Wiktionary talk:About Chinese/Wu

I like this document, it's very interesting and educational and will help understand the transliteration and IPA you use. However, for me personally it's not very useful at the moment in terms of me contributing in Wu, as I don't know what the pronunciation for each character is and what tone is used on the 1st syllable to determine the numbers (1 to 5). Is there a database somewhere on this? How do I know how to transliterate any given Hanzi and it's tone? I'll watch your Wu edits, though. Meanwhile, I'm able to contribute a bit in Cantonese and Min Nan a bit. Cantonese is not that alien to me, I had a bit of exposure. Min Nan is new for me but it's well documented.

Sorry to ask you again, I haven't saved the link to Shanghainese pronunciation online. What is it again? I might at least get exposed to the sounds of Wu Chinese. --Anatoli ^{(обсудить}/^вклад) 01:24, 9 May 2014 (UTC)

I partly translated what I wrote in w:zh:上海话 two years ago... I have added some resources for Shanghainese and Wu. Note that Wu Chinese MiniDict may be erroneous sometimes, and the most reliable resource is the comprehensive dictionary. Websites which have Wu recordings include: forvo, shh.dict.cn and Phonemica. Wyang (talk) 02:18, 9 May 2014 (UTC)

Thank you. I might the resources for my own benefit but I'll refrain from actively editing Wu. Just occasionally maybe, with your guidance. --Anatoli ^{(обсудить}/^вклад) 02:24, 9 May 2014 (UTC)

I particularly like the third link there. It is very interesting. :) Wyang (talk) 02:30, 9 May 2014 (UTC)

@Wyang, Atitarev, Justinrleung, Mteechan: Are we aware of https://www.wugniu.com/? —Suzukaze-c◇◇ 07:11, 16 May 2018 (UTC)

@Suzukaze-c: Yes. — justin(r)leung _{{ (t...) | c=› }} 07:13, 16 May 2018 (UTC)

Wu transliteration for 北京

@Wyang I used "4poq jin" to transliterate 北京 according to wu-chinese site ("4 poh" + "cin") and the sound changes you described. Is this correct? I tried to do 人類 but it gives variant readings for "人" - "1 nyin" and "1 ze". --

Yes, well done! Shanghainese has many more variantly pronounced characters than Mandarin (called the "Literary and colloquial readings of Chinese characters"), an example is 人. 'zen' is used in high-level literary words, such as '人類', '人民', and 'nyin' is used in colloquial words, such as '上海人'. 人類 would be '3zen le'. Wyang (talk) 03:34, 10 May 2014 (UTC)

Simplified, toneless transliteration for Wu Chinese outside pronunciation sections

Can we use simplified, toneless transliteration outside pronunciation sections - usexes, translations, see also, etc.? E.g. 中国 (tson koq), 刑法 (hhin faq). --Anatoli ^{(обсудить}/^вклад) 00:03, 12 May 2014 (UTC)

I'm not sure. The purpose of using romanisations is to help readers pronounce words in foreign scripts. Toneless transliteration fails the criterion of transliterative faithfulness in that it doesn't carry all the information that is necessary to reproduce the actual pronunciation. In this way it is a bit like toneless Pinyin in usexes and translations. But the case is also slightly different as there is no faithful alternative to it. Wyang (talk) 00:09, 12 May 2014 (UTC)

Yes, that's my reason too - no faithful alternative (what about diacritics, are they possible with Wu?). Well, If Wu entries are regularly created (or rather Wu pronunciation is provided on Chinese entries), then a link, such as 中国 (tson koq) would guide users to the proper pronunciation section with IPA. I'd compare it with romaji without the pitch accent. It seems Wu tones are complicated, not as strong, depend on the position in a sentence (I might be wrong in any of these claims). Besides, Wikipedia doesn't usually provide tones for Wu either. Using this transliteration would also promote it and make users more comfortable with it. As an alternative for short words, a Wu translation for China, might be 中国 (tson koq) (T1), 中国 (tson koq (T¹)) or 中国 ((T¹) tson koq). For usexes, it would be messy though.--Anatoli ^{(обсудить}/^вклад) 00:21, 12 May 2014 (UTC)

