Talk:General Chinese/Archive 1

Middle Chinese

How does this project of romanised "General Chinese" compare to attemts to reconstruct Middle Chinese? There are hints at historical phonology:

"The difference between the yin and yang tones is indicated by the voicing of the initial consonant, which is possible because the original voicing distinctions are retained." (my emphasis)

But it's that easy. — Babelfisch 01:46, 9 November 2005 (UTC)

History

赵元任早在1967 年就在美国发表了《通字方案》，1983被翻译为中文在北京出版。“所谓通字（general Chinese）是为了研究和写作的方便，取中国语言当中的一部分作全部的代表”。通字以官话、吴语、闽语为基础，从常用字中挑出来的2085字。我们挑出的“西方汉语学习者写认读汉字表”主要是为西方部分汉语学习者学习汉语阅读写作用的字，我们提出区分“写读认汉字”与“读认汉字”与赵元任先生的《通用字表》有很多差别。(excerpt from 丁崇明: 西方汉语学习者汉字教学策略)

If that's correct, the reference given in the article (1983) is a bit misleading. There is also a lack of historical context: How was this project received? Was it ever used by anyone except Chao himself? There are have been hundreds of projects like this, and they were all ignored by Chinese linguists as well as by Chinese speakers - they're all one-man pet projects. Isn't that true for this one as well? — Babelfisch 01:46, 9 November 2005 (UTC)

I don't know if that's the case or not. But Chao is a significant figure in this area, and I think that alone makes it notable. kwami 02:24, 9 November 2005 (UTC)

It's not a popular project,but the spelling scheme and the phonological system shown in it are often used for reference.No government support nor powerful organisation promotion, it doesn't work in contemporary China.It's not an one-man pet,there are few people they do like this work, and got inspiration from it.Blankego 05:18, 2 July 2007 (UTC)

Although this proposal was never popularized, that doesn't lessen its importance. This is the finest example of of a number of proposals (including Lamasse and Jasmin) using historical spelling to represent Chinese, a potentially important innovation and counterpoint to the claim that only characters can unify all dialects. —Preceding unsigned comment added by 222.28.89.37 (talk) 15:51, 5 July 2010 (UTC)

Online write-up?

Is there any online write-up of General Chinese, preferably in English? Besides the Wikipedia article, the only resource appears to be Chao's original text, and I'm not sure where there is a copy outside the National Library of Australia. Amazon hasn't heard of it. m.e. (talk) 07:48, 3 October 2010 (UTC)

Notes

There's an early version at least partially visible here.

Branner 2006 The Chinese rime tables is partly visible online and discusses the development of GC. — kwami (talk) 10:36, 3 October 2010 (UTC)

Ao (1999) and Norman (2006)

The article says "More recent diaphonemic systems are Ao (1999) and Norman (2006)", but does not go into further detail. What are these works? Please cite properly. m.e. (talk) 15:22, 22 November 2010 (UTC)

More examples/info please

Currently there are only three characters (and not even any 入聲 examples). That doesn't do much to describe this diaphonemic system. Asoer (talk) 12:14, 6 December 2010 (UTC)

I've done the initials, but I don't want to type out the rimes by hand. Do you know of a site online that has the rimes for me to copy & paste? — kwami (talk) 22:58, 6 December 2010 (UTC)

Thanks. I'm not sure what you're looking for, but would the table on Middle Chinese#Finals help? Asoer (talk) 20:56, 9 December 2010 (UTC)

Yeah, that's it. — kwami (talk) 04:35, 10 December 2010 (UTC)

Hm, they're just irregular enough that I don't know if its worth it. Perhaps half of the GC rimes correspond to more than one MC rime. — kwami (talk) 08:57, 13 December 2010 (UTC)

expansion on sample

The sample text of this scheme that is supposed to represent multiple dialects only uses one dialect, this clearly is suboptimal, since it only shows GC vs Han vs Pinyin vs English; All that shows is Mandarin encoding. It should show something else, like Cantonese or Min-Nan (preferably both Standard Cantonese and Shanghaiese). 65.93.12.101 (talk) 06:42, 24 March 2011 (UTC)

It does show how little difference there is between character GC and VC, which was Chao's point for that example. — kwami (talk) 07:52, 24 March 2011 (UTC)

A well-rounded Min dialect would be a preferable inclusion since much research specifically excludes Min for being troublesome to include. LCS (talk) 16:31, 28 July 2011 (UTC)

Great work!

