAAAMAAJ&q=paper+manchu+turki+chinese+tibetan+mongol&dq=paper+manchu+turki+chinese+tibetan+mongol&hl=en&sa=X&ei=4L27UIrNCIf00gHY_YCwAg&ved=0CEcQ6AEwBw

Page xli

"The Chinese have developed an ingenious system of taking exact facsimiles of inscriptions by means of inked paper ... to Chinese, but there are many in other languages like Manchu, Tibetan, Mongol, also in Si-hsia, Arabic, Turki, and ..."

http://books.google.com/books?ei=4L27UIrNCIf00gHY_YCwAg&id=GKe0AAAAIAAJ&dq=paper+manchu+turki+chinese+tibetan+mongol&q=+manchu+turki+tibetan+mongol#search_anchor

Page 621

"selected papers on the art, folklore, history, linguistics, and prehistory of sciences in China and Tibet Berthold Laufer, Hartmut Walravens. the equestrian ball-game . Both the ... 272), corresponding to Manchu mumuhu, Mongol bumbuge (the pronunciation is thus fixed by the Manchu transcription in the Pentaglot edition), aud eastern Turki lob \Jyi (transcribed lob in Manchu). The latter word is listed by ..."

http://books.google.com/books?id=hX4bAAAAIAAJ&q=paper+manchu+turki+chinese+tibetan+mongol&dq=paper+manchu+turki+chinese+tibetan+mongol&hl=en&sa=X&ei=Hb-7UI7OG6e30gH11IHIBw&ved=0CDsQ6AEwBDgK

Page 7

"background papers and proceedings Oliver Edmund Clubb, Eustace Seligman, Lyman M. Tondel Association of the Bar of ... of a Chinese Federated Republic by the unification of China Proper, Mongolia, Tibet and Sinkiang into a free federation. ... are many Turki and Mongol groups that then resided, and still reside, outside the formal boundaries of the Chinese state. ... that Republican Chinese traditionally have considered that any vassal that had been wrested from the Manchu rule, ..."

http://books.google.com/books?id=IugsAAAAMAAJ&q=paper+manchu+turki+chinese+tibetan+mongol&dq=paper+manchu+turki+chinese+tibetan+mongol&hl=en&sa=X&ei=Hb-7UI7OG6e30gH11IHIBw&ved=0CEEQ6AEwBjgK

Page 2

"In keeping with the orientalist traditions in this country, most space is given to Turcology, with Mongolian linguistics and philology a close second. Among Altaic studies, Tungus-Manchu philology and ethnography are represented and some articles have been devoted to ... Also fairly well represented is Semitic philology, and there are papers in Sinology, Iranology and Indology, as well as ancient ... One article discusses the Salar language spoken in the Chinese People's Republic."

http://books.google.com/books?id=DqlMAQAAIAAJ&q=paper+manchu+turki+chinese+tibetan+mongol&dq=paper+manchu+turki+chinese+tibetan+mongol&hl=en&sa=X&ei=Hb-7UI7OG6e30gH11IHIBw&ved=0CD4Q6AEwBTgK

Page 131

"The Manchu ft is the same Chinese word but from a later period. ... «pencil» is of Russian origin, where it supposedly has a Turkish etymology ,59 sambara « writing tablet* is a Tibetan loanword.*0 The remaining words are of Mongolian origin. ... «paper» a derivative from the obsolate stem *(ä- bir-ün bariyói (К.) «pain ter» — T pir thogs-pa ; bir bariyéi (SD II 19) «id. ... M bir-tür beke silemede- (SD II 19); bir turki- (SD I 1229) «to smear, draw with a brush soaked in ink» = T snag ris rgyag, ..."

http://books.google.com/books?id=D9MTAQAAMAAJ&q=paper+manchu+turki+chinese+tibetan+mongol&dq=paper+manchu+turki+chinese+tibetan+mongol&hl=en&sa=X&ei=Hb-7UI7OG6e30gH11IHIBw&ved=0CEQQ6AEwBzgK

More sources on the dictionary

please put these primary sources into wikisource and link them to this article.

http://books.google.com/books?id=_Is4AAAAYAAJ&pg=PA578#v=onepage&q&f=false

http://books.google.com/books?id=uZ8LAAAAIAAJ&pg=PA578#v=onepage&q&f=false

Rajmaan (talk) 04:13, 3 February 2013 (UTC)

Pronunciation
Regarding this edit, I don't want to jump in but I did want to add my two cents. [ʔʊjˈʁʊr] is the correct pronunciation of the ethnonym, but in the language the word they use for the language itself is always [ʔʊjˈʁʊrtʃɛ] ("Uyghur" + adjectival suffix) or [ʔʊjˈʁʊr tili] ("Uyghur language"). So I think either of those is correct here (based on my understanding of, where it says that the pronunciation paramater should be the pronunciation of the native name).