Another resource

@Wyang I've just bought a book - "Follow me to say Shanghai dialect" (根我说上海话) →ISBN. It's not a dictionary but it seems like a good introduction for beginners. Every sentence has Mandarin and English translations, transliteration is provided throughout and it has audio. It uses different numbers for the five tones:

• 52 阴平
• 34 阴去
• 23 阳平
• 5 阴入
• 12 阳入

However, in the transliteration, they also use "1", e.g.:

侬早，我看到侬真高兴。 [Shanghainese, trad. and simp.]

Good morning, I am very glad to see you

The transliteration is "nong23 zao3, ngu23 koe3 dao5 nong1 zen52 gao5 xin1" Which tone is number 1? Is it neutral? --Anatoli ^{(обсудить}/^вклад) 05:43, 12 May 2014 (UTC)

There is not much difference between their notation and the one here. The tonal values for the five tones (when in isolation) are essentially the same as the values here (Table 1, Column "Tone values"). The values in the example sentence are IPA values, after tone sandhis when the syllables are not in isolation. It can be analysed into words as:

Here	侬早 (phrase TS, 33-34)	我 (1 syl word, T3, 23)	看到侬 (3 syl word, T2, 33 55 21)	真 (1 syl word, T1, 53)	高兴 (2 syl word, T1, 55 21)
There	侬早 (phrase TS, 23-3)	我 (1 syl word, T3, 23)	看到侬 (3 syl word, T2, 3 5 1)	真 (1 syl word, T1, 52)	高兴 (2 syl word, T1, 5 1)

As seen in the comparison, there is basically no difference, apart from preferences for slightly different tone values. There are only minor differences in the romanisation:

	侬早	我	看到侬	真	高兴
Here	non33-tsau34	ngu23	khoe33-tau55-non21	tsen53	kau55-xin21
There	nong23-zao3	ngu23	koe3-dao5-nong1	zen52	gao5-xin1

Wyang (talk) 05:59, 12 May 2014 (UTC)

Thanks for the detailed reply! I'll try to cross-reference the book to WT methods. --Anatoli ^{(обсудить}/^вклад) 06:13, 12 May 2014 (UTC)

@Wyang could you explain the abbreviations, please? How do I know that "3" stands for "33" (not "23"), etc. Is there a reference for numeric values and tones, IPA somewhere? --Anatoli ^{(обсудить}/^вклад) 02:08, 13 May 2014 (UTC)

Which abbreviations? '3' stands for the mid-level tone, ˧. It is essentially the same as '33', ˧˧. Wyang (talk) 02:10, 13 May 2014 (UTC)

Sorry for my ignorance. I'm just not very used to numeric notations. So, 3=33, 5=55, 1=21. What about 52=53, 23=33? Are the identical or just almost the same? --Anatoli ^{(обсудить}/^вклад) 02:17, 13 May 2014 (UTC)

3=33, 5=55, 1=11 essentially. Some people use the former for shorter (checked-tone) syllables and the latter for longer syllables (i.e. /pak3/, /pa33/), some people do not make that distinction. 52~53, 11~21, 23~33 are almost the same - the difference is within the range of normal populations, another native speaker or even the same speaker may reproduce '33' as '23'. Wyang (talk) 02:21, 13 May 2014 (UTC)

Thanks, it's easy for me to use my book now. I might try adding translit to some short phrases, since I can convert the numbers and perhaps make a conversion table. They use "sh" for /z/, "fh" for /v/, xh for /ɕ/. Are you familiar with such notations, are they common? --Anatoli ^{(обсудить}/^вклад) 02:31, 13 May 2014 (UTC)

Is it Qian Nairong (錢乃榮)'s work? Wyang (talk) 02:54, 13 May 2014 (UTC)