Before I ask my question below, I want to thank the editors of this page, above all kwami, for the great work improving it. I had looked at it a year ago, and couldn't do much with the information presented, but now it's become a treasure trove of information. — Sebastian 08:07, 1 September 2011 (UTC)

Glad it's helpful. — kwami (talk) 03:01, 4 December 2011 (UTC)

Onsets table

The table at Onsets has two extra columns for tenues and muddy consonants; which appear to be all fricatives, as opposed to the affricates further left. There is no top headline, but it seems that these two columns actually belong under the voiceless and voiced headlines, respectively. Personally, I would even put the fricatives before the affricates, being the more basic ones. Is there a reason why these columns are singled out? — Sebastian 08:07, 1 September 2011 (UTC)
Another way of arranging these would be by giving them rows, instead of columns. That would make the table more compact (60 cells instead of 66), but of course, the nice arrangement of having one row for each place of articulation would suffer. — Sebastian 08:17, 1 September 2011 (UTC)

OK, I think I found a good compromise between the two considerations; I put each of the pairs into one cell, one word above the other. Of course, side by side would make sense, too, but it wouldn't be so easy to align them nicely.

Now I have a different question about the table: What about "mh", "nh" and such; shouldn't they be in the table, too? Of course, they are no real phonemes, but the primary point of this table, after all, is not to provide an overview over the phonetic inventory of Middle Chinese, but to lay out the building blocks of Character-based General Chinese. — Sebastian 03:07, 3 September 2011 (UTC)

More consistency between the tables

Currently, the onset table looks very different from the other tables in that it has a much more complex structure for each entry. I hope we can find a way to make the way the tables work more consistent. The optimum probably lies between the two extremes; I propose to lay out all tables as follows:

1. Put the GC letters first, without brackets. Rationale: This is (half) the main topic of this article, and is what people want to find in the table in the first place.
2. Add IPA to all tables, except the tones table. (I presume there is no reconstruction of MC tones that we could use.)
3. Leave out the hanzi. While 音節索引 are a tried and proven tradition, I believe they are more to the benefit of people who are more familiar with Chinese characters than with latinizations, and don't provide much benefit to most of our readers. The benefit they do have is providing examples, but that can better be provided in a dedicated place, as we already have as part of the Romanized General Chinese and Sample text sections. Anyone who wants a more complete list can obtain that from the external links. — Sebastian 04:08, 3 September 2011 (UTC)

OK, people, how about that proposed change? As for the brackets, I should add that by now they look much better (at least on my screen), having been changed from ‹› to ⟨⟩ . So we could also go the other way, and introduce them consistently in all tables. That, however, would make some of them much wider, which would mean that a table like Combinations of medials and rimes might not fit on everyone's monitor. See also Talk:Grapheme#Brackets for graphemes. — Sebastian 00:27, 29 November 2011 (UTC)

Putting the GCR letters first makes sense, but the hanzi names are part of Chao's system, and some readers will find them useful for cross-referencing with Middle Chinese. The IPA, on the other hand, doesn't seem relevant – GCR provides formulas that cover diverse modern pronunciations, rather than a single pronunciation. Position in the table should suffice. Although his initials follow the traditional LMC initials, you'd have to more or less invent pronunciations for the finals.

One thing I would change about the initials table would be to put the fricatives in separate columns, following the traditional presentation as well as Chao, and exposing a bit more of the structure of his system. Kanguole 01:00, 29 November 2011 (UTC)

Oh, yes, good idea. I had arranged the fricatives in rows as a compromise (per my post of 3 September) but that can be changed. Now that you made me aware of the MC tables, I'm not sure why the arrangement is at all different from those. What is the meaning of the two "Semi-" rows; these two terms are nowhere explained. And why are 曉 and 匣 under "Gutturals"?

You raise a good point about IPA: That information is indeed redundant since it is already defined by the column and row headers. But I have to admit that it did help me in putting sounds to the onsets, since it's easier to see the pronunciation from the IPA character at a glance.(Actually, that might have been misleading; I'll start a new section about that below.) Still, I wouldn't object to leaving it out for the sake of consistency.

