Talk:Middle Chinese

Unihan Tang Dynasty pronunciations
The Unihan Database gives both "ɑ" and "a" in romanizations of Tang Dynasty pronunciations. Is there a difference between these two vowels or do they both mean a regular "a" phoneme, and just an inconsistency in their typing?

An example of their use of "ɑ" can be found here and "a" is found here. Badagnani (talk) 02:14, 10 July 2008 (UTC)


 * The Unicode Database says "derived from various sources" in regard to the Tang pronunciations. In the unihan.txt file, it tells you that other than the kIRG- prefixed data fields, all other data is provisional, and cannot be taken as accurate. The answer to your question therefore lies in where the data comes from. Unicode's page,  is not forthcoming, nor is there any info about it on the unihan.txt file. Dylanwhs (talk) 06:52, 10 July 2008 (UTC)

OK, I just wrote to the Unihan people and received a response. According to this webpage (which seems not to be linked on the main page you presented above), Their Tang readings come from T'ang Poetic Vocabulary by Hugh M. Stimson (Far Eastern Publications, Yale University, 1976). So, how did Stimson derive his 4,000 or so reconstructed pronunciations? By comparing rhymes, or by comparing with Cantonese, Vietnamese, and Korean readings? Or a mixture thereof? Badagnani (talk) 21:55, 15 July 2008 (UTC)


 * I don't have Stimson's book, so I can't help verify any reading provided in Unihan.txt. I guess the methodology he must have used to get to his reconstructed readings is found there. gives a listing of those reading in Unihan.txt. Dylanwhs (talk) 22:55, 15 July 2008 (UTC)

They also told me that the ɑ and other symbols that appear in many Tang readings in the Unihan database are actually IPA symbols. Thus, ɑ is different from a, ɛ is different from e, etc. Badagnani (talk) 21:55, 15 July 2008 (UTC)


 * Each letter of the IPA gives some phonologically different property, whether they be consonants, vowels or diacritic marks. Whether or not Stimson's use of IPA is correct or not one had to careful with free unchecked data. Dylanwhs (talk) 22:55, 15 July 2008 (UTC)


 * I managed to obtain a copy of Stimson's 55 Tang Poems, and the short introduction tells you mostly about Stimson's reconstruction. However, the system in 55TP is a simplification of the Tang Poetic Vocabulary readings, such that certain rimes are grouped together. For example in 55TP the -om/-op rime are a conflation of both 覃 (-om) and 談 (-ɑm) rimes in TVP etc. This is shown by the kTang data which derives from TVP not 55TP. Dylanwhs (talk) 13:36, 4 July 2016 (UTC)


 * Yes, the introduction to Tang Poetic Vocabulary gives a system that distinguishes all the traditional finals, and gives groups of finals that may be unified, though he says he's not unifying them in that book. Kanguole 21:13, 4 July 2016 (UTC)

Phonology table
(UPDATE) I think the table should be consistent with the one in the Chinese article about the "36 initals" and the English rime table article to eliminate confusion between them. So it should look a lot like this - 60.53.47.157 (talk) 06:13, 13 June 2009 (UTC)


 * Note: I do not speak any Sinitic languages, nor do I have any scholarly experience with them.
 * Why is the phonology table written unlike many other tables on the website? Why are the places of articulation the y-axis? Wouldn't it be more consistent for the chart to resemble the official IPA chart? Also, x and ɣ are velar fricatives, not laryngeal fricatives.
 * I don't know how to edit Wikipedia articles or how to use HTML, but I really think this table should be reworked. AnnaColon3 (talk) 03:52, 16 June 2024 (UTC)

William Baxter
Why among all of the scholars whose systems appear here is William Baxter not one?Tibetologist (talk) 08:30, 6 January 2010 (UTC)


 * It seems to have been remedied now. 08:34, 27 March 2011 (UTC)

Category:Extinct languages vs. Category:Extinct languages of Asia
Category:Extinct languages of Asia is itself a category within Category:Extinct languages. — Robert Greer (talk) 09:06, 14 August 2009 (UTC)

Video for discussion

 * 李白 月下獨酌 中古漢語朗讀
 * 李白 將進酒 中古漢語朗讀
 * 李白 靜夜思 中古漢語朗讀
 * 杜甫 佳人 中古漢語朗讀
 * 崔顥 黃鶴樓 中古漢語朗讀
 * 李白 清平調 lix brak, chieng bieng deuh Arilang   talk  12:40, 14 March 2010 (UTC)

Exemplars?
What would really help here, more than the rhyme-tables, is a transcription of a text (preferably a poem, for obvious reasons) into the various reconstructions used, much like the Example passages at Latin Spelling and Pronunciation. I would recommend a text either by Du Fu or Li Bai from the 唐詩三百首, since these are both (a) very well known to native Chinese and (b) often used in texts for students studying classical Chinese as a second language. If there are no objections, I'll go ahead and do this. Or someone else can if they feel inclined.