Yes, Qian Nairong, Wang Xiaoming, [1]. You can see a sample page on that link. I got it for $11 in Melbourne (I get 10% discount in a Chinese bookshop). The audio is a bit too fast for beginners, phonetics are only explained in Chinese, English translations are not always right and only a few things are explained from a Mandarin speaker's point of view but it's not a bad book at all, plenty of dialogues. --Anatoli ^{(обсудить}/^вклад) 03:07, 13 May 2014 (UTC)

Seems to be well worth the money. Qian is the most (and only) prolific author of Shanghainese learning material. His romanisation is a little strange though, he uses 'h' to consistently mark voicing, so that the 'b-p-ph' contrast becomes 'bh-b-p', and 'v' becomes 'fh'. 'sh' is then unintuitively used for 'z'. Wyang (talk) 03:16, 13 May 2014 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ Oh, this is what "h" is for. "jh" is for /dʑ/ and "xh" for /ʑ/ (I made a mistake above, I thought it was kind of weird because, I'm not familiar with Wu phonology). --Anatoli ^{(обсудить}/^вклад) 03:30, 13 May 2014 (UTC)

I expanded the table a bit, using what I could make of the sample page you linked to. Wyang (talk) 03:54, 13 May 2014 (UTC)

Whath is the translit for 雨? Qian uses "yhu". --Anatoli ^{(обсудить}/^вклад) 09:30, 13 May 2014 (UTC)

3hhy. Wyang (talk) 10:09, 13 May 2014 (UTC)

Neutral tone

@Wyang 𠲎 is marked as 0 (neutral tone) in wu-chinese.com/minidict/. "0" didn't work. Please check the transliteration. --Anatoli ^{(обсудить}/^вклад) 15:16, 12 May 2014 (UTC)

Thanks, added. Please check the entry. Wyang (talk) 23:42, 12 May 2014 (UTC)

Great, thank you! --Anatoli ^{(обсудить}/^вклад) 23:48, 12 May 2014 (UTC)

Tone conversion wu-minidict to Wiktionary

@Wyang

Now that major work on the merger is complete (I can't do much apart from manually converting/creating entries), I could spend some time understanding the logic of using Wu/Shanghainese tones, using 吴音小字典·吴语小词典 and this page. Are you able to write out the rules of tone conversion, if there are any rules? Does minidict correctly display the isolated tones? Unfortunately, I didn't understand your explanation at User_talk:Wyang/Archive1#Other_Topolects_for_.E5.8A.A0.E6.B2.B9. --Anatoli ^{(обсудить}/^вклад) 03:27, 26 May 2014 (UTC)

Minidict is mostly correct. Rules of tone conversion are:

Tone (MiniDict)	Voicing of initial	Tone category
平	voiced	3
	voiceless	1
上	voiced	3
	voiceless	2
去	voiced	3
	voiceless	2
入	voiced	5
	voiceless	4

Wyang (talk) 23:42, 26 May 2014 (UTC)

Thank you. What does the final "uae" become as in 台湾 (1the uae)? --Anatoli ^{(обсудить}/^вклад) 00:24, 27 May 2014 (UTC)

'uae' is 'ue'. Wyang (talk) 10:35, 27 May 2014 (UTC)

吳語拉丁字 / Un̂ȳ-Latìnzw

Dukàhóŭ!

WT romanisation seems to be very complicated, and at many levels. Why not using Un̂ȳ-Latìnzw? it would be much more simple. I am also making a proposal for standardised Wu with a dictionary using exclusively Un̂ȳ-Latìnzw romanisation system. Have a look, and you'll see that it is not a bad idea!

