闽南语学习网 & 闽南语教程

[Andrew]

1. G is approximated by Ng. Minnan has G, K, K' whereas mandarin/cantonese only has K, K'.

Then how do you represent ng- , as ngO 5 (literary)?

[Mark Yong]

tadpole wrote:
Another thing is that Minnan has many series of contractions: two syllables contracted into one syllable. If you use character writing, you need to invent many more characters to take care of the contractions.

Correct. But then again, Mandarin also has its fair share of contractions, too - 別 for 不要 and 甭 for 不用 being two examples. These have been accepted into the Standard Modern Chinese lexicon (though, here I vehemently voice my objection to them ).

Apart from contractions, Standard Modern Mandarin has a whole plethora of borrowed characters 假借字, e.g. 這, 那. To be fair, even Hong Kong Cantonese has its fair share of borrowed characters 假借字 and incorrectly-assigned ones (by that, I mean the actual characters exist, but are not used).

tadpole wrote:
Using characters is to show friendliness towards other Chinese dialects...

Agreed. In a previous thread, I voiced my disapproval at how the "Mandarin-isation" of written Chinese was at the expense of isolating the other dialects, and that the other dialects going about creating their own characters to represent their own quirks would further fragment written Chinese's unique feature of binding the dialects together.

My conclusion: If Romanised script is used, then by all means, show the contractions. However, if Chinese characters are to be used, then I would suggest that the contractions be ignored and the correct individual morphemes be retained in the written form. The reason being, even if the morphemes are bound ones, they are still morphemes that carry a particular lexical or grammatical function to the phrase as a whole.

To give an example in Hakka: The negative "not yet" (還沒有 in Mandarin, and simply 未 in most of the Southern dialects) is pronounced măng/miăng in Standard Moi Yen Hakka 梅縣客家. It is really a contraction of the two individual morphemes mŏi-ch'æn(g) 未曾. Granted, writing them as individual characters does introduce a certain degree of artificiality, in that the natural process of contraction has not been captured. But the advantage is that when written in their actual morphemes, it becomes clear to the reader (of any dialect) what it means (the two characters 未曾 translate as "not yet <v./adj.> before"). Compare that with 盲 'blind' as the commonly-adopted borrowed character 假借字, which phonetically may a convenient choice, but written-wise makes no sense. My opinion is that contractions (as well as shifts in tones, etc.) will continue as a natural evolution of the language, even if Chinese characters continue to distinguish the individual morphemes.

'Now/at present' is pronounced yĭ-gâa or gâa-hāa in Cantonese, which are really 而今下 and 今下, respectively - with the final -m lost from 今 when 今下 fused (the distinction in 今下 is still retained in Hakka, i.e. kīn-hâ). However, Hong Kong Cantonese has taken to writing it as 而家, which works fine for the local Cantonese readers, but befuddles just about anybody else who does not speak the dialect.

Another reason for my slant towards more use of Chinese characters and less Romanisation is because of the homonyms 異義同音字. Now, one can argue (as did the Chinese proponents for Romanised script for writing Chinese during the last century) that context within the sentence normally resolves the ambiguity. But how often is that really so? And especially so for the Southern dialects like Hokkien, which do not have as many compound words as Mandarin? Even in the case of the Korean language, despite the almost-universal use of hangeul, official documents still make use of Chinese characters to resolve ambiguities created by the homonyms. I own a copy of Nicolas C. Bodman's Spoken Amoy Hokkien - it is written entirely in Romanised form complete with tone markers, and quite well, I must say. But even then, I find myself struggling at times to figure out which morpheme a particular Romanised syllable refers to (albeit for a second or two, but try compounding that seemingly-negligible time delay per word across a full-page text!), so much so that I have resorted to foot-noting the ambiguous ones with hand-written Chinese characters!

Perhaps a new thread is warranted here, as I stand totally guilty of throwing everyone off-tangent from the original thread topic!

[SimL]

xng & tadpole:

Thank you to you both for the explanations. I have to apologise for being too negative, based purely on my own ignorance. I had of course heard that Quanzhou was a major variant of Hokkien, but I hadn't realised that its phonology was different on the two points about initial consonants which I referred to.

(Having said that, it is a pity to have a major effort on Hokkien based on a variant which doesn't have distinctions which are so important in other variants. While *any* variant is just as valid - and worthy of study and presentation - as any other variant, such a variant is less helpful for speakers of the other variants than the other way around. What I mean is that a Quanzhou speaker looking at a Amoy website can just merge the distinctions he/she sees, whereas an Amoy speaker looking at a Quanzhou website can't *create* the distinctions (if he/she doesn't know the words to begin with).)

Thanks too for the explanation of "-h". It seems strange to me to accept "-p", "-t", "-k" (which would also have to be marked as ru-tones), and not accept "-h", but ok. I think a Hokkien enthusiast once said to me that anyone interested in learning *about* Hokkien just has to accept a large variety of transcription systems - an opinion which I (reluctantly) had to agree with. (I also once read somewhere that the definition of a "real sinologist" is "someone who has designed his or her own improvement of pinyin" .).