But doesn't the same apply to the hanzi in the onset table? Assuming the row and column labels are correct and consistent with the MC usage (see below), doesn't that provide the same cross-referencing? I'm also not sure if the currently used characters really lend themselves to cross-referencing; they are not consistently the same: E.g., 穿 is replaced by 初. Lastly, I don't see how they are a necessary part of the GC system; the system works nicely without them, as far as I can tell. In conclusion, I don't think any benefits outweigh the disadvantage of inconsistency: Why should we keep the characters in the onset table, if we don't have any in the rime tables? — Sebastian 01:35, 30 November 2011 (UTC)

It's not so much that the IPA is redundant, more that it is unfounded. The pronunciations given here are a strange hybrid of EMC and LMC distinctions. They don't come from Chao, who sets out to define a system that can be read with the sounds of a range of modern dialects. GC isn't Early or Late Middle Chinese; it's a reduction of those, keeping only the distinctions that are reflected in the modern varieties.

The nomenclature currently used in the article is traditional for MC, but a bit different from what Chao uses for GC. He really does call the series in question palatal stops. The table below shows his terminology:

		stops and affricates			fricatives		nasals	lateral
bilabial		b 幫	p 滂	bh 並			m 明
dentilabial					f 非~敷	fv 奉	v 微
dental	stops	d 端	t 透	dh 定			n 泥~娘	l 來
	sibilants	z 精	ts 清	dz 從	s 心	sz 邪
retroflex		dr 莊	tr 初	jr 崇	sr 生	zr 士	r 日
palatal	stops	dy 知	ty 徹	dhy 澄
	sibilants	j 章	ch 昌	dj 船	sh 書	zh 時
velar		c 見	k 谿	g 羣	x 曉	h 匣	q 疑
laryngeal		∅ 影; y~w 以; y~w~h 云

Here the stops and affricates are placed on separate rows. One could also spread them out, making a larger table. But there are strong parallels between the development of the stops and affricates, which is why it's usually done this way, I think.

Chao starts with the initials and rhyme groups of the traditional analysis, with their traditional names, and combines those not distinguished in any modern speech. He then renames some of them so that they give the right sounds in the modern varieties. Thus a modern speaker knows how "zh" sounds from their pronunciation of 時 in their own variety. The character names of the initials and rhyme groups are thus a significant part of the system. As for the difference between the presentation of the initials and finals, I would change the presentation of the finals, following Chao, to relate them to the traditional rhyme groups, like so (illustrated for velar initials and level tones):

Yunjing Rhyme	開 Open			合 Closed
	Div. I	Div. II	Divs. III & IV	Div. I	Div. II	Divs. III & IV
果假	-o 歌	-ea 麻	-ie 戈麻	-uo 戈	-ua 麻	-iue 戈
效	-au 豪	-eau 肴	-iau 宵蕭
流	-eu 侯		-ieu 尤幽
遇	-u 模		-iu 魚虞
蟹	-ai 咍	-eai 皆佳	-ei 祭廢齊	-uei 灰	-uai 皆佳	-uei 祭廢齊
止			-i 支脂之微			-ui 支脂微
咸	-am 談	-eam 咸銜	-iem 鹽嚴添凡
咸深	-om 覃		-im 侵
山	-on 寒	-ean 刪山	-ien 仙元先	-uon 桓	-uan 刪山	-iuan 仙元先
瑧	-en 痕		-in 真瑧欣	-un 魂		-iun 諄文
宕江	-ang 唐	-eang 江	-iang 陽	-uang 唐		-uang 陽
梗		-aeng 庚耕	-ieng 清唐青		-uaeng 庚耕	-iueng 清唐青
通	-ung 東		-iung 東
	-ong 冬		-iong 鍾
曾	-eng 登		-ing 蒸	-ueng 登		-iuic 蒸

Notes:

The current presentation is overly focussed on the vowel letters, but Chao never means these to stand for phonemes, or even diaphonemes. His unit of analysis is the final, and he happily simplifies the spellings as long as he can keep them distinct. Kanguole 16:53, 3 December 2011 (UTC)

The reason for focusing on vowel letters was not to claim they are diaphonemes, but merely to simplify the presentation. Providing all the details, as you've started to do, would be so involved that IMO it would need to be split off as an appendix. I think it would be best to keep the current presentation (or s.t. close to it), note that it is a simplification, and direct the reader to the full set of tables if they're interested.

As for the IPA, granted this is not a phonemic system, but IMO it would be nice to provide some warning that the palatal stops may sometimes be better thought of as retroflex, and the ⟨r⟩ as palatal. The IPA was a way of providing that dual identity. Perhaps better to use slashes, though, since that would make no claim as to actual pronunciation.

Mandarin w and r are not nasals, so that label could be a bit confusing.

It would be good to keep the 'voiced' and 'voiceless' columns, as the tones depend on these. That includes the laryngeals. Also, 云 and 以 are not distinguished in GC, so IMO we do not need to list them separately. —kwami (talk) 20:56, 3 December 2011 (UTC)

Maybe s.t. like,

		stops and affricates			fricatives		sonorants
		tenuis	aspirate	voiced	tenuis	voiced
bilabial		⟨b⟩ 幫 /p/	⟨p⟩ 滂 /pʰ/	⟨bh⟩ 並 /b̥/			⟨m⟩ 明 /m/
dentilabial					⟨f⟩ 非敷 /f/	⟨fv⟩ 奉 /v̥/	⟨v⟩ 微 /w̃/
dental	stops	⟨d⟩ 端 /t/	⟨t⟩ 透 /tʰ/	⟨dh⟩ 定 /d̥/			⟨n⟩ 泥娘 /n/, ⟨l⟩ 來 /l/
	sibilants	⟨z⟩ 精 /ts/	⟨ts⟩ 清 /tsʰ/	⟨dz⟩ 從 /d̥z/	⟨s⟩ 心 /s/	⟨sz⟩ 邪 /z̥/
retroflex		⟨dr⟩ 莊 /tʂ/	⟨tr⟩ 初 /tʂʰ/	⟨jr⟩ 崇 /d̥ʐ /	⟨sr⟩ 生 /ʂ/	⟨zr⟩ 士 /ʐ̥ /	⟨r⟩ 日 /ȷ̃/
palatal	stops	⟨dy⟩ 知 /tʲ, ʈ /	⟨ty⟩ 徹 /tʲʰ, ʈʰ/	⟨dhy⟩ 澄 /d̥ʲ, ɖ̥ /
	sibilants	⟨j⟩ 章 /tɕ/	⟨ch⟩ 昌 /tɕʰ/	⟨dj⟩ 船 /d̥ʑ/	⟨sh⟩ 書 /ɕ/	⟨zh⟩ 時 /ʑ̥/
velar		⟨c⟩ 見 /k/	⟨k⟩ 谿 /kʰ/	⟨g⟩ 羣 /ɡ̊/	⟨x⟩ 曉 /x/	⟨h⟩ 匣 /ɣ̊/	⟨q⟩ 疑 /ŋ/
laryngeal		∅ 影 /ʔ/					⟨y, w, h⟩ 喻 /j/

— Preceding unsigned comment added by Kwamikagami (talk • contribs) 21:26, 3 December 2011

I agree that voiced, etc labels are useful. Chao lists the nasals as m, v, n, r and q. I see he also puts r next to the palatal sibilants. He presents 云 and 以 as separate initials with overlapping spellings, though he gives very little detail.

It is difficult to justify giving pronunciations when Chao says "there is no such thing as General Chinese pronunciation" (p86), and says his system is only intended to be read with pronunciations according to the various modern dialects. No form of Chinese had /ʈ/ (EMC), /f/ (LMC) and merged /n/ and /ɳ/ (post-LMC). This system of pronunciations is an artificial combination not supported by our source.

Regarding finals, that table is not intended as a start, but as the whole thing. The finals take their fullest forms after velars; the forms after other initials can be obtained from these by dropping non-distinctive medials (with a couple of exceptions). Kanguole 01:47, 4 December 2011 (UTC)

Okay, that's good then. That table is helpful.

喻 comes from Chao, I think. We could just lump 云 and 以 together, as we do with conflated forms elsewhere. I don't see any point in listing them separately and then explaining in a footnote that they're the same.

We're not really giving pronunciations. The IPA shows the phonological correspondences between forms, which is what Chao's tables do. They no more give actual pronunciation that GC does, but they do make the parallels easier to follow. After all, Chao says that ⟨v⟩ is labiodental, and that's just as much instruction in how to pronounce it as /w̃/ is.

Chao may have called them 'nasals', but besides that being confusing, the relevant feature phonologically (and historically) is that they're sonorants. All sonorants affect tone the same, whereas nasality is irrelevant.

Perhaps we could show the conflation of palatal and retroflex in the table, like this?:

palatal	sibilants	⟨j⟩ 章 /tɕ/
	stops	⟨dy⟩ 知 /tʲ, ʈ /
retroflex
	sibilants	⟨dr⟩ 莊 /tʂ/

— kwami (talk) 02:58, 4 December 2011 (UTC)

Thanks for both your replies. I now understand why the hanzi are there, key for me was the statement "Thus a modern speaker knows how "zh" sounds from their pronunciation of 時 in their own variety.", and the explanation preceding it. I'd vote for adding that to the article.

As for the IPA, I now am weakly for taking them out. On the pro side: They are a good mnemonic, which is especially helpful since Chao's onset characters are not very systematic. (If we keep them, I would however urge to change the symbols to those for palatal sounds, since that discrepancy adds yet another level of confusion. On the con side, in addition to Koncuohle's arguments, I feel that it adds more clutter to the tables, and even more so if we try to introduce them to the other tables.

I agree with using "sonorants" rather than "nasals" in the table; it also has the advantage that 日 and 喻 fit better in the same column.

I have one request: Could we move the part covering the rime table (from "As for the difference between the presentation of the initials and finals" on) to a separate section? — Sebastian 05:17, 4 December 2011 (UTC)

Does the current version of mine above address the palatal/retroflex issue? — Kwamikagami 05:40, 4 December 2011 — continues after insertion below

Oh, I only just now saw the interlaced rows. It's a great idea. This certainly would be the perfect solution if our focus here were phonemics. However, at least as I see it, the row headings are primarily there to classify the GC letter combinations. Unfortunately, the split row prevents identification with an unambiguous name. — Sebastian 07:56, 4 December 2011 (UTC)

We could give both IPA transcriptions even if we don't interlace the rows. — kwami (talk)

Yes, they are intended as a mnemonic. I don't think it would be profitable to try extending them to the other tables.

Either of you know which dialect of Yue includes Bobai? — Kwamikagami 05:40, 4 December 2011 — continues after insertion below

Not me. Who would know about exotic languages here, if you don't?! — Sebastian 07:56, 4 December 2011 (UTC)

Darn. I keep seeing refs to 'Bobai dialect' (it's famous for having a 4-way split of entering tone), but have no idea where I should link it from. — kwami (talk)

What in the world is going on with this? Is anyone else getting the name of this page as Leonard Cohen, Talk:General Chinese ? — kwami (talk) 05:40, 4 December 2011 (UTC)

Weird - did you get this from the talk page history? This seems only a past issue, though; afaik everything on this page is still fine. — Sebastian 07:56, 4 December 2011 (UTC)

That's the change that popped up on my watch list. And Leonard Cohen itself isn't on my watch list. — kwami (talk) 21:03, 4 December 2011 (UTC)

It seems the revision number was off by one, and gave you a diff between a version of Leonard Cohen and a version of this talk page. Kanguole 01:14, 5 December 2011 (UTC)

The IPA /v/ gives similar information to Chao's labels (which would be present in the revised table), but the problem arises when you want to describe a pronunciation that differs from Chao's labels, e.g. in the case of his "palatal stops" above, which most people believe were retroflex in EMC. Kanguole 01:14, 5 December 2011 (UTC)

That's why I now list both. — kwami (talk) 02:56, 5 December 2011 (UTC)

But that is a description of EMC values, not GC, which the definitive authority says has no pronunciation except in the modern varieties. We ought to follow Chao in that regard, and add extra information about EMC, LMC, etc in footnotes or other separate text. Kanguole 08:29, 5 December 2011 (UTC)

By that argument, we have to remove all labels from the table, even though they're from Chao. "Voiceless velar stop" is a pronunciation, after all. The IPA transcription is no different; it merely makes the correspondences more explicit so that they're easier to follow. I suppose we could transcribe the palatals c cʰ ɟ to be more in line with Chao's descriptions. — kwami (talk) 10:02, 5 December 2011 (UTC)

The argument is that we ought not to give GC a hybrid phonetic description that is not supported by the authoritative source. Chao uses these labels to describe the rules that give the pronunciations in modern varieties; he's explicitly unconcerned with specifying a GC pronunciation, and avoids precision that isn't required for his purpose. To interpret his label "palatal stops" with precise IPA transcriptions would depart from his intentions, and would not help anyone. More useful would be a note that these initials are usually held to have been retroflex dental stops in EMC, and to have later merged with the the retroflex and palatal affricates everywhere except in Min. Kanguole 14:31, 5 December 2011 (UTC)

You're incorrect in two assumptions: (1) it is not a phonetic description, hybrid or otherwise. (2) it is no more precise than what Chao already gives. [k] is defined as a 'voiceless velar stop', and so is equivalent to saying 'voiceless velar stop'. Both conventions indicate pronunciation, regardless of Chao's claims to the contrary: a voiceless velar stop has voiceless phonation, velar position, and occlusive articulation, all elements of pronunciation. /k/, on the other hand, does not indicate a pronunciation—it is merely suggestive of pronunciation, much like CG transcription itself, and so is more in keeping with Chao's intent. — kwami (talk) 22:06, 5 December 2011 (UTC)

The velars (and the labials and dentals) are fairly straightforward, but it really comes unstuck when you get to Chao's "palatal stops". To give them as c cʰ ɟ would be to take the label overly literally, when Chao tells us there's no such thing as a GR pronunciation, and no-one thinks their realization in EMC was anything like that. But to give anything else means going beyond the source. As for Chao's intent, I think we should take him at his word. Kanguole 00:36, 6 December 2011 (UTC)

But their realization was like that. /c/ is simply a segment; it is less precise than Chao's wording. We could use /tʲ/ or /ʈ/ or /tʲ~ʈ/ or /♥/. It doesn't really matter which symbols we use, because none of them give an actual pronunciation. What they convey is the relationship between the segments, which is where romanized GC is somewhat obscure. — kwami (talk) 01:05, 6 December 2011 (UTC)

/♥/? Clearly this is pointless – I'll leave you to it. Kanguole 15:19, 6 December 2011 (UTC)

I prefer //, too, if we agree on having these entries at all. — Sebastian 05:49, 7 December 2011 (UTC)

"Coronals" and "Sibilants", "Palatal" and "Retroflex"

I have always been bemused by the arrangement of rows and columns in the onset table. I already posed one question about columns for tenues and muddy consonants above. Another question is: Why are "Coronals" and "Sibilants" presented on the same level, as rows? These are not alternatives: One is a place, and the other a manner of articulation. I now realize that that is the same as at Middle Chinese#Late Middle Chinese, so I suppose that it's something traditional, rather than linguistic. Equally confusing from a linguistic point of view is that the row "Palatal stops" contains the IPA symbols for retroflex consonants, which is quite the opposite. But that, too, may be some fad among LMC people, who don't seem to care about this distinction as it had vanished by LMC. Given that apparently no recent language has traces of these original palatals, I think it would reduce confusion if the column were renamed "retroflex". Is "palatal" really used here by Chao? If so, we should probably stick with it to remain true to our source, but it would be nice if we could at least add a parenthetical explanation. — Sebastian 01:35, 30 November 2011 (UTC)

Sibilant is as much a place as a manner of articulation, and in this case behaves as place. — Kwamikagami 21:22, 3 December 2011 — continues after insertion below

Well, in some cases, it may make sense to regard them that way (The Cambridge Encyclopedia of Language never does; neither does this definition, which I feel is clearer than our current one), but that's beside the point. My point is that sibilant vs coronal is an absurd distinction, since some of our most common sibilants are coronals. — Sebastian 05:20, 4 December 2011 (UTC)

The palatals and retroflexes have been conflated to some extent, and sometimes are more profitably thought of as one, sometimes as the other. "Retroflex" ⟨r⟩, for example, is palatal /ɲ/ in many dialects, whereas "palatal" ⟨dy⟩ is retroflex /tʂ/ in others. Agreed, a note would be helpful. — kwami (talk) 21:22, 3 December 2011 (UTC)

Terminology

I'm getting confused with some of the terminology such as 'coronal' which is used in the article (eg under Combinations of medials and rimes), but not defined anywhere. Comparing the table with the Middle Chinese article, 'coronal' must mean dental (= 'alveolar'), retroflex or palatal; please reply if this is wrong. m.e. (talk) 09:35, 14 July 2012 (UTC)

(Also there, -on is described as rounded, but in the table it is in the unrounded column.) m.e. (talk) 09:12, 14 July 2012 (UTC)

Korean, Japanese, and Vietnamese

The article says "It can also be used for the Korean, Japanese, and Vietnamese pronunciations of Chinese characters", which is very intriguing. However, it does not really deliver on this - there's nothing beyond the short teaser in the introductory section Romanized General Chinese, and a parenthetical mention in the Onsets section. I wish there were information that helps readers to actually apply the system to those languages. — Sebastian 04:19, 3 September 2011 (UTC)

The problem is entering the data. It's too great a task for the time I have available. If you know of a computerized database it would be doable. — kwami (talk) 03:05, 4 December 2011 (UTC)

The Unihan.txt[1] file contains pronunciation data, for "Mandarin, Cantonese, Tang-dynasty Chinese, Japanese, Sino-Japanese, Korean, and Vietnamese". m.e. (talk) 08:25, 20 June 2012 (UTC)

The quality of that data is suspect. There are obvious typos that I’ve found in the data for Tang pronunciations (such as “bgyɛn”). I suspect there are many more since I applied some transformations to the Cantonese data as well and found some improbable pronunciations for some characters. YMMV. — LCS (talk) 00:38, 31 July 2018 (UTC)

I can imagine! 3 months ago I found at least lists with character pronunciations for Mandarin, Vietnamese and Korean, and I combined that with the GC pronunciation list. However, I only got some general rules for Mandarin, with the help of Four tones (Chinese) and some original research, like the following:

Onset* \ tone	平Level	上 Rising	去 Departing	入 Entering
C tenuis	˥	˨˩˦	˥˩	˧˥ for frequent ones, ˥˩ for most, or any other
K aspirate				˧˥ or ˥
Q some sonorants	˧˥			˥˩
G muddy (affric., obstr.,..)	˧˥ → K	˥˩ → C	˥˩ → C	˧˥ → C

* C, K, Q, G are my abbreviations for the columns. (I just used the letters for the velars, since these have single letters for each column.)

I did something similar for the onsets, but it's in a less compact form. More importantly, it's even more OR, and if I did that for Vietnamese and Japanese, it would be completely OR. I stopped when I realized that I wouldn't be able to use this here. IIRC, this is why I asked if something like this exists. — Sebastian 07:29, 4 December 2011 (UTC)

Baxter I think has some of the correspondences, but I don't know what's publicly available. — kwami (talk) 05:38, 5 December 2011 (UTC)

The Vietnamese phonological system allows for finals with both /w/ as an on-glide and off-glide (e.g.: khuỷu tay). Middle Chinese had no such finals. Unless there are some rules that create such syllables from Middle Chinese, I’m skeptical that General Chinese can provide much coverage for Vietnamese. Vietnamese has several layers of borrowings and re-borrowings and the General Chinese is much too simple to account for all the layers. I’m not even sure if comprehensive coverage was even intended by Chao. LCS (talk) 00:26, 31 July 2018 (UTC)

Combinations of medials and rimes

Matching the new table of combos with the existing one isn't as straightforward as I expected: It apparently lists fewer rimes - e.g. "-a", "-iue" and "-an" seem to occur only in the old one. Also, is the sequence of rows tradional, or can it be arranged so that it better corrensponds to the first table? — Sebastian 17:59, 9 December 2011 (UTC)

The only rhyme that seems to be missing is -ia. As for rearranging, I don't know if there's any particular reason to stick to the current order, which isn't particularly intuitive, but it won't be possible to get them completely in sync. — kwami (talk) 20:32, 9 December 2011 (UTC)

Hmm, am I blind? Where is e.g. "-a"? As for the order, of course it's not completely possible, but even a bit would help, don't you think? — Sebastian 22:02, 9 December 2011 (UTC)

You're right. I was only looking for combos. The simple finals -a and -an are missing, as you say, plus -ia. Chao noted some sort of conflation w -o, but I forget what: maybe it is lost after velars? — kwami (talk) 22:06, 9 December 2011 (UTC)

Thanks to the above misunderstanding, I realize that "Combinations of medials and rimes" can be understood in two ways. I understood it to include the medial Ø, while Kwami excluded that. Based on my understanding, I just assumed that the head row of the first table just doubled as the Ø row. Eventually, I think it would be better to include Ø and dedicate a row for that, but I'm not proposing an immediate change since I first want to find agreement about consistency among tables (see section "consistency" above).

Maybe Kwami intended the second table to fill in the information missing in the first, namely the the hanzi, which, from a consistency point of view, are missing there. Using the second table for that is an interesting idea. On one hand, that keeps the first table from growing beyond the common page width. On the other, however, it requires an additional look up step, which counters our desire to make the information more accessible. — Sebastian 00:57, 10 December 2011 (UTC)

Improvable wording

Came across this while binging on text-to-speech stuff, and noticed that the ending of the first introductory paragraph: "and challenges the claim that Chinese characters are required for interdialectal communication in written Chinese." is a bit spotty. In it's current form, it's roughly equivalent to: "and challenges the claim that the color cyan is an example of blue". I'm familiar enough with the Chinese writing system to get the meaning that written Chinese communication can be accomplished with pronunciation characters instead of "whole meaning" characters, but I feel that the wording can be improved for clarity. Maybe a single change like: "and challenges the claim that Chinese logograms are required for interdialectal communication in written Chinese."?

Forgot the sig, sorry. 2602:301:7764:AC00:31ED:B39B:6908:AB64 (talk) 23:26, 20 May 2017 (UTC)

It seems the OP only read the first section - Character-based General Chinese . If GCC were all there were to GC, then it could indeed only prove that statement for “logograms”, not for “Chinese characters”. ◅ Sebastian Helm 🗨 07:44, 25 February 2024 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified one external link on General 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/20110720022911/http://ishare.iask.sina.com.cn/f/8841122.html to http://ishare.iask.sina.com.cn/f/8841122.html

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

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 02:35, 7 January 2018 (UTC)

Syllabary

Where is this fabled syllabary of 2,082/2,085 glyphs? I cannot find a list. The English and Chinese Wikipedia articles also differ on the count by 3. The list from *rimeime* (which is referenced in this article) has 2,079 distinct syllables—one of which is just “r”. The article does not mention syllables with only an initial consonant and no final. (Maybe the list is wrong.) — Preceding unsigned comment added by LCS (talk • contribs) 03:13, 30 July 2018 (UTC)

The source of our frustration

It seems to me we are trying the squaring of the circle. On one hand, we want to be instructive to our readers, on the other hand, we want to be true to our source. Since Chao himself put such considerations as notational brevity over consistency and learnability, the two are sometimes in conflict. Since Wikipedia has to strictly stick with its sources, this article can not be as easy to understand as we would like it to be, no matter how hard we try. One way to solve this would be to keep the textbook text in a different location. How about http://en.wikiversity.org/wiki/Topic:Chinese for that? I would enjoy working with the two of you on that. — Sebastian 05:49, 7 December 2011 (UTC)

Maybe a comparison can clarify how I see it. This feels to me as if we were writing an article on classical mechanics while slavishly clinging to Newton's antiquated mathematical notations. GC is a great concept, but its realization/notation has its idiosyncrasies, making it hard to learn, which seems to me a big factor in why it hasn't caught on. Unfortunately, linguistics isn't as logical as physics, so it's much harder here to just take the good part of a concept and change the notation. I'm writing this to point out that the frustrations we have with each other in trying to put sense into our tables are not any editor's fault, but are inherent in the subject. — Sebastian 17:47, 7 December 2011 (UTC)

These idiosyncrasies annoy me. I’m new to this, and would it would make understanding so much easier if I didn’t have to comb through a spreadsheet trying to match up these convoluted spellings with actual diaphonemes. The media+nucleus chart is a good start, but it doesn’t exactly help me work backwards to get the diaphonemes. Should I get a syllable containing *aeng*, how do I know it’s not *eaeng*? There should be some disambiguation algorithm described for people who are less intimate with the topic. What should the spelling look like before discrepancies are introduced? --LCS (talk) 03:07, 30 July 2018 (UTC)