Lingstudisto (talk) 22:01, 16 July 2015 (UTC)

It shouldn't be too hard to get accustomed to this system, considering many Chinese editors on Wiktionary seem to have learnt it well. Wyang (talk) 05:36, 19 July 2015 (UTC)

Excuse me, but it seem to be a bad mix between Fawu and Pinyin... with only bad sides of them I mean. Spaces between each syllable looks also like Vietnamese Quoc ngu... Lingstudisto (talk) 15:14, 21 July 2015 (UTC)

Sorry but if you have nothing but bad things to say, feel free to leave and work on your good project. Wyang (talk) 02:05, 23 July 2015 (UTC)

I just criticise. If you take it for yourself, i'm not responsible. I'm here to help building Wiktionary, so what the problem if i use my own tools? If you look from only one spot, you can't get everything of something. So, why not using both my and your systems together?Lingstudisto (talk) 16:13, 24 July 2015 (UTC)

Doing so, the users will decide which of them they prefer.

• Dukàhóŭ is just plain wrong! It's an abomination to use the vernacular pronunciation "du" here. It must be "da". --2.245.76.18 22:26, 23 December 2016 (UTC)

Pronunciation for phoneme "u"

I believe the more accurate phonetic value for phoneme "u" is [v̩ʷ], instead of [ɯ]. — This unsigned comment was added by Mteechan (talk • contribs).

(@Wyang) —suzukaze (t・c) 07:26, 24 June 2016 (UTC)

Yes thanks! Changed. Wyang (talk) 01:50, 25 June 2016 (UTC)

&, +

@Wyang Could these symbols be documented on this page? —suzukaze (t・c) 07:50, 22 August 2017 (UTC)

+ is already documented in #Right-prominent tone sandhi (phrase sandhi). I've never seen &... Wyang (talk) 08:14, 22 August 2017 (UTC)

& is in 日腳. —suzukaze (t・c) 08:29, 22 August 2017 (UTC)

Ah, I had forgotten about it. I've added the documentation now. Wyang (talk) 09:27, 22 August 2017 (UTC)

@Wyang What are hyphens (硬碰硬 (yìngpèngyìng)) for? —Suzukaze-c◆◆ 00:09, 9 April 2018 (UTC)

@Suzukaze-c It forces a mid-tone on a monosyllable. Added now. Wyang (talk) 09:30, 9 April 2018 (UTC)

影 Initial Voiced or Voiceless?

There is a table of consonants which includes ʔ (glottal stop) as an initial phoneme. In another page, 影 is given as an example character for it.

http://wu-chinese.com/romanization/consonant.html

The mini-dict, begins with vowels with an implicit initial glottal stop. Does that make it voiced or voiceless for the purpose of predicting tone?

72.89.72.93 13:49, 24 September 2017 (UTC)

A glottal stop would have ' (apostrophe). It's voiceless, according to Minidict ("in" and 上), if there is no initial consonant, so the syllable is "2in". --Anatoli T. ^{(обсудить}/^вклад) 14:11, 24 September 2017 (UTC)

Pan-Wu Phonetic System

Is there any reason for or against the establishment of a diaphonemic system for Wu? The Wu Minidict seems to distinguish ae and e for the Shanghai dialect, and that seems to correspond to an aen and e finals in the Shengzhou dialect respectively. But WT Romanization here conflates ae and e.

LCS (talk) 17:52, 29 October 2017 (UTC)

@LCS It's nothing against a Pan-Wu system. Wiktionary records 中派上海话, or middle-style Shanghainese (i.e. not old-style or new-style), which does not distinguish between ae and e, so only e is used to write this vowel. Wyang (talk) 06:47, 30 October 2017 (UTC)

Some groups like Yue have pronunciation information for different dialects such as Guangzhou and Taishan. I have a dictionary from Glossika with pronunciations for both the Shanghai and Wenzhou dialects. Would such a source be accepted if Wenzhou pronunciation information is added? LCS (talk) 23:14, 9 November 2017 (UTC)