Mark:

No problem that the topic of the thread drifted - it happens on all forums, and I think this is a really great discussion. I agree with parts of what everyone is saying, so it's hard for me to write a detailed response. But I really enjoy reading the different points of view about how best to write Hokkien.

[xng]

xng & tadpole:

Thank you to you both for the explanations. I have to apologise for being too negative, based purely on my own ignorance. I had of course heard that Quanzhou was a major variant of Hokkien, but I hadn't realised that its phonology was different on the two points about initial consonants which I referred to.

I didn't know that quanzhou differs so much from xiamen.

Are there any good xiamen online website around that is as good as UCLA ?

Or somebody can email UCLA to provide two phonologies, one for quanzhou and one for xiamen.

[SimL]

Tadpole:

I really wish there are more 泉州 dialect materials on the net. UCLA has an OK Quanzhou dictionary (http://solution.cs.ucla.edu/~jinbo/dzl/lookup.php). But as for audio-visual materials, there is just not much on the net. Youtube has a few story-telling video clips in Quanzhou dialect, but that's about it.

You said this quite a while ago, but I'd like to ask you to post links to this sort of story-telling clips here (Quanzhou and non-Quanzhou). Every time I search on YouTube for "Hokkien" or "Minnan" or "Taiwanese", I get 99.99% songs. Songs are well and good, but because tones are lost in singing, I think a lot of the feeling and flavour of any particular variant of Hokkien is lost in a song. Story-telling, or any other sort of just plain speech (like the bits in Niuc's architectural clips) would be many times better, and I'd love to see and hear such clips.

Thanks.

[Andrew]

niuc wrote:
As languages evolve, they affect one another, and invariably absorb the best of each other's features. Cantonese in Hong Kong has evolved today, in no small part because by the adoption of the Mandarin model as the written standard, it facilitated the absorption and adoption of Modern Mandarin vocabulary that emerged from technological and commercial jargon of the last century, thus allowing it to evolve alongside Mandarin as a result (to a certain extent, this evolution also took place with the Shanghai dialect). Hokkien, however, suffered this disadvantage of not being able to get on-board the mainstream of evolution in the Chinese language as a whole. I have little doubt that had Taiwan not suppressed Hokkien in the years between 1949 to the 1990's, the evolution of Hokkien would not have been as stunted as it is today.

I don't speak Cantonese, but the impression I get is that when reading written (Mandarin) Chinese they often read specifically Mandarin phrases literally, e.g 我們 and 我的. You could do the same with Hokkien, with ngO-bun and ngO-tit, but most of us just wouldn't accept that as being Hokkien. I suppose the same could apply equally to 文言文, but for the fact that there is a long history of Hokkien Hanbun and so it doesn't seem so strange.

I think for the Chinese-educated there is not the same problem with the non-readability of Hokkien because they can map many words and phrases between Hokkien and the same words in written Mandarin Chinese. The problem arises with the words that are no longer used in Mandarin, but this is simply a matter of teaching a limited register of "Hokkien" characters.

[niuc]

Hi Andrew

Your quote has Mark and my id, actually that's from Mark's posting.

The 2, 3 or more tiers of hanji pronunciation (usually termed colloquial vs literary) in Hokkien usually have separate meanings. 大人 can be pronounced as tua7-lang5 or tai7-jin5, but they mean differently. This indeed makes reading hanji in Hokkien much more difficult than Mandarin or Cantonese.

[Mark Yong]

Andrew wrote:
I don't speak Cantonese, but the impression I get is that when reading written (Mandarin) Chinese they often read specifically Mandarin phrases literally, e.g 我們 and 我的. You could do the same with Hokkien, with ngO-bun and ngO-tit, but most of us just wouldn't accept that as being Hokkien. I suppose the same could apply equally to 文言文, but for the fact that there is a long history of Hokkien Hanbun and so it doesn't seem so strange.

I think for the Chinese-educated there is not the same problem with the non-readability of Hokkien because they can map many words and phrases between Hokkien and the same words in written Mandarin Chinese. The problem arises with the words that are no longer used in Mandarin, but this is simply a matter of teaching a limited register of "Hokkien" characters.

The same problem may exist in the other direction, i.e. mapping Mandarin words to Hokkien pronunciation. A Chinese-educated person reading a Chinese text today would intuitively render 跟 as tue/te instead of kin1.

I agree with you that one could render 我們 and 我的 as ngO-bun and ngO-tit, but like Hong Kong Cantonese, it would have to be taught as such before it can become intuitive to the reader. Therein lies the gap - and the difference between Cantonese in Hong Kong vs. Hokkien just about everywhere on the globe (except maybe Taiwan today). I often wonder how the Hokkien newsreaders on Malaysia's 106.7FM radio station manage to do it without such formal education (though, I do catch a few of them struggling at times).

niuc wrote:
大人 can be pronounced as tua7-lang5 or tai7-jin5, but they mean differently.

Interesting example you chose, niuc. A similar problem exists with 人 yan in Cantonese. When pronounced with a low-flat tone, it simply means 'person'; when pronounced with a rising tone, it refers to the gender of the person.