That being said: Hahaha, a content disagreement is not vandalism, so don't accuse others of vandalism because of a legitimate disagreement. <b class="IPA">r ʨ anaɢ</b> (talk) 14:17, 25 April 2013 (UTC)
 * Thanks for your kind word. I just thought [ʔʊjˈʁʊr(tʃɛ)] is not accurate. Or did I misunderstand it? Dose [ʔʊjˈʁʊr(tʃɛ)] means " [ʔʊjˈʁʊr] or [ʔʊjˈʁʊrtʃɛ] " ? As you said, [ʔʊjˈʁʊr] is not the language name. So I wrote [ʔʊjˈʁʊrtʃɛ]/[ʔʊjˈʁʊr tili], that is the IPA of the Uyghur text above. BTW: Uyghurche should be [ʔʊjʁʊrˈtʃɛ]. --Hahahaha哈 (talk) 15:23, 25 April 2013 (UTC)
 * I assume the parentheses in that version were meant to indicate that the "-che" is optional. <b class="IPA">r ʨ anaɢ</b> (talk) 13:36, 26 April 2013 (UTC)


 * My objection was not to how the Uighur was handled, but to how the English was handled. — kwami (talk) 01:39, 26 April 2013 (UTC)
 * We know what you do. It is happy to see that finally you understand what we do. Cheers.--Hahahaha哈 (talk) 03:28, 26 April 2013 (UTC)

The Holy Bible in Eastern (Kasiigar) Turki (1950)
https://archive.org/details/rosettaproject_uig_gen-1

06:48, 8 February 2014 (UTC)

Chinese wikipedia's policy on Historical Uyghur related articles
People who were called Uyghur "Huihu" 回鶻 at the time of the Uyghur Khaganate, get linked to Uyghur Khaganate. So Bayanchur Khan will get linked to Uyghur Khaganate for his ethnicity.

https://zh.wikipedia.org/wiki/回鹘

People who were called wèiwùér 畏兀儿 at the time of the Kingdom of Qocho, get linked to Kingdom of Qocho. So Wang Baobao's mother's ethnicity is linked to Kingdom of Qocho in Chinese wikipedia.

https://zh.wikipedia.org/wiki/畏兀儿

Note that Old Uyghur alphabet is called Huihu script and not Old Wéiwú'ěr script in Chinese wikipedia.

https://zh.wikipedia.org/wiki/回鹘文字母

Modern Uyghurs get linked to the modern Uyghur people article

https://zh.wikipedia.org/wiki/维吾尔族

The Old Uyghur language has its own separate article from modern Uyghur, on the Chinese, German, Russian and Uzbek wikipedias but nobody created an article about it on English wikipedia yet.

Chinese wikipedia calls it "Huihu" language 回鹘语, and indicates that this was spoken in the Kingdom of Qocho (Gaochang)

https://zh.wikipedia.org/wiki/回鹘语

There are also a Chinese, Russian, and Turkish wikipedia articles on the language of the Turkic Khaganate, Chinese wikipedia calls it "Tujue" language. Note that it is an entirely different article from the Old Turkic language article.

https://zh.wikipedia.org/wiki/突厥语

We need to create a separate "Old Uyghur language" article, and an article on the language of the Turkic Khaganate which would both be separate from Old Turkic language article.