Also, it would be nice to have corresponding audio recordings which attempt to put the various reconstructions into practice. I can do this, too, if there are no objections. — Preceding unsigned comment added by Szfski (talk • contribs) 02:20, 15 September 2010 (UTC)

Tables
The table of initials attempts to present both phonetic values and transcriptions. However it shows that there is almost complete agreement between recent accounts of the phonetic values (Li, Zhou, Pulleyblank, Pan and Baxter). So I'd suggest splitting this into two tables, one giving these consensus values in a grid like at Old Chinese (with notes to cover any variations), and another to compare transcriptions.

With the finals there is no such consensus on phonetic values; indeed it's widely held that no single variety contained all the distinctions, so the comparison of transcriptions is all we have, apart from some broad statements about codas and numbers of vowels. I'd further suggest hiving off these comparisons of transcriptions to a separate article, to make this one more readable. Kanguole 14:23, 24 March 2011 (UTC)


 * I agree for the most part. For the initial tables, I don't think we really need the comparison of transcriptions at all.  Just include a table of consensus values, arranged like normal phonological grids (essentially like the one at Old Chinese), where each entry gives both Baxter's "ASCII-convenient" notation (since it appears frequently) and the normal IPA equivalent.  Describe variations using footnotes and/or a set of notes after the whole table.  The only actual table of "comparison of transcriptions" we need is a small one that compares Karlgren's values to consensus agreement and lists only the entries where they differ.


 * For the finals tables, I agree that we need a table of reconstructions. I actually think we need two tables: One covering the finals purely from a MC perspective, and one covering the diachronic perspective.  As it is, I've filled in some diachronic info (Pulleyblank's LMC reconstructions, Modern Mandarin outcomes) into this table, but I realize the table will get too big to include everything I want and needs to be split out.  For the synchronic table in this article, I'd take out Pulleyblank's LMC outcomes but would consider leaving the Mandarin outcomes in, compressed or perhaps just in the form of one or more examples to help relate the otherwise very dusty table to reality.  I'd also take out Wang Li's column; it appears to be hardly different from Karlgren's, and the few changes he made that aren't just notational variants are incorporated into Li Rong's reconstruction (still very Karlgrenian, but with a number of significant changes).


 * As for the diachronic table: First, it should probably go into Chinese historical phonology, which I've been expanding by adding material that it needs to cover and moving the diachronic info from the Old Chinese and Middle Chinese articles into it. My ideal diachronic table would contain one row per EMC final, probably in the same order as currently, giving the following columns, in order:
 * OC final(s) that lead to this EMC final, given either according to Baxter's OC reconstruction or according to Schluessler's Old Chinese Minimal (OCM), or perhaps both when they differ
 * Late Han (LH) final(s) that lead to this EMC final, according to Schluessler
 * EMC final according to Karlgren
 * EMC final according to Baxter
 * EMC final according to Pulleyblank
 * LMC final according to Pulleyblank
 * Modern outcomes in a representative Min Nan dialect, presumably either Amoy or Taiwanese
 * Modern outcomes in Cantonese
 * Possibly, modern outcomes in Shanghainese
 * Modern outcomes in Standard Mandarin


 * Probably, split final rows so that -p/-t/-k are separate from nasals, since they turn out :very differently in various modern forms. Possibly, also split the rows according to places where an EMC retroflex sibilant or dental sibilant bends a following high-front vocalic.  Possibly move the Min column between the Late Han and first EMC column since Min split off before EMC.


 * It's still a mess in a table like this to have even three different EMC reconstructions, but they're all very different one from another and they're all relevant. Baxter's is probably closest to modern consensus (to the extent it exists), but Karlgren is still relevant -- many authors (e.g. Schluessler in his OC dictionary) still uses Karlgren's reconstructions in his OC dictionary -- and since the LMC reconstructions are from Pulleyblank, including his EMC reconstructions is important in working out how much EMC->LMC change is due to genuine differences and how much just to notational or conceptual disagreements.