Such multiply-pronunciation/tone ambiguities also exist in Classical texts. In the 論語 Confucian Analects, we have:
"齊景公問政於孔子，孔子對曰：“君君、臣臣、父父、子子。"
For each couplet, the first character is a noun, and the second character is a verb. So, the first 父 means 'father', and the second 父 means 'to be/act like a father'. And depending on whether it is a noun or a verb, the tone is different.

Of course, the difference here is that these Classical texts have been annotated, such that the ambiguities are well-documented and clarified to the readers over the centuries. I guess in the case of written Hokkien using Chinese characters, in the case of resolving the ambiguity of 大人, one would either have to write 'adult' as 成人 seng-jin, or if 'adult' really must be written with the Chinese characters 大人 for any particular reason, that some annotation be built-in to clarify what reading to use (similar to the kana markers found alongside kanji in Japanese texts today). In any case, my opinion is that while such ambiguities admittedly do exist in Hokkien, they are relatively few in number to disrupt the use of Chinese characters.

[niuc]

Hi Mark

A similar problem exists with 人 yan in Cantonese. When pronounced with a low-flat tone, it simply means 'person'; when pronounced with a rising tone, it refers to the gender of the person.

How is it indicated in writing to point out the correct pronunciation? Is it purely by context?

Such multiply-pronunciation/tone ambiguities also exist in Classical texts. In the 論語 Confucian Analects, we have:
"齊景公問政於孔子，孔子對曰：“君君、臣臣、父父、子子。"
For each couplet, the first character is a noun, and the second character is a verb. So, the first 父 means 'father', and the second 父 means 'to be/act like a father'. And depending on whether it is a noun or a verb, the tone is different.

Thank you for the information! That's interesting! I know about the noun-verb thing but not the tonal difference. I suppose the tonal difference is lost in Mandarin and Hokkien, or is it not? Could you help to give the romanization with tones of " 君君、臣臣、父父、子子 " ? Which Chinese languages still preserve this?

Of course, the difference here is that these Classical texts have been annotated, such that the ambiguities are well-documented and clarified to the readers over the centuries. I guess in the case of written Hokkien using Chinese characters, in the case of resolving the ambiguity of 大人, one would either have to write 'adult' as 成人 seng-jin, or if 'adult' really must be written with the Chinese characters 大人 for any particular reason, that some annotation be built-in to clarify what reading to use (similar to the kana markers found alongside kanji in Japanese texts today). In any case, my opinion is that while such ambiguities admittedly do exist in Hokkien, they are relatively few in number to disrupt the use of Chinese characters.

Yes, using of markers would be great. I think ambiguities regarding dual (colloquial & literary) or more layers of pronunciation is not that few. However, it depends on how the words are written in hanji. I have Hokkien ("Amoy Vernacular") Bible and Hymnal written in full hanji. The former is a "transliteration" from POJ version and it uses POJ in brackets to clarify the ambiguous readings. The latter uses small circle mark beside each hanji that has literary reading, occasionally having footnotes in POJ. Regardless the methods, again the lack of standardization is a big disadvantage.

[Mark Yong]

niuc wrote:
How is it indicated in writing to point out the correct pronunciation? Is it purely by context?

I'm afraid so. I have never seen any Cantonese texts with tonal markers or annotations.

niuc wrote:
I know about the noun-verb thing but not the tonal difference. I suppose the tonal difference is lost in Mandarin and Hokkien, or is it not? Could you help to give the romanization with tones of " 君君、臣臣、父父、子子 " ? Which Chinese languages still preserve this?

Actually, the tonal difference is preserved in Mandarin. I am writing this from out-of-town, so unfortunately I do not have my reference material with me. But off-the-cuff, I do recall that in the case of 王, wang2 means 'king' and wang4 means 'to be king' (i.e. 'to rule').

I am not sure if such noun-verb tonal differences are well-preserved (if at all) in Hokkien, but I do know that pronunciation differences are preserved in Cantonese. One example that comes to mind is 度. When pronounced dou, it is a noun meaning ' degree' (in the sense of a measurement); when pronounced dOk, it is a verb meaning 'to measure (normally the length of an object)'. No other major dialect group I know has preserved 度 as a verb.

niuc wrote:
I have Hokkien ("Amoy Vernacular") Bible and Hymnal written in full hanji. The former is a "transliteration" from POJ version and it uses POJ in brackets to clarify the ambiguous readings. The latter uses small circle mark beside each hanji that has literary reading, occasionally having footnotes in POJ.

Wow... how in the world did you manage to get hold of a Bible and Hymnal in Hokkien using full hanji? Could you share a scanned page of it, or perhaps some sample verses (Psalm 23 or John Chapter 1 comes to mind!)?

If the Hymnal was written with Hokkien readings in mind, then I am assuming that it dates back to the 19th century, i.e. the Literary Chinese period. If so, then the bulk of the words would use literal readings, which means the majority of the hanji would have small circles beside them, right? Or was the Hymnal deliberately written in fully-colloquial Amoy? And if so, how did it get around those words with no known hanji?