@LCS This has been discussed before (Template_talk:zh-pron#Wenzhou_dialect and Wiktionary_talk:About_Chinese#Looking_to_improve_Wenzhounese_coverage). I think most people are supportive of adding additional Wu dialects, such as Wenzhounese. For that to take place, the technical infrastructure that handles the romanisation (and converts it to IPA) needs to be put in place, which will be reliant on a good understanding of the phonology of the Wenzhou dialect. We need to produce a mature Wiktionary:About Chinese/Wenzhounese page detailing how the romanisation to be used will correspond to IPA. There is already a romanisation system proposed on that draft page, which should be further refined. As a whole, I find Glossika's romanisation acceptable, but I'm not really fond of their diacritics system of marking tones in the romanisation. Glossika's material can be used as a source of inspiration but hopefully we can avoid having to mark Wenzhounese tones with diacritics. Wyang (talk) 08:50, 10 November 2017 (UTC)

Wu minidict dictionary is now fixed

The Wu minidict dictionary has been broken for a while. I have emailed them and they have fixed it now. --Anatoli T. ^{(обсудить}/^вклад) 21:37, 29 April 2019 (UTC)

Template

The template converting syllables to IPA is very helpful and I'm used to the weird system now, but I find some aspects counterintuitive. I still have too be very careful with the voiced/voiceless/aspirated pairs, as they are different in the dictionary and in the Wiktionary source code. The current system seems to be a mixture of Qian Nairong's most recent revision and something else. It is close to standard Chinese pinyin, using j, q, x, but ts instead of z. You use double consonants for voicing the sibilants jj, xx, but single consonants for voiced stops (b, d, g). Aspiration is denoted with h (ph, th, kh, tsh), however hh is voiced. Personally, I prefer something as close as possible to pinyin like Qian Nairong's current version, with b, d, g and z, c, s as in pinyin and strict doubling for voiced sounds (bb, dd, gg in addition to current xx, jj). Would it be possible to have it changed? --Explosivo (talk) 15:16, 4 April 2020 (UTC)

@Explosivo: I think it'd be hard to change since it's already implemented on so many entries. This change would also be confusing since b, d, g (voiced) in the current system conflict with b, d, g (voiceless) in the proposed system, as an example. — justin(r)leung _{{ (t...) | c=› }} 15:42, 4 April 2020 (UTC)

It is already quite confusing due to the combination of two different approaches. I think some consistency would be nice, but that's just my observation. I wonder if the changes could be automatically implemented, but I'm not familiar with bots. --Explosivo (talk) 16:43, 4 April 2020 (UTC)

I agree that Qian Nairong's system is very elegant.

I think that a change using bots is feasible, but we would have to be very careful. —Suzukaze-c◇◇ 03:40, 5 April 2020 (UTC)

I guess this is about {{zh-pron}}. I don't know Wu, but I wonder if some editors will fail to notice that the transcription system has changed, and contribute old-style transcriptions that happen to be valid new-style transcriptions with a different meaning, causing the template to show the wrong IPA. To avoid that, perhaps the new-style transcriptions should be marked in some way for a while after the transition. For instance it could have a "new" in the parameter name or value. Then the template could accept the old system and generate the correct IPA, or throw an error telling the editor to use the new system. — Eru·tuon 04:19, 5 April 2020 (UTC)

Should tone 3 be allowed for "mm"?

For 畝／亩 (mǔ), it seems that the vernacular reading should be /m̩²³/ (it's even already in the dialectical data). This would appear to require tone 3, suggesting the pronunciation of "3mm". However when I try this I get: "Lua error in Module:wuu-pron at line 75: Invalid syllable: mm3. Voiceless initials only occur in tones 1, 2 and 4."

But from this page it looks like "3mm" only has a final not an initial, so is this error wrong, or is is "mm" a voiceless final without an initial? Or if the difference between initial and final doesn't really exist in this case, then there even such a thing as voiced and voiceless? If not, I don't see why mm can't potentially take any tone, but I feel that I'm not familiar enough with the linguistics.