Benwing (talk) 10:00, 4 April 2011 (UTC)


 * A couple more comments:
 * The diachronic table above should also (ideally) have modern outcomes in Sino-Japanese Go-On, Sino-Japanese Kan-On, Sino-Korean and Sino-Vietnamese. The Japanese info is available for the finals is available (to some extent) online from Baxter's old OC etymological dictionary, at .  You can also get Sino-* (I think) using the Unihan tables (although they don't clearly distinguish Go-On vs. Kan-On), and with a little program, you could take the info from Baxter's new tables with Sagart, rearrange them into normal text database form, add the relevant info from the Unihan tables, bring them up into a spreadsheet or whatever, sort on any column you want, and use that to figure out the relevant reflection of an initial or final into any language you want, provided the Unihan database has the reading for that language.
 * There should also be a diachronic table for initials, giving the same columns. Some info for this is found on the Middle Chinese page, some on the Sino-Japanese vocabulary page, some in Baxter-Sagart's recent info.

Benwing (talk) 11:01, 4 April 2011 (UTC)

English translations of Chinese terminology
I think adding English translations of Chinese terminology detracts from the quality of these articles, for instance 重紐 chóngniǔ as 'repeated button' in this one, and 韻目 yùnmù was translated as 'rime eye' in the Qieyun article. This is plainly inaccurate and silly. Anyone can babelfish the meaning of the individual common modern meaning. If you're unsure or don't know the reasons why the character was chosen, in these cases 紐 and 目, then don't translate. For your information, 紐 is related bonds as in realtionships rather than buttons. 目 on the other hand derives from lists (biaomu) rather than eye. Dylanwhs (talk) 23:44, 29 March 2011 (UTC)


 * I didn't do those translations myself. rime eye wasn't added by me at all (AFAI remember), while "repeated button" comes from another paper on Old and Middle Chinese.  Rather than not translate them, I think it's better to just translate them right.  For example, in place of "repeated button" maybe you should say "repeated connection" or "paired connection" or "bound pair" or something, whatever seems to you to most accurately convey the meaning.  I do think it's useful to give the English meaning of Chinese terms, and for many people it makes the material much more approachable. Benwing (talk) 10:05, 4 April 2011 (UTC)


 * We need to be careful to use translations given by our sources, to avoid the risk of interpretation. I think these ones were, though they are very literal.  And even literal translations can be helpful, provided they're marked as such.  Kanguole 12:31, 4 April 2011 (UTC)


 * The gloss "button" is given by Norman (1988) and Baxter (1992), who say that 紐 niǔ refers to the small circle used to delimit homophone groups in the Guangyun. Kanguole 15:07, 12 April 2011 (UTC)


 * Right, I agree that if a source gives a translation, use that one. Otherwise probably a literal translation is best -- it minimizes interpretation, since the interpretation is only in choosing among possible definitions if there is more than one. Benwing (talk) 15:38, 15 April 2011 (UTC)

/j/ is called 以 or 喻?
Wondering why 喻 is used for /j/. 174.27.253.240 (talk) 01:17, 22 July 2011 (UTC)

Problem with transcription in Table
Karlgren definitely never used IPA symbols. Consequently the table with these headings is very misleading. Also, by giving a column just called 'initial' it suggets that this column is the truth, which Baxter and Karlgren interpret somehow. Such a perspective is absurd.. If anything the first column should have Chinese characters in it and the Karlgren column should have the symbols which Karlgren actually used. Tibetologist (talk) 09:15, 13 March 2012 (UTC)

Tang reconstructions?
What are these, who does them, and how are they made? They look like another presentation of the Qieyun data, in a system showing pinyin influence. Kanguole 17:36, 1 June 2012 (UTC) This is also being discussed at Template talk:Infobox Chinese. Kanguole 10:23, 2 June 2012 (UTC)

When ?
It is amazing that in all of the long turgid introduction to this, there is no information about the simple and obvious question,  when was this language in use ? 100 years ago ? 1000 years ago ? 5000 years ago ?Eregli bob (talk) 11:58, 23 July 2012 (UTC)


 * The article lead does contain some clues as to dates. However, the date range over which Middle Chinese existed is problematic. Middle Chinese is especially associated with the Qieyun (dated 601 AD, although even that is somewhat problematic, since it is mostly known through later versions). How long before and after this date Middle Chinese was spoken is unknown, even in terms of centuries. Using a figure of 1400 years ago for Middle Chinese would be misleading and inaccurate (although true enough for the Qieyun, with which MC is associated). Dcattell (talk) 17:29, 25 July 2012 (UTC)