In fact the Chinese article linked to the English Old Turkic language article is a disambiguation page, it lists Tujue language, 回紇 Huihe language (alternate name for Uyghur Khaganate language), and (Gaochang) Huihu language

https://zh.wikipedia.org/wiki/突厥－回鹘语

Rajmaan (talk) 22:11, 5 May 2014 (UTC)

Eastern Turki Materials in European Archives Part I
http://dissertationreviews.org/archives/1884

Rajmaan (talk) 17:57, 27 September 2014 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 1 one external link on Uyghur language. Please take a moment to review my edit. If necessary, add after the link to keep me from modifying it. Alternatively, you can add to keep me off the page altogether. I made the following changes:
 * Attempted to fix sourcing for http://www.uwm.edu/~vaux/uyghur.pdf

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

Cheers. —<sup style="color:green;font-family:Courier">cyberbot II <sub style="margin-left:-14.9ex;color:green;font-family:Comic Sans MS"> Talk to my owner :Online 10:35, 18 October 2015 (UTC)

Modern sources


Ibrahim Muti'i

Ibrahim's story

http://faculty.washington.edu/stevehar/Ibrahim.pdf

The Ethnogenesis of the Uighur

http://faculty.washington.edu/stevehar/ANTH470Gladney2.pdf

07:27, 19 October 2015 (UTC)

Uyghur transliteration
Could someone well versed in Chinese-to-Uyghur-to-Latin script transliteration please contact me? I find myself editing Xinjiang-related articles once in a while, but don't know how to transliterate their names (I don't want to do so in Pinyin). For example "阿力木江•买买提". Colipon+ (Talk) 19:14, 22 January 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 3 one external links on Uyghur language. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Corrected formatting/usage for http://www.uyghurdictionary.org/excerpts/An%20Introduction%20to%20LSU.pdf
 * Corrected formatting/usage for http://kuscholarworks.ku.edu/dspace/bitstream/1808/5624/3/EngYakDwy2009_Uyg1full_10.pdf
 * Replaced archive link http://www.webcitation.org/6McjVeWyi?url=http%3A%2F%2Fweb.archive.org%2Fweb%2F20060208045946%2Fhttp%3A%2F%2Fwww.uwm.edu%2F%7Evaux%2Fuyghur.pdf with https://web.archive.org/web/20060208045946/http://www.uwm.edu/~vaux/uyghur.pdf on http://www.uwm.edu/~vaux/uyghur.pdf

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.— InternetArchiveBot  (Report bug) 11:32, 9 November 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 3 external links on Uyghur language. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://archive.is/20130415173450/http://www.rfa.org/uyghur/xewerler/tepsili_xewer/turki-dillar/diwani-02122010022941.html to http://www.rfa.org/uyghur/xewerler/tepsili_xewer/turki-dillar/diwani-02122010022941.html
 * Corrected formatting/usage for http://www.uyghurdictionary.org/excerpts/An%20Introduction%20to%20LSU.pdf
 * Corrected formatting/usage for http://kuscholarworks.ku.edu/dspace/bitstream/1808/5624/3/EngYakDwy2009_Uyg1full_10.pdf

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 23:59, 3 September 2017 (UTC)

Completely ridiculous edit war over phrasing
, I really don't get why we are fighting over the minor edit of Special:Diff/1020526335. The paragraph has already started with stating the current Uyghur language is a descendent of one of the Karluk languages, a classification that includes all Middle Turkic langauges such as Khaqani. The content should not be repeated simply for the fact that it comes from some Gerard Clauson: the "according to" repeat is only necessary when WP:BIASED, while we are looking at two widely-accepted facts about language classification. --Artoria2e5 <small style="font-weight:lighter">🌉 01:00, 30 April 2021 (UTC)

Bad Font
The fonts being used for the Uyghur are wrong. They are being rendered in either Diwani or Ryqa. Uyghur (and central Asian languages in general) tended to write in either Tuz or Nastaliq. On the Ottoman Alphabet page we can perhaps use either Ryqa or Diwani, however for Uyghur it is probably best we use Tuz as it is the most unique and contrastive style of Uyghur that isn't shared with other languages. — Preceding unsigned comment added by Dimashlar (talk • contribs) 23:36, 24 October 2021 (UTC)

Wiki Education Foundation-supported course assignment
This article is or was the subject of a Wiki Education Foundation-supported course assignment. Further details are available on the course page. Peer reviewers: LBrieuc.

Above undated message substituted from Template:Dashboard.wikiedu.org assignment by PrimeBOT (talk) 12:12, 17 January 2022 (UTC)

Wiki Education Foundation-supported course assignment
This article was the subject of a Wiki Education Foundation-supported course assignment, between 6 July 2020 and 16 August 2020. Further details are available on the course page. Student editor(s): Michiguoo.

Above undated message substituted from Template:Dashboard.wikiedu.org assignment by PrimeBOT (talk) 12:12, 17 January 2022 (UTC)