Characters vs. romanization

I’m starting this thread to deal not with the romanization of Mandarin but with the use of Chinese characters for languages/dialects other than Mandarin.

An article in last Friday’s Taiwan News had something about this. I’ve been looking for followup information but haven’t seen any. So I’ll just get things started with this:

quote:
Minister of Education Huang Jong-tsun yesterday said the invention of a character system for the various Chinese languages is more important than choosing a romanization spelling system for the Chinese languages in Taiwan. ...

Huang said yesterday that the romanization spelling system could not completely represent a language and the phonetic symbols are just part of the language. “What is more significant is to build up a written character system, which would be the true identity of a language,” Huang said…

“The MOE has promised the Legislature to design a character system for Hoklo, Hakka, and the aboriginal languages within three years, and the current drafting of the bill will be completed in two months and submitted to the Executive Yuan for review,” he added.



In short, the Ministry of Education want to use Chinese characters, not romanization, not only for dialects of Chinese other than Mandarin but also for languages from completely different language groups. This is nothing short of astonishing.

I’m trying to figure out the reasoning – and, perhaps more to the point, politics – behind this.

Has anyone else heard news of this?

John DeFrancis, where are you when we need you?

My friend’s mom does it this way, and it’s the dumbest thing; Chinese isn’t even a phonetic language and she uses traditional characters to say English (sort of) words, right down to the Chinese tones; kind of takes away the whole purpose, doesn’t it?

Though i’ve been guilty of using bopomofo to drive some English words home…

Looks like a bunch of guys inventing jobs for themselves.

The idea of developing character systems for Chinese languages is not entirely stupid. After all, cantonese newspapers exist all over the world. I suppose this kind of system makes sense to native Chinese speakers. It isn’t going to be much use for your average foreigner, but then I guess that isn’t really the point.

Doing the same for Aboriginal languages seems a little harder to see the sense in. But having said that I do have aboriginal friends who write letters in the Tayal (spelling?) language using Chinese characters. I don’t think they follow much of a system. They just write each word using set of characters that sound the same. (some of my friends can’t even write roman letters at all).

A romanization system would seem far less hassle than the above method for aboriginal languages, but I guess some people have just got used to using Chinese characters to write.

Overall, I guess the plan could work OK for Chinese languages, but it seems a bit silly for Aboriginal languages. Roman letters do work rather well, otherwise most of the world wouldn’t be using them.

quote:
Originally posted by kiwi: Roman letters do work rather well, otherwise most of the world wouldn't be using them.

Most of the world? I don’t think America, some of Europe, Australia and New Zealand count as most of the world.

There is no need at all to “design” a character system for Chinese dialects. For the most part, the characters already exist - you just have to match them up correctly with the spoken syllables. Hokkien (“Taiwanese”) is one of the dialects farthest removed from Mandarin, but almost all the vocabulary in Hokkien that is different from Mandarin is derived from ancient Chinese. A great deal of research has been done on this, notably at Xiamen University. What we see in Taiwan is that spoken Hokkien is often sloppily matched up against inappropriate Chinese characters. For example, the Hokkien word for “fragrant” - pang - is often represented by the character 香 whereas it should be 芳. In Hokkien, the character 香 is pronounced hiu (Juba pinyin - I think it would be hniu in Xiamen University’s pinyin,) and it is only used in the sense of incense.

As to Hakka, it is more closely related to Mandarin than is Hokkien, and is also related to Cantonese, for which special dialect characters are already well established. So establishing the right characters for Hakka should be easier than for Hokkien. However, since I can’t speak a word of Hakka, I will leave any further comment up to our resident expert, Jidanni.

It is absurd to try to write Taiwan’s aboriginal languages with Chinese characters. They are related to the languages of the Philippines, Malaysia and Indonesia, which are now all written with Latin characters. The best way to develop Taiwan’s aboriginal languages is to spell them in a similar way to other languages of the same language family, like Tagalog and Malay.

quote:
Originally posted by sandman:

Most of the world? I don’t think America, some of Europe, Australia and New Zealand count as most of the world.


How about: All of North America, Central America, South America, most of Africa, Indonesia, the Philipines, East Timor, Vietnam, a large percentage of India, western Europe, parts of Eastern Europe, Australia, New Zealand, Micronesia, Polynesia and all of Antarctica.

quote:
Originally posted by sandman: Most of the world? I don't think America, some of Europe, Australia and New Zealand count as most of the world.
Add into the equation Central and South America, the Caribbean, Africa, Southeast Asia and China. That's right, while Taiwan is content to use Zhuyinfuhao for their pronunciation schemes, China, the heavyweight of world populations, uses Roman letters to phoneticize Mandarin. Throw Canada into the mix (How could you have forgotten Canada! [img]images/smiles/icon_confused.gif[/img] ), and I do believe that you have "most of the world". [img]images/smiles/icon_razz.gif[/img]
quote:
Originally posted by sandman:

Most of the world? I don’t think America, some of Europe, Australia and New Zealand count as most of the world.


I think you mean all of western, central and southern Europe (except Greece), Turkey, all of northern, central and southern America, most of Africa, most of southeast Asia, Australia, New Zealand and all other countries in Oceania.

***Oops - O’Brian, Maoman and I all posted at the same time. Great minds think alike. While we’re about it, let’s not forget Japan (romaji widely used) and India, Pakistan, Bangladesh, Sri Lanka, Nepal etc., where English is an official language.

Russian can also be spelled with Latin letters e.g. V samoe blizhajshee vremya budet otkryt novyj site projecta “Srednee Uho” s podrobnoj informaciej o koncertah i drugimi interesnymi i poleznymi materialami. (Source: http://ulitsky.freeservers.com/lat/aurismedia.html)

Another web page in romanised Russian: Biografia Aarona Rosanda (Biography of Aaron Rosand)

quote:
Originally posted by Juba: ***Oops[img]http://oriented.org/ubb/icons/icon11.gif[/img] - O'Brian, Maoman and I all posted at the same time. Great minds think alike.

I’m going to save this quote.

So my brain isn’t connected to my … erm… anything. But its my ball and I’m taking it home with me, so yar boo sucks to you all.

I agree entirely with Juba. Get a standard character system for Taiwanese. It’s not too hard, but pretty stupid to use Chinese charcters for aboriginal languages. I also agree with the education minister in that article where he says a standard character set for Taiwanese and Hakka is porbably more important (for Taiwan, not for us foreigners) than choosing the romanisation system.

Bri

It never ceases to amaze me how a country can educate the majority of its citizens to read and write more than 3,000 Chinese characters yet is completely unable to come to terms with the 26 letters of the Roman alphabet. (Which is the most universally recognised and used system of writing in the world).

That said I can never understand the argument that Tongyong pinyin has been adopted because it can be used to romanise other languages in Taiwan. Everybody in Taiwan needs to use Mandarin and almost all foreigners, even if they are proficient in Mandarin will never use other Taiwanese languages. It just seems crazy to use a half-arsed system of romanisation because, perhaps, a few hundred people might use it for other languages (when and if a system of romanisation using Tongyong for those languages is developed). In the meantime millions of people suffer.

And why can’t they develop a universal method for writing Taiwanese (in Chinese characters)? Surely if people are interested in promoting Taiwanese for reasons of politics or national identity the best thing they could do is develop a system a system for writing the language.

There is also the issue of language education. This seems to be another area where Taiwan fails dismally and perhaps another thread could be devoted to this.

I want to revisit this topic.

I think it’s generally agreed here that using Chinese characters for the languages of Taiwan’s tribes is inappropriate.

But I’m a bit surprised by the general talk of the use of Chinese characters for dialects other than Mandarin.

Why should the world’s most complicated writing system be applied to a language if simpler means are available? And as DeFrancis (link in my first post), working to dispel the myths about Chinese, says, “Chinese characters are a phonetic, not ideographic, system of writing.”

The political twists are also curious. This proposal is being put forward in a DPP administration. The DPP is not generally inclined to believe that traditional ways from China are superior to more “modern” approaches – quite the opposite, in fact.

Note, too, the curious and contradictory relationship between this proposal and tongyong. The big selling point of tongyong is its supposed unified approach – that is, letters represent specific sounds consistently regardless of the language/dialect.

But I think it highly unlikely that a character-based approach would follow this same principle. For example, if someone wanted to write “Banqiao” in characters, would that person need to use one set of characters for Mandarin, another for Taiwanese, and another for Hakka? After all, the name of the place is pronounced differently in each. To use different characters for the different sounds would involve an astronomical number of characters (esp. if the proposal to include the languages of the tribes is taken into account), such that no one would be able to remember them all.

There’s no way that’s going to happen. So that leaves supplementing Mandarin forms. No more “universalist” pronunciation approach.

Really, this is all extremely curious.

It’s not necessary that different character be used for different pronunciations. Many European languages have identical spellings for identical concepts with differing pronunciations (‘restaurant’). Japanese, Mandarin, Cantonese all use many of the same characters to denote the same thing. There is rarely confusion because you know which language you’re in, since the languages don’t overlap (anything near) perfectly.

Does anyone know what proportion of e.g. Hakka characters/syllables do not correspond to a mandarin equivalent? That seems crucial to the viability of using characters.

For comparison, how did written cantonese develop - did it have its own baihua written tradition?

quote[quote]"Chinese characters are a phonetic, not ideographic, system of writing." [/quote]

That’s a gross simplification. Leaving aside that many other factors go into them (I seem to remember 6 character ‘types’, phonetic probably being the most common), the sounds reflected are more often than not those of a long extinct language or one separated from modern Chinese by 1000+ years of language evolution.

Originally posted by salmon:

quote[quote]It's not necessary that different character be used for different pronunciations. [/quote] I know that, of course. Please reread my previous post. My point is that it is extremely strange that the Ministry of Education announces in one breath two fundamentally opposite approaches to language policy: pronunciation that is universalist (tongyong) and separatist (characters).
quote[quote]Does anyone know what proportion of e.g. Hakka characters/syllables do not correspond to a mandarin equivalent? That seems crucial to the viability of using characters. [/quote] Lumping together characters and syllables -- esp. when referring to something other than Mandarin -- is not accurate. But that aside, the question is interesting. I haven't seen a comparison for Hakka and Mandarin. DeFrancis cites a 1981 study of difference between Mandarin and Taiwanese as finding that "30 percent of the vocabulary as a whole is different (apart from the overall difference in pronunciation), a figure that rises to 50 percent in the case of function words (adverbs, prepositions, demonstatratives, measures, questions words, conjunctions, participles)."
quote[quote]That's a gross simplification.[/quote] Yes, but a more accurate one than most blanket statements (myths) that get passed around about Chinese. Here's a more measured statement: "There is much justification for considering the Chinese script to be basically -- that is, more than anything else -- a phonetic system of writing." From your statement below, it seems you would agree with this.
quote[quote]Leaving aside that many other factors go into them (I seem to remember 6 character 'types', phonetic probably being the most common), the sounds reflected are more often than not those of a long extinct language or one separated from modern Chinese by 1000+ years of language evolution.[/quote] Which is again my point: Why use the world's most complicated writing system, one that is predominantly phonetic but outdated by millennia in its application, for a language without a standard writing system?

I guess the western (English???) world stuffed it up a long time ago, giving different spellings and pronunciations to places in the world, even if those countries used roman writing. eg Koln - Cologne, Rome - Roma, Firenza - Florence, etc and even the countries too. Sure some of them are direct translations that are probably acceptable - like Etats Unis (USA in French) , but why cannot English speakers say Espana, Italia, etc

Maybe in China there has been the best recovery from this “stupidity”. Few English speakers now refer to my current location as Canton, and Peking is a long dead usage to my knowledge.

“Peking” is actually the Wade Giles spelling of “Beijing”. When China started using Hanyu Pinyin instead of Wade Giles (around 1960), the rest of the world finally started pronouncing “Beijing” the right way.

Originally posted by Mark Nagel:

quote[quote]"Peking" is actually the Wade Giles spelling of "Beijing". When China started using Hanyu Pinyin instead of Wade Giles (around 1960), the rest of the world finally started pronouncing "Beijing" the right way.[/quote]The Wade-Giles way of rendering Beijing would be Pei-ching. "Peking" is the old Chinese Postal System rendering. I don't remember hearing "Beijing" instead of the ridiculous "Pee-KING" in the United States until the late 1970s. Until then, Taiwan -- with its messed up Wade-Giles and Chinese Postal System remnants -- was "China." Perhaps the media in countries that recognized the PRC earlier (Britain, for example) caught on to proper pronunciations of Chinese much sooner than their U.S. counterparts.
quote[quote] Why should the world's most complicated writing system be applied to a language if simpler means are available? [/quote]

Becuase reading pinyin is shit. It’s much easier to read charcters once you know them. The nature of Chinese languages - the huge amount of homonyms - makes it most suited to a system like this.

I want to post a link or copy from the zhongwen.com faq, but I can’t. If you go there and look, they say that English has aprox 8000 syllables, whereas Mandarin has 1250-1300, or 400 if tones are ignored. I’m not sure but I think other Chinese languages are similar. That makes it real hard to read a purely phonetic system like pinyin.

Chinese could still be simplified to make it more phonetic, but still keep it’s special features. Ie you could take the 400 syllables and choose just one radical to represent the sound instead of the current situation where you would have a number of different radicals representing the same sound. Then you could meomorise those 400 and by looing at the phonetic part of a chracter you would know how to say it even if you didn’t understandt he whole character. But that’s all theoretical, becuase it’ll never happen, but I think tey could bear the principle in mind when standardisinmg a character set for Taiwanese and/or Hakka.

Bri