 * Well, no more "misleading and inaccurate" than no information at all.  One would think that the evolution of a language is a continuous process, so although you would not expect a definite date when a particular phase of the language "starts" and "ends" (unlike, for example, a dynasty, which does have a start and end date),  you need some idea of when it was in use.Eregli bob (talk) 18:27, 26 July 2012 (UTC)
 * The difficulty is that Middle Chinese does not refer to a language, but to the phonological system of the Qieyun and its interpretation in the rime tables. And that phonological system was itself a composite of several reading traditions rather than a description of a particular form of speech.  Norman and Coblin (1995) put it rather bluntly: "The conclusion must be that Ancient Chinese (or Early Middle Chinese, which is only another name for the same thing) has no proper phonology of its own, no lexicon and no grammar. It is not a language."  Kanguole 18:20, 25 July 2012 (UTC)

An addition to general vagueness, the lead contains several inaccuracies: It also does not summarize the article, which is solely concerned with phonology, as are just about all the available sources. Here's my suggested version:
 * MC is not a constructed language, or even a linguistic reconstruction as commonly understood.
 * There is no "extensive surviving literary record"; MC cannot be identified with Classical Chinese (Warring States and Han) or Literary Chinese (a style modelled on the former).
 * Traditional vs. simplified characters is irrelevant.


 * Middle Chinese (formerly known as Ancient Chinese) is the system of Chinese pronunciation recorded in the Qieyun, a rime dictionary first published in 601 AD and followed by several revised and expanded editions. As the original form was difficult to use, this system was further analysed in a series of rime tables dating from the 12th century, but probably representing an earlier tradition.  The rime tables attest a number of sound changes that had occurred over the intervening centuries.  Many linguists refer to the system of the Qieyun as Early Middle Chinese and the variant revealed by the rime tables as Late Middle Chinese.


 * The Swedish linguist Bernard Karlgren believed that the dictionaries recorded a speech standard of the capital Chang'an of the Sui and Tang dynasties, and produced a reconstruction of its sounds. However, based on the more recently recovered preface of the Qieyun, most scholars now believe that it records a compromise between northern and southern reading and poetic traditions from the late Southern and Northern Dynasties period.  As such, the system contains important information for the reconstruction of Old Chinese phonology (1st millennium BC).  The system is also used in the study of Tang poetry, and as a framework for describing modern varieties of Chinese, most of whose distinctions can be traced to it.

Kanguole 11:49, 30 July 2012 (UTC)


 * Very good. Still, I wonder: Do Sinologists and other scholars (for example, linguists studying the loanword layers of Bai, Hmong–Mien, Tai–Kadai, Vietnamese, Korean and Japanese) never feel any need to refer to a stage (any stage) earlier than Qieyun's Early Middle Chinese but later than Old Chinese? Like, a stage in the first half of the 1st millennium AD. After all, this period covers the Han and the Six Dynasties (the Three Kingdoms, Jin and Southern and Northern Dynasties) – not exactly a dark or illiterate period. That's clearly not Old Chinese anymore according to our definitions (which are conflicting: Old Chinese is sometimes said to end already before the end of the 1st millennium BC, or extend somewhat beyond the start of the Christian era). "(Late) Proto-Chinese", i. e., the last common ancestor of all Chinese dialects including Min (and Ba-Shu, presumably) seems to also fall into that period, and should be reconstructible at least in principle, to some considerable extent. Would the (standard/court?) language spoken and written under (and by) Emperor Zhang (ruled 75–88) be described rather as a late form of Old Chinese, an early form of Middle Chinese or a transitional form? Can the Midnight Songs be said to be already in a form of Middle Chinese? (Not altogether irrelevant, since not only pronunciation changes but also grammar and lexicon, and pronunciation is important in poetry for rhymes and other wordplay, after all.) What would we reconstruct their original pronunciation to be like? Zu Chongzhi and Emperor Wu of Liang – did they already speak Middle Chinese in any meaningful sense, despite living earlier than the Qieyun? Do the go-on readings already essentially reflect a type/variant/dialect of Middle Chinese? --Florian Blaschke (talk) 00:30, 21 November 2013 (UTC)


 * I think it's dominated by the available documentary evidence. There are two points for which we have phonetic information about almost the entire lexicon: the Qieyun (Middle Chinese) and the phonetic compounds together with the Shijing rhymes (Old Chinese).  (That also means that these are somewhat formal constructs, with an unclear connection to actual speech.)  There's more sporadic information from other periods, e.g. extensive contemporary discussion of pronunciation from the Eastern Han, and rhyming practice in various periods.  Perhaps due to a combination of the availability of documentary sources and the complexity of the history, there has been little use of the comparative method, except as a supplement to the documentary sources.  (cf Norman & Coblin (1995) cited in the article)
 * Linguists do try to date the various changes. There's plenty of literature, but speech was diverging from writing in that period.  Also the Eastern Han sources show considerable regional variation, and it's assumed that goes back at least to the Warring States period.  Go-on is usually compared with the southern component of the Qieyun system.  Kanguole 18:15, 21 November 2013 (UTC)

please include Middle Chinese reconstructions in main body of Sino-Xenic/modern table of numbers
It is difficult to make sense of the sound changes involved in this table if the reconstructed forms aren't given, or are stashed inconveniently in a footnote. I notice that an IP user put some forms in, but user Kanguole inexplicably moved them to a footnote. I'd strongly recommend putting them back, along with Li's alternate reconstructions (if, as the text says, Li's and Baxter's reconstructions are the most commonly used). That way, readers can compare directly.

BTW it is absolutely standard linguistic practice to include reconstructions in such tables. Putting them in a footnote seems very strange to me and I suspect reviewers of a paper structured this way would complain and ask to have them in the main body. Benwing (talk) 05:36, 22 February 2013 (UTC)


 * I still think the MC transcriptions are out of place in the Sources section, but I've added them anyway. Note that Li's and Baxter's notations do not claim to be reconstructions, but transcriptions of the Qieyun categories, which are no longer considered to represent a single variety.


 * I've also replaced the Hokkien examples with a Wu dialect. Min is usually considered to have branched before MC, while Wu dialects are important for their retention of voicing (though there's only one example of that here).  Kanguole 10:14, 23 February 2013 (UTC)

Reversion Revision 731429721. Character, pinyin format

 * Many articles already use the patterns seen in {{zh and {{nihongo. In my meticulous edit, I was actually trying to make the article self-consistent, as most of the article puts the Character first. Why should that section be different from the rest? Also, given the context of Middle Chinese, modern Mandarin readings seem a bit distant. Yes, they may serve as "reference" readings, but should probably not be misconstrued as "primary" readings, so why not let the Character stand first? You also reverted yīn (阴/陰), and the "lower", or yáng (阳/陽). The simplified characters are not used elsewhere in this article. Hongthay (talk) 22:09, 25 July 2016 (UTC)
 * I agree on aiming for consistency within the article, but the only place characters were placed before pinyin in the text was in the first paragraph of the Tones article. If we were talking about the characters (e.g. in illustrations of rhyme books and rhyme tables), it makes sense to put them before the pinyin.  But in the text we're talking about words, not of MC, but the terminology of Chinese phonology.  For readers of English (the target audience of this wiki), 聲母 does not speak, but shēngmǔ does, so placing the pinyin form first makes the text flow more easily.
 * I also agree about consistency on the simplified/traditional issue, though I'm not sure which way to jump. The reason for giving the characters for these terms is so that readers who can handle them can relate them to Chinese texts on phonology.  But some of those texts are written in simplified characters.  Kanguole 00:29, 26 July 2016 (UTC)
 * Thanks for replying. Parts of Rime table were copied and pasted here, introducing a sudden high density of IL's followed by details in parentheses. WP:CHINESECHARACTERS. I wonder if there is a better way to achieve flow since there's a main article with the details already. Thanks. Hongthay (talk) 04:57, 26 July 2016 (UTC)
 * I think that's right – we could drop a lot of the traditional Chinese terminology from this article, since that detail is provided by the rime dictionary and rime table sub-articles, and that would make it easier to read. A possible exception is shè, which is sometimes used in English-language works to avoid confusion with the Qieyun rhyme groups.  Kanguole 23:42, 26 July 2016 (UTC)
 * Thanks for your efforts, looking a lot better now! Hongthay (talk) 15:12, 27 July 2016 (UTC)

External links modified (January 2018)
Hello fellow Wikipedians,

I have just modified one external link on Middle Chinese. 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://web.archive.org/web/20100403045014/http://www.eastling.org/tdfweb/midage.aspx to http://www.eastling.org/tdfweb/midage.aspx

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) 14:34, 29 January 2018 (UTC)

Wiki Education assignment: CHIN 330 Introduction to Chinese Linguistics
— Assignment last updated by PewPew924 (talk) 03:09, 28 September 2022 (UTC)