ILLITERACY IN CHINA.

I would urge everyone to read Victor Mair’s guest post at Language Log. Its conclusion seems to me indisputable:

The people of China — now, as they have been for the past three millennia — are constantly challenged by an enormously complicated script suitable only for an elite consisting of a tiny proportion of the population…
If only China would adopt a policy of true digraphia (PINYIN plus HANZI) and actively promote it, the problems of illiteracy would vanish within a decade or two.

I’ve said that before myself, but people seem unwilling to accept it; the romance and storied history of the characters blinds people to the obvious.

Comments

  1. Someone has to actually demonstrate the causal relationship between a non-alphabetical writing system and illiteracy. It can’t be done, of course.
    While you’re at it, you can explain to us why the Japanese have the world’s highest literacy rate. 🙂

  2. “With alphabetic literacy, one can forget how to spell a word properly but still get one’s idea across by misspelling it. If one forgets a crucial character, like the TI4 of DA3 PEN1TI4 打喷嚏 (“sneeze”), which very few Chinese know how to write, you’re stuck.”
    Nothing is preventing people from using a homophonous character, as Mair surely knows, or otherwise fudging it. People do it all the time. It’s not a situation where it’s the correct character or nothing.
    And while I’m at it, pinyin is not the only available “digraphia” option. How about bringing back fanqie? or using zhuyin fuhao like Japanese furigana?

  3. Wouldn’t digraphia basically mean the end of the hanzi, though? Given the choice between two systems, one a straightforward system with 20-some characters and the other an arcane labyrinth with thousands of characters, who would bother to learn the latter? And if you tried to keep hanzi alive by requiring it for certain purposes, then the majority who were illiterate in hanzi would be unable to function in those ways in society.
    That’s a pretty strong argument against digraphia, at least if you consider hanzi an indispensible part of Chinese culture.

  4. Wouldn’t digraphia basically mean the end of the hanzi, though?
    In popular use, yes, though they would continue to be studied by scholars. That’s why there’s so much resistance to any proposed change. Some of us think the history and romance of the hanzi are a poor tradeoff for illiteracy.
    if you consider hanzi an indispensible part of Chinese culture
    Which of course they’re not, any more than the Perso-Arabic alphabet was an indispensible part of Turkish culture, though before Atatürk’s reform many people would have said it was. Cultures are very resilient. And do all those illiterate people not count as part of “Chinese culture”?

  5. The high Japanese literacy rate may well be a myth. The writer quotes deFrancis, who along with Mair counts Mark Swofford as a fan, so perhaps such ideas as presented are all ultimately derived from deFrancis.

  6. Knowing next to nothing about chinese, how many homophonic characters are there? And how easy would it be to turn it into a syllabic writing system? To an outsider that would seem a reasonable compromise, and aparently in practise people already use it that way…

  7. Supplementing the characters with zhuyin fuhao would also allow mixing, with phonetic characters inserted where the hanzi was forgotten.
    I used to know a Japanese/American science PhD, educated both in English and Japanese, whose Japanese father rebuked him whenever he substituted phonetic writing for kanji. It was purely a class-dialect issue. Basically his science studies had kept him too busy to keep refreshing his Japanese.
    “Japanese / American”: Born in a relocation camp, taken to Japan and educated there through high school, bilingual from birth, college and grad school in the US. He studied WWII in a Japanese HS, an American University, and the US Army War College. Very interesting guy.

  8. I believe that there are about 400 possible syllables, 1600 counting tones, but not all of them are actually ever used. I think that with about 350 syllables plus tone marks you could have a syllabary, but an alphabetic system seems like a better idea.

  9. Time to plug Mark Rosenfelder’s wonderful essay on yingzi, or English written with Chinese-style characters.

  10. Culture notwithstanding, I always thought that the Chinese logographic system served to bind the country together; they’ve all diverged phonetically, but have a close enough word-stock that they can all use the same writing system if it’s a logographic one. No?