For now I saved the page with "2mm" to avoid the error, but this might not be correct. ChromeGames923 (talk) 08:28, 7 September 2021 (UTC)

Alternatively, is there supposed to be an "hh" initial? Like in 五 for example, but then that seems to differ from the supposed pronunciation. ChromeGames923 (talk) 09:03, 7 September 2021 (UTC)

@ChromeGames923: I believe it should be 3hhmm. — justin(r)leung _{{ (t...) | c=› }} 14:26, 11 September 2021 (UTC)

@Justinrleung: I see, how can you tell whether that is the case? It seems reasonable and is also suggested by 吳語學堂, but going off the Wiktionary dialectical data and Wu MiniDict I probably would not think of it. ChromeGames923 (talk) 23:42, 11 September 2021 (UTC)

@ChromeGames923: Tone 3 needs a voiced consonant, so for syllabic mm, there needs to be a voiced hh. Also, it is better to follow 吳語學堂, which is closest to our system. The dialectal pronunciation data in the dropdown table might not be right all the time. — justin(r)leung _{{ (t...) | c=› }} 01:45, 12 September 2021 (UTC)

@Justinrleung: I see, and hh is the only plausible voiced initial. Thanks for the advice; I've generally found 吳語學堂 to be quite good, though sometimes there's readings missing (and vice versa). ChromeGames923 (talk) 05:33, 12 September 2021 (UTC)

How to get 11＋22＋22＋23 tone?

@Suzukaze-c This page says that tetrasyllabic words starting with a tone 5 character can be either 11＋22＋22＋23 or 22＋55＋33＋21. But am I correct in saying that Module:wuu-pron always lists the tones in such a situation as 22＋55＋33＋21? If so, how can I get the former tones to appear? (eg, 上海话大词典 claims that 日中心裡 is 11＋22＋22＋23 but the article doesn't reflect that.) ChromeGames923 (talk) 06:10, 11 September 2021 (UTC)

wuu-pron is a mystery to me. @Justinrleung —Suzukaze-c (talk) 06:12, 11 September 2021 (UTC)

@ChromeGames923, Suzukaze-c: The module only spits out 22+55+33+21 AFAICT. I'm not familiar enough with Shanghainese to know if both sandhis are available for all words. 上海市区方言志 seems to suggest that the two patterns are interchangeable (based on the examples). I'm not sure if certain words only use one of the two patterns. — justin(r)leung _{{ (t...) | c=› }} 06:20, 11 September 2021 (UTC)

@Justinrleung: I think you're right, in which case I suppose that would be good to acknowledge in our pronunciations. I wonder if it has to do with the tone mergers in Shanghainese, but I'm also not familiar enough with tones to say. (And while I'm on the topic, do you have any thoughts on the 2mm/3mm pronunciation from the above section?) ChromeGames923 (talk) 07:52, 11 September 2021 (UTC)

Just a note: we've added the LLLH chain. LHLL for tone 8 will be notated like 勿 from now on: ⁶x₈. See 突然之間 — nd381 (talk) 00:53, 26 March 2024 (UTC)

Wu Question

@ND381 The Suzhou pronunciation of the word 刮痧 seems to be like "kuaq", according to 現代漢語方言大詞典, but it doesn't seem to be part of the current module? Mar vin kaiser (talk) 18:27, 25 March 2024 (UTC)

It's kuaeq

Also, please exercise more caution when editing/adding Northern Wu entries. Muset and I have had to clean up mistakes you have made regarding pronunciation, labelling, etc. and I hope that you would be more careful in the future. Particularly, please label Shanghainese entries with "Shanghainese" and make sure the tone sandhi is correct in Suzhounese and Hangzhounese. The former is something that we frequently have to fix and the latter is something that would be a relatively newer feature and thus I would just like for you to be aware of this. Please read WT:AZH/Wu before editing. All of this, including the -uaeq rime, can be found there. Thank you — nd381 (talk) 00:52, 26 March 2024 (UTC)

Wiktionary:About Chinese/Wu#Initials table columns

Are the Wugniu and WT headers switched? ([2]) —Fish bowl (talk) 00:24, 28 March 2024 (UTC)

Where in particular do you think it's switched? I don't recall discussing this on the Discord — nd381 (talk) 22:44, 3 April 2024 (UTC)