Jonathon Delacour has figured out what’s going on with the Sino-Japanese variant character he couldn’t find or display, and he shares his findings in his latest post. I would have made it an addendum to my earlier post on the subect, but I liked this quote so much I decided to give it its own entry: “I was able to find Kafu’s boku on my Windows PC with the aid of the Mojikyo Character Map because the Mojikyo Institute regards Chinese characters as ‘a very important cultural asset of the human race’ and—like Apple—is committed to making that wonderful variety of characters widely available.” I thoroughly agree about the cultural importance of characters, which is why I’m so torn about the issue of romanizing Chinese; from a practical point of view the arguments for it are unassailable, but there will be a real cultural loss, even though the characters will still be there for those who wish to devote the necessary years to learning them. Sometimes I feel like a one-man culture war.
Addendum. OK, ignore my tossed-off, oversimplified, and unduly provocative formulation and go straight to Eric Lien‘s detailed analysis of the problems and possibilities of simplified spelling systems for Chinese. (Romanization is probably not a good idea.) As always, I’m grateful to have readers who have given more thought to a subject than I have.
Further addendum. Jim of UJG has been following up on the topic, with a post on Chinese and another on Hmong. The first quotes the marvelously named Wi-vun Taiffalo Chiung‘s “Old Taiwanese saying” Hanji na thak e-bat chhui-chhiu to phah si-kat (‘You cannot understand all the Han characters even if you studied until you could tie your beard into a knot’); googling Taiffalo led me to his paper Romanization and Language Planning in Taiwan, which discusses the early (Dutch) romanization system “Sinkang writing” and the late-19th-century Peh-oe-ji romanization, revived by the Taibun movement in the 1980s. You can read more about Taiwanese (Holo or Hoklo, a variant of the Amoy dialect of Southern Min Chinese) here, and there’s even a Wikipedia in Taiwanese.
MORE ON MOJIKYO.
October 5, 2003 by 24 Comments
I’m so torn about the issue of romanizing Chinese; from a practical point of view the arguments for it are unassailable
There are some very good reasons not to romanize Chinese. The more Chinese I learn, the less appealing romanization seems to me. Not just because I’ve made a personal commitment to learning all those characters, but because it is simply easier to read something in characters than in pinyin. Don’t get me wrong, I love pinyin, and I use it for looking words up in the dictionary as well as computer input – but for reading characters strike me as superior. The structure of Chinese is such that there are a limited number of sounds – hence the need for tones to distinguish words in speech and characters for distinguishing them in writing. But characters work better than tones. I can’t tell you how many times when speaking to someone I’ve seen them clarify what they mean by drawing the character on the palm of their hand and then showing me the palm – as if I could still see the character they wrote with their finger ….
Addendum – there is one person who came up with an ingeious idea for reforming Chinese writing which I actually like. It would entail the use of radicals and romanization together. This seems like an ideal system – the radicals would give you a sense of the meaning (fire/water/beast/etc.) and the phonetics would give you a sense of the pronounciation….
Wait, wait, wait. People are actually considering using romanized Chinese instead of the characters, rather than in conjunct with the characters? I’m confused.
If the issue is pronunciation, I learned how to speak Chinese using a sort of phonetic alphabet that used characters (Zhu Yin Fu Hao) to represent sounds–in conjuct with the word itself. The characters are written beside the word as a pronunciation aid. The, err, spirit of the characters are preserved in a sense, and no romanization is necessary. I hear that it’s still used in Taiwan but no longer popular on the mainland.
The idea of replacing Chinese characters with phonetics is a very old one – and it gained some popularity in the 1920s when the “Baihua” or vernacular langauge movement sought to overturn the orthodoxy of Classical Chinese, which was seen as preventing China from modernizing. In the end, China chose to simplify the characters instead of eliminating them altogether, but some even now regret this change, and my understanding is that traditional characters are used more and more on the mainland (although it is unlikely that they will make a total comeback).
As far as romanization is concerned. There are MANY different systems. Zhu Yin Fu Hao is in the process of being replaced in Taiwan, but even though the government has already chosen a replacement romanization system, there are many who prefer to use the system currently in use on the Mainland. The result is that some streets are marked in one system, with other streets marked in another. Each mayor seems to chose a system depending on what political party they belong to.
This is an interesting site on Taiwanese orthography.
See #12 here for different Romanization conversion charts.
I just wanted to address some confusion regarding Tao Zi’s comment. Although these symbols are not made using roman letters- there is essentially no difference between such a system and Hanyu Pinyin (as the charts listed above demonstrate). While there are some trade-offs between various systems, they only represent the sound of the character, not anything about the meaning, as is (often) done in Chinese characters.
The more Chinese I learn, the less appealing romanization seems to me.
Exactly! The more Chinese (characters) anyone learns, the less appealing romanization seems. That’s not the issue. What you have to consider is the massive number of people who speak Chinese but are functionally illiterate because the characters require too much of an investment in time and effort. Traditionally they were the province of a tiny elite that had the leisure to spend years on such a demanding course of study; the Communist government has put a great deal of effort into teaching the masses (and of course claims absurdly high levels of literacy), but there will never be mass literacy unless and until the characters are replaced with a simpler system, whether romanization or some sort of syllabary. Believe me, I know how you feel; I resisted the conclusion for years, even though I personally had learned only a few hundred characters and thus had a minor investment. But eventually I was convinced. Think about it: how can someone who has to work most of each day just for food and shelter, whether peasant or city dweller, maintain such an intricate system? Even if they truly learned the mandatory few thousand characters in school, they quickly forget them for want of repetition. There have been studies done on this; I don’t have a bibliography to hand, but it shouldn’t be hard to find some titles, and if you give it a little thought, it should be intuitively obvious once you get past that resistance caused by your personal attachment. I’m sure the ancient Hittite scribes, once they had mastered their incredibly complicated writing system with its three different linguistic layers (a given cuneiform symbol might be read as Sumerian, Akkadian, or Hittite), would have fought tooth and nail against any attempt to replace it with a simpler system. But of course in those days writing systems weren’t meant for the masses. Chinese characters are a hangover from those days—the downside of a civilization that’s been basically uninterrupted for millennia.
I disagree about literacy rates. Taiwan has very high literacy rates, and uses Traditional characters. I’m not convinced that Chinese literacy is any harder to aquire for a native speaker. True, high levels of literacy are difficult – there is a big difference between knowing 4,000 characters and knowing the 8,000 that an elite scholar might know. But it is arguable that the same is true of English…
Even if Chinese statistics are inflated, I think it is arguable that China has significantly higher literacy rates than India, which uses alphabetic scripts.
…until the characters are replaced with a simpler system, whether romanization or some sort of syllabary.
How would dialects be handled by a phonetic system/syllabary? Various provinces of China have pronunciation systems that are completely different from Mandarin and each other. Cantonese is an example so exaggerated that it has been made a whole other language.
In my experience, many of the people who live in a certain area speak the dialect of that area exclusively, and the Mandarin phonetic system does not correspond to their own pronunciation of the word.
For example, my name is “An Qi” (“angel”) in Mandarin, but that is pronounced “Oojy” (approximation of the sound) in Shanghainese, which also has completely different tonal sounds. And “oojy” sounds like “[something] chicken” in Mandarin. The spoken dialects are all so different that someone who comes from one province or another would not be able to understand someone who comes from, say, Beijing and speaks Mandarin.
Peasants in particular do not usually live near Beijing or other large cities where Mandarin would be spoken, preferring instead to speak their own vastly different dialect. For that reason, this proposal would benefit only educated Mandarin-speaking people (including city workers) who are already literate, which defeats the purpose.
How can someone who has to work most of each day just for food and shelter, whether peasant or city dweller, maintain such an intricate system? Even if they truly learned the mandatory few thousand characters in school, they quickly forget them for want of repetition.
I’m not sure where you’re getting the statistics for the average workday in China, but it’s an 8-hour workday for five days a week, same as it is in the U.S. And it’s not really for food and shelter. Um.
And I don’t understand how they would forget the language. My relatives do read and write once they’ve learned the words; forgetting would be difficult.
Traditionally they were the province of a tiny elite that had the leisure to spend years on such a demanding course of study.
Isn’t this true of most written languages, though? If I were a peasant during the Renaissance, I would not be literate despite the fact that the written language is easy to learn.
Finally, most Chinese people would tell you that Chinese is not difficult to learn; the visual cues and symbols make sense to most native speakers who do in fact go to school. Just ask anyone; if they haven’t gone to school, they haven’t learned how to write, same as it is everywhere.
Sigh. I’m not going to spend any more time and effort arguing this here, since I don’t have any references for you. If it’s not intuitively obvious to you that literacy involving thousands of characters, many of them extremely complex, is a great deal harder to achieve than literacy with an alphabet that can be learned in a day, nothing I say will convince you. A couple of points:
it’s an 8-hour workday for five days a week, same as it is in the U.S.
Uh huh, sure. China is just like the US, and the average person has the same workday. And they’re all just as literate as we. Absolutely. And the Russians under Stalin were free as little birds, because the Soviet Constitution guaranteed more freedoms than any in the world. Official truth is the only truth.
China has significantly higher literacy rates than India
Very likely. And this proves what? Surely you’re not implying that the reason Indians can’t read is because of devanagari?
Oh, and the dialects/languages: of course this is a serious problem, and a better (practical) reason for maintaining the present system than cultural history. It would be impossible to maintain the fiction that the separate Chinese languages are merely “dialects” if they were written phonetically, and the open linguistic divisions might have political consequences. (Of course, this is bad only if you’re involved with a government grimly determined to rule a billion people in enforced unity from Peking/Beijing.) In any case, romanization is only a thought experiment; it’s highly unlikely to come about unless there’s some sort of major political upheaval (comparable to the end of the Ottoman Empire, which allowed the Turks to romanize their own writing system). So fear not, you traditionalists!
China is just like the US, and the average person has the same workday.
I’m Chinese. My parents are Chinese. My father’s side of the family farms and my mother’s side are city workers. Neither side is wealthy.
I can call up the relatives on my father’s side and ask, if you like. I emailed my cousin (mother’s side) when I began composing my second reply and asked her how long her workday was and how long her parents’ workdays were and she told me eight hours. If she’s wrong, she’s wrong. *shrug*
Also, a law was passed some time ago that restricted weekly hours from 44 to 40 hours and gave workers two rest days a week.
I’m perfectly willing to believe those members of your family work eight-hour days. Lots of Chinese do. My understanding is that there are a great many more poor Chinese who have to work long days and don’t have much leisure than there are poor Americans, though of course that’s changing rapidly, and I don’t vouch for my understanding as the Truth. At any rate, my point is that government literacy stats are not to be trusted. You may disagree.
Sigh… I wasn’t going to weigh in on this one because I hate this debate…. LH, you may feel it’s “intuitively obvious” that literacy with characters is more difficult than literacy with an alphabet, but there is a gap in that reasoning. It does seem obvious to me that memorizing 2000 characters is more difficult than memorizing 26 letters, but knowing the orthographic system doesn’t make you literate, though it may contribute to the condition that is implied by the term “functionally illiterate” (i.e. having some ability to read letters and even words but not the ability to make sense of written language).
Here’s one thought: as far as I know there is some evidence to suggest that people who are literate in English read words and groups of words, rather than individual letters — which implies to me that literate English readers treat words somewhat like Chinese characters, that is, as indivisible chunks of meaning rather than as assemblages of letters. The letters may give us more access to the pronunciation of unfamiliar words, and they sure make dictionaries easier to use, but they don’t necessarily simplify the problem of literacy.
On top of this, there is a component to literacy that has to do with understanding genres of text (the traffic sign, the restaurant menu, the instruction manual). In other words, you have to not only be able to recognize the letters, *and* the words, you also have to know what words can be used for. I might be more convinced by an argument that suggested that literacy in China was hampered by the fact that written Chinese is still not a true vernacular, as any non-native learner of the language soon discovers: you still can’t write just what you would say, though it’s closer than it used to be. Thus the genres of text are further removed from the genres of spoken language in Chinese than they are in English.
As for your debate with Tao Zi, may I suggest that you two are talking about different groups of mainland Chinese citizens: TZ seems to be making reference to the class of Chinese citizens in small cities and the better-off parts of the countryside (mainly where there is a lot of water and/or overseas relatives, typically the south and southeast), who are doing much better both in terms of standard of living and levels of literacy (and other education) than they were twenty years ago; while you, LH, seem to be referring to the residents of the most poverty-stricken parts of the mainland (mainly the water-poor north and northwest, and parts of Sichuan and Tibet).
The fact that levels of literacy in large cities and in the areas TZ is talking about are way up seems to me to have more to do with money than with the inherent difficulty of the language: money, and the tendency of educated people to move to the cities and more affluent areas of the country because they can; and also the way in which the children of literate people everywhere have it easier in becoming literate themselves. Education breeds education, as we’re finding in the US; in China this is true even among the children of the peasantry — at least the peasants in Fujian and Guangzhou who go to spiffy new schools endowed by overseas relatives. Meanwhile, economic pressures mean that members of the “floating population” move to big cities to work in the shadow economy; but because they don’t have legal permission to live in the cities, their children don’t have access to public education at all.
It’s worth knowing that public education in mainland China (a) is compulsory but NOT free — parents must pay tuition for their children, plus buy books and supplies; and (b) is only guaranteed through primary school — not everybody gets to go to junior high, still fewer get to go to high school, and let’s not talk about college. If mainland China could or would devote enough resources to institute a public education system in which everyone could go to high school, perhaps literacy rates would even out between poorer and richer parts of the country. At that point perhaps we could start talking about the implications of a character-based system for literacy in Chinese. Before that it seems that we can only make conjectures based on what is “intuitively obvious” to us, and that’s a point on which we’re never going to agree.
You are, however, right to suspect government stats. YMMV.
And this proves what? Surely you’re not implying that the reason Indians can’t read is because of devanagari?
Of course not. The point is a reverse argument – simply showing that script doesn’t matter as much as providing basic standard education for improving literacy. As someone who spent a year living in a rural elementary school in Taiwan, amongst the Aborigine population which has some of the highest school drop-out rates in the country, I didn’t see characters as an impedement to learning. Nor did anyone I met and worked with. Broken families, absent parents, broken homes, alcaholism, etc. were much more on people’s minds than was orthographic reform – and people laughed at me for being so facinated with the orthographic debates. I think the difficulty of learning Chinese characters is only “intuitvely obvious” to someone who learns Chinese as a second language.
I also agree with xiaolongnu that how we read has to do with much more than spelling. But I think it is important to distinguish writing and reading. I myself can not write many characters – I decided it wasn’t worth my time. I use pinyin input and I just need to recognize if the comptuer is telling me it is the correct character or not. So I can recognize thousands of characters that I can’t write. This is obviously not the case in English, where even though I make lots of spelling errors, I can usually get pretty close even with words I only write on rare occasions. So learning to write Chinese is difficult – but not insurmountably so. As far as I can tell, most people learn to write Chinese by simple repetition. It is learned in the “hand” more than in the “head.”
I also think it is a good point that literacy vs. illiteracy is not a black and white issue – there is a tremendous grey area inbetween. How many American’s can read a text in literary criticism? These days I’m not even sure I can with some of the stuff that’s published. But learning to read is about a lot more than learning to spell, and while it may be harder to learn to read with Chinese – I think the difference is not that great for a native speaker child who receives a decent education.
But the main point is simply that given the proper learning environment, a child will learn to read whether it is devanagari or Hanzi. The solution to illiteracy is providing such an environment – not changing the writing system.
Interestingly, one of the reasons Taiwan is moving from Zhuyin Fu Hao to a Romanization system is that Chinese themselves initially learn Chinese with such a phonetic system, and they feel it will be easier for the children to learn English if they don’t need to learn a third writing system. But here to, I think it is important to distinguish between learning literacy in your mother tongue and learning a second language. One of the main arguments for bilingual education is that it is easier to learn literacy in your native language. I think that is true even if your native language has a very difficult writing system.
I’m tired and rambling … so I’ll stop here.
Excellent points all around. I also agree with everything xiaolongnu said, and am glad she overcame her reluctance to weigh in. Obviously to some extent I was being provocative with my simple formulation, but I certainly wasn’t claiming that characters are an insurmountable difficulty or that it’s impossible to achieve mass literacy with them — just a lot harder. It does seem obvious that it takes longer to learn a system with thousands of characters, and the time spent doing that could be spent learning physics or tennis, but life is full of unnecessary burdens and we mostly get by anyway. And the whole discussion is purely theoretical, since nobody’s in a hurry to give up characters! But I deeply appreciate everyone’s input.
“There’s literacy and there’s literacy”. Chinese literacy is sort of graded, and basic functional literacy isn’t that hard. But I think that Chinese, more than other languages, reminds you how much more you have to learn. For example, newspaper-reading is a separate course. (I worked at it for awhile until I realized I had spent ten minutes figuring out something like “Louis Robert, Belgian Undersecretary for Transportation, and Roger Schultz, the Swiss chairman of the International Trade Commission” — and it was in an AP story translated into Chinese. Similiarly, my Chinese teacher, a college-graduate Chinese major, couldn’t read the Hung Lou Meng (XVI C.?) with much confidence.
However, it’s the same for English. The average, bright, literate HS graduate can’t read anything of any difficulty unless he’s spent time studying that specific topic. I remember as a college freshman picking up a fairly basic book on economic history (which I was actually interested in) and being able to understand almost nothing.
Here are some links:
A list of books on orthography. (With sample chapters.)
Samples of another book.
A page on Chinese writing reforms.
Contents from a book on Chinese children learning to read.
If anyone finds any more interesting links, please post them here!
To go on, traditionally marking class distinctions and enabling the display of cultural refinement were considered a very positive aspect of the system, but less so now. The same thing happens, more weakly, in English of course. And the inefficient semi-phonetic English spelling system moves it in the direction of the Chinese system — whole-word learning is often the quickest. (It’s a tough trough to get through, though).
I’ve spent twice as much time on Chinese as on Tibetan, but I know twice as much Tibetan as Chinese. The only difference I can see in the learning situation is characters vs alphabetic script — the languages are equally foreign to me, as a native English speaker. The statistics I’ve seen suggest that Chinese children spend one or two more school years to acquire the same level of literacy as students learning alphabetic languages. (“The Chinese Language,” I think the book was? by a romanization partisan, I forget his name, late of Yale)
It’s a complex question. I think technology is going to decide it, not policy makers. A Chinese woman I work with tells me that she writes email home in English, not Chinese, because she finds typing Chinese so laborious — her English-literate brother then translates the email to her parents. It takes a pretty large obstacle to push someone into corresponding with their parents in a second language, I think. My guess is that Chinese characters are doomed to take the place of, say, Chaucerian English in the English-speaking world — an elective branch of study for people who delight in the difficult and obscure. A shame, in many ways. I agree that they’re one of the treasures of civilization. I love them, and if I had another lifetime to spend I’d gladly spend it acquiring 8000 characters and reading Chinese poetry.
Kerim:
Addendum – there is one person who came up with an ingeious idea for reforming Chinese writing which I actually like. It would entail the use of radicals and romanization together. This seems like an ideal system – the radicals would give you a sense of the meaning (fire/water/beast/etc.) and the phonetics would give you a sense of the pronounciation….
Yeah, the trouble with a plain old romanization system, or even a spelling system that uses multiple separate ways of writing the same syllable to pack in more information (a la English two, to, too), is that it’s not very resistant to the introduction of noise. Mainly cuz Chinese is so syllable-poor. So you need to add something in there, like radicals, to ensure that meaning is retained even if someone makes a typo. (I have much longer thoughts on this matter, but I didn’t want to hijack the comments section with an essay, so I posted it here).
Language hat:
Oh, and the dialects/languages: of course this is a serious problem, and a better (practical) reason for maintaining the present system than cultural history. It would be impossible to maintain the fiction that the separate Chinese languages are merely “dialects” if they were written phonetically, and the open linguistic divisions might have political consequences. (Of course, this is bad only if you’re involved with a government grimly determined to rule a billion people in enforced unity from Peking/Beijing.)
Hi, my first time here, I added a link to you. Anyway, regarding this argument: promoting speaking in dialects and devaluing Mandarin would be bad for more than just the government. It would add huge transaction costs across the entire economy because of the increased need for translators. Not exactly what a poor country needs. Also I think you’re overexaggerating the extent to which these linguistic divisions might result in political consequences. Certainly those regions inhabited by Mongolians or Uighurs aren’t too happy about being stuck in the Qing Empire as revived by the Communists, but the last thing a bunch of poor peasants in Sichuan aspiring to become workers in Beijing want is for the central government to start blowing them off and treating them like a separate country just because they speak a different dialect.
“the radicals would give you a sense of the meaning (fire/water/beast/etc.) and the phonetics would give you a sense of the pronounciation” — I thought that’s how most characters already worked – combination of a radical and a phonetic hint.
Baloney: The more characters I learn, the more their “etymologies” seem like mnemonic stories à la ars memorativa (well documented by Yates) and less like a true semantic component. There’s no denying there’s a phonological component, just that it’s based on pronunciations that are two to four thousand years old.