  11. I find Mair’s conclusions disputable because he doesn’t provide any real facts – just assertions that “sound right.” Where does he come up with the figure that only 20-25% of the population can recognize 3000 characters? That strikes me as complete crap – it certainly is not true in Taiwan. Is Taiwanese literacy a myth or not? It’s above 95%. Maybe the answer is to dump the illogical simplified character set.
    Increased literacy of the general population is generally a good thing, but let’s not pretend that there won’t also be negative consequences from such a radical change, some probably unpredictable. I think there is no question that Turkish culture, and literature in particular, has suffered since the break. There has certainly been much economic growth and positive political change in Turkey in the last 80 years, which is probably, but not necessarily, related to the new alphabet. I doubt anyone has actually tried to prove a relationship, it is just something we all “know.”
    It’s also naive to ignore the nationalistic element of characters – hanzi are one of the glues that hold the various “Chinese” peoples together. The PRC government is very worried that a move to phonetic alphabets will encourage more development of regional languages and weakening of the Han identity. I think for that reason alone any discussion of China moving away from hanzi is a pipe dream.

  12. Another common argument. As with the beauty/tradition, yes, it’s a nice feature, but not an acceptable tradeoff for illiteracy. Look at it this way: would you be willing to have all European languages written in an ideographic system with thousands of characters in order to be able to communicate in writing without having to learn another language?

  13. Er, that was a response to Ran, but I see vanya makes the same argument. And yes, it may be a pipe dream, but I wish people would acknowledge that it’s desirable but politically impossible rather than trying to show that the present system is worth preserving for its own sake. (And if more people would acknowledge that, maybe it would get less impossible, especially as the common people start getting more of a voice on the mainland.)

  14. would you be willing to have all European languages written in an ideographic system with thousands of characters in order to be able to communicate in writing without having to learn another language? I wasn’t using the regional identity issue to argue for the utility of hanzi, simply stating a political fact. If the the Chinese elite believes hanzi are important to keep China from disintegrating than the elite will not budge on this issue no matter how good the arguments of the Mair’s of the world are.
    My problem with the illiteracy argument is that, in my experience, Chinese characters aren’t that hard to learn if you’re talking about reading. I learned to recognize well over a thousand kanji in the space of a year in Japan – and kanji are worse than hanzi because of the many different readings. Learning to write hanzi, I grant, is difficult, but even that, with computer software, is not the obstacle it once was. So yes hanzi are cumbersome, and I wouldn’t start a system from scratch using them, but I think Mair really exaggerates the day-to-day burden they create in people’s lives. Certainly Japan and Taiwan have not suffered materially from using characters, whereas many countries using phonetic alphabets cannot boast anywhere near the same economic or educational progress. I think writing reform ranks pretty close to the bottom of “what needs to be done” in China.

  15. Of course, if you’re a native speaker of something other than Mandarin, alphabetic writing would have obvious advantages, and hanzi obvious disadvantages. Much of the resistance to digraphia would come from Mandarin chauvinists hostile to the idea of Wu, Cantonese, Min, and other “dialects” flourishing independently and developing literatures of their own.

  16. In the comments a I see little mention of the fact that the article mainly deals with imparting (or rather the failiure to impart) a working knowledge of hanzi on sections of the population with scarce access to schools. And employed in occupations not too concerned with reading and writing.The conclusion being that this is very hard indeed.
    To bring into this discussion assertions that Japan, a country with 9 years of compulsory education (not a right, a must), has a litteracy rate of 99% seems exceedingly irrelevant.
    Now, my experience is limited to Japanese but: I’ve seen enough Japanese born, raised and educated professors of the Japanese language working in Japanese universities ask a dictionary carrying student “Did I get this kanji right?” to be convinced that… I don’t exactly what I’m convinced of but, Japanese writing is hard as hell and once you get past the magic barrier of a 1000 characters it kinda becomes a blur.
    If the Koreas could free themselves of hanzi/kanji so can Japan and China! (You haven’t forgotten that they did that have you?)

  17. I forgot Vietnam!
    The brave Vietnamese, ignoring Chinese taunts about cultural flimsiness, also got themselves a more sensible (if less beautiful) writing system.
    It would seem only the Koreans managed both sense AND beauty.

  18. michael farris says

    Looking for something else (vietnamese calligraphy using quốc ngữ ) I stumbled across this, which may be of interest (research comparing the reading skills of taiwanese character learning children and vietnamese alphabet learning children).
    http://www.twl.ncku.edu.tw/~uibun/chuliau/lunsoat/english/phd/index.htm
    food for thought, maybe

  19. ntartifex says

    Long time reader, first time commenter.
    I have patiently and humbly spent the last year studying Mandarin Chinese with zero experience. Until yesterday, I had spent the semester envying my Chinese classmates who grew up with ambient phoneme exposure/speaking parents/Classical and seemed so much more at ease during speaking.
    As it turned out, two of my native classmates had to ask me to decipher an unknown word in the lessons. I only managed to figure it out because I had seen and remembered the characters in other compound words. I think that’s the problem — deconstructing vocabulary is much easier than calling it out of your head. I suppose you could do the “insert homonym character” trick (incorrect spelling) but that would change the message.

  20. I really would like to understand better what the opponents of hanzi think reform will achieve, big picture. Will China grow even faster economically than it already is growing? Will China get even more foreign investment? Will China make even faster gains in computer, automotive, and aerospace technology? Would increased literacy encourage faster democratization? – that hasn’t happened in Vietnam. Would Chinese learn English more easily? Doesn’t seem to be an issue so far. Would Han nationalism become less aggressive? Possibly, but the benefits would be offset by increasingly violent regional nationalism. Really, what would be better about life in China through the reform Mair proposes? I know the efficiency argument – “People spend too much time studying characters” – yes, Chinese writing seems inefficient at first glance, but most people spend countless hours learning useless trivia, memorizing bad poetry, imbibing sports statistics, etc. Why is memorizing characters so harmful? What fantastic things would these students otherwise be doing with their time? Somehow the millions of Chinese who are quite good at math, music, cooking, film-making, dance, engineering, etc. don’t seem to have been handicapped by having to learn characters. “well, more people would be literate!” OK,sounds good in isolation, but then what? Will their lives really improve if they’re working in jobs that don’t require literacy? It may be true that a modern industrial state can actually function just fine with 30% illiteracy, particularly in the video age. If smart ambitious people are being denied opportunities in China because they have no chance to learn to read, that would demand a remedy. But is there any evidence this is a widespread problem? I’m shocked people are willing to casually throw out 3000 years of tradition and knowledge when they can’t even specify what great tangible improvements their reform is supposed to bring other than “efficiency”. I say pshaw.

  21. David Marjanović says

    Culture notwithstanding, I always thought that the Chinese logographic system served to bind the country together; they’ve all diverged phonetically, but have a close enough word-stock that they can all use the same writing system if it’s a logographic one. No?

    No. It works only for Mandarin. In the other Sinitic languages, many of the most common words simply don’t have a character. For Cantonese many characters were made up to fill the gap, but not for the others. As an example, here is a sample of Taiwanese written in characters to the greatest extent possible; scroll down to compare that to its Mandarin translation.

    Maybe the answer is to dump the illogical simplified character set.

    Hah. That would make reading a bit easier, granted. It would make writing a lot more difficult, however. Compare the simplified and the traditional characters for “how much”.
    And what exactly do you mean by “illogical”?

    in order to be able to communicate in writing without having to learn another language?

    The “without” part is not even true. Speakers of any Sinitic language other than (to a large extent) Cantonese are currently forced to learn Mandarin in order to learn to write — not the whole language, but much of it.

    Would Han nationalism become less aggressive? Possibly, but the benefits would be offset by increasingly violent regional nationalism.

    Really?

    Somehow the millions of Chinese who are quite good at math, music, cooking, film-making, dance, engineering, etc. don’t seem to have been handicapped by having to learn characters.

    But, you see, it could be a billion.

    Will their lives really improve if they’re working in jobs that don’t require literacy? It may be true that a modern industrial state can actually function just fine with 30% illiteracy, particularly in the video age.

    You don’t believe that yourself. Come on. Please.

    people are willing to casually throw out 3000 years of tradition

    That’s not going to happen anyway. What about the Japanese model (with pinyin, zhuyin or whatever instead of kana)?

  22. Etienne says

    A seemingly silly question, if I may: is written Chinese really that much harder to learn than, say, English spelling? Professor Mair’s posting seems to repeatedly stress the flaws of Chinese characters without ever proving, or indeed attempting to prove, that Chinese characters are indeed of an order of complexity such as to be in and of themselves an obstacle to literacy.
    Having little knowledge of China or Chinese, I will simply add that similar such illiteracy rates can be found in much of (especially rural) Latin America, despite the simplicity of Spanish and Portuguese spelling. Would Professor Mair propose that the solution to Latin American illiteracy is to spell Spanish and Portuguese with the script/spelling of a highly literate culture (Japanese? Hebrew?)

  23. David Marjanović says

    A seemingly silly question, if I may: is written Chinese really that much harder to learn than, say, English spelling?

    English spelling is an unfair comparison, but even then, yes, it is easier. I’ve tried both :o)

    without ever proving, or indeed attempting to prove, that Chinese characters are indeed of an order of complexity such as to be in and of themselves an obstacle to literacy.

    Because it’s self-evident. Just try it. 🙂 Or take the fact that it takes native speakers in good schools years to get to the level necessary for reading, well, just about any text of any length.

    I will simply add that similar such illiteracy rates can be found in much of (especially rural) Latin America, despite the simplicity of Spanish and Portuguese spelling.

    That’s because most of those people have never even seen a school from the outside. That’s not so common in China.

    Would Professor Mair propose that the solution to Latin American illiteracy is to spell Spanish and Portuguese with the script/spelling of a highly literate culture (Japanese? Hebrew?)

    Nope, these are all more difficult than Spanish/Portuguese spelling. OK, hiragana is easier, but doesn’t work for a language with more than 50-odd syllables.

  24. The PRC government is very worried that a move to phonetic alphabets will encourage more development of regional languages…
    I would argue that linguistic diversity is a good thing, even if it doesn’t directly benefit the PRC government.

  25. Noetica says

    A fascinating discussion.
    Here are my ruminations, as one who has spent a little time in China, and much time in dialogue with Chinese students in Australia, and also in China (by email and telephone, these days) as they struggle with English and I struggle intermittently with Chinese.
    Alphabetic scripts have always been superior for representing the sound of a language. Sure, there is representation of sounds in some hanzi (not very reliable representation). And sure, hanzi have a romance and a beauty beyond the dreams of our debased and worn-out Phoenician coinage.
    Alphabetic scripts also have the advantage that they can be learned and applied rapidly by most people possessing competence in the spoken language. Sure, there are mechanisms hidden below the surface that we still don’t understand well; for many people these mechanisms fail, and they never do learn the code fully or efficiently. By contrast, perhaps hanzi do not call on those fallible mechanisms, and might be learnable by those who would be dyslexic with an alphabet. But then, other hidden mechanisms might be in play in acquisition of codes like hanzi, and those mechanisms might fail for other learners. These are empirical matters, and we’d just be guessing if we didn’t engage with empirical research. Still, we do know that people’s abilities with recognition of shapes vary widely, and some deficits are very specific (think of prosopagnosia). It would be surprising if this variance did not yield wide variance in acquisition and use of hanzi.
    If there were only one Chinese language to deal with, and it had little dialectal variation, an alphabetic code would, I think, be vastly better than hanzi. It could represent sound simply and transparently, and that would be enough for its principles to be grasped easily by almost all speakers of the language. It would also be better than a syllabary, because it would be extensible beyond the very limited range of possible syllables in the main Chinese language. (Pinyin itself is not good for that purpose; but at least users of it get the basic alphabetic idea.)
    Generally, China implements a policy of imposing one language (putonghua) for “official” purposes. To the extent that this draconian policy will be successful, an alphabetic code will clearly be the way to go. Until it is successful, millions of Chinese who use other languages would be disadvantaged. Acquiring literacy in putonghua, they might lose literacy in their own languages. Digraphia may be a luxury not available to all.
    These problems arising from diversity in the language group bear some resemblance to our English-language problems. I do not favour spelling reform to match some spurious standard pronunciation. Which standard? (Educated South-East Australian would be the obvious choice for world communications, yes?) And since the linguistic situation in China remains complex despite the central dictatorship’s best efforts, Chinese may be right to resist abandoning a system that somehow clunks along, with some success in maintaining several cultures and enabling, um… a modest achievement in commerce.
    Then of course there is the matter of encoding for “modern” communications. If the problems discovered with telegraphy and the typewriter did not clinch things more than a century ago, the internet and the PC surely should now. I have a hanzi-inputting macro installed in Word, and I have watched highly literate Chinese use such a system. For me it’s glacial; but it’s slow for them too.

  26. David,
    But, you see, it could be a billion.
    Well, that’s the point, isn’t it? We wouldn’t want to have around more than a billion people who are actually good at something else than producing cheap stuff, now would we?
    vanya,
    interesting point about efficiency. If I understand Mair correctly, to him, “efficiency” is means to an end, not the final goal. It’s all about ease of access to … to … to all the things those of us who learned to read at the age of six take for granted.

  27. I thought Chinese had the problem of having a ridiculous number of homophones that made converting to an alphabetic system quite difficult.
    How much of an obstacle is that for the language’s alphabetisation?

  28. Not much of one. As I always point out to people who raise this objection, if the Chinese can understand a sentence when they hear it, they can understand the same sentence when they read a phonetic rendition of it. The whole homophone issue tends to be overblown — it ignores the fact that many words in modern Chinese are multisyllabic, for one thing.

  29. David,
    But, you see, it could be a billion.

    No, it couldn’t. Not under any economic system currently operating on this planet. No more than every literate American, Indonesian or Mexican is somehow magically granted the right to live a life of care-free creativity and prosperity. Reform of peripheral issues like hanzi is not going to lead to any real meaningful social change in China. The responses I see are leading me to revise my thesis – attacking hanzi is not just efficency worship, it is also often a proxy for the naive progressive idealism that I thought went out of fashion in the 1950s.

  30. Here are some additional readings that may be of interest and use in this discussion. (I hope all these links don’t trip the spam filter.)
    Comparing Chinese Characters and a Chinese Spelling Script — an evening conversation on the reform of Chinese characters, by Lü Shuxiang, editor in chief of the Xiandai Hanyu cidian (现代汉语词典)
    Also available in the original Mandarin: 漢字和拼音字的 比較 —-漢字改革一 夕談 / 汉字和拼音字的 比较 —-汉字改革一 夕谈.
    The Prospects for Chinese Writing Reform, by John DeFrancis
    Hanzi Bu Tebie Biaoyi, by Zhang Liqing (PDF — written in Hanyu Pinyin)
    An Outsider’s Chats about Written Language, by Lu Xun
    Of course, The Ideographic Myth, by John DeFrancis
    And on the subject of literacy in Japan, see Unger: The Price of Tradition and Literacy and Script Reform in Occupation Japan.

  31. “Reform of peripheral issues like hanzi is not going to lead to any real meaningful social change in China. ”
    (loosely paraphrased from Mair’s post)
    Literate rural municipality official claims “Everybody here can now read”. Intra- and interregional (and international) media report this fact. Unaffiliated non-local literate person happens to takes an interest, goes there and ask local supposedly literate person “Can you read?”. The answer is “No”.
    Now consider a ficitional scenario.
    Rural municipality official claims “Everybody here is well fed”. Intra- and interregional (and international) media report this fact. Local starving person read said media and sends in a counter-claim stating that aforementioned “fact” is a lie. Intra and interregional debate and fact-finding ensues.
    Literacy equals agency in a modern society. It in it self IS a meaningful social change. The choice of writing system is by no means peripheral, if it makes a difference in MILLIONS of people being able to read and write.
    Labeling hanzi a peripheral issue does in fact sound quite like the old arguments about the struggle for gender equality being unnecessary since everbody would be equal once socialism was realized. A line of argument I thought had gone away with the 50s as well 🙂

  32. vanya, you’re a smart guy, so I’m having a hard time understanding how you can seriously argue that illiteracy is some sort of minor handicap, like a runny nose, that needn’t be taken seriously. You can’t possibly believe that a significantly improved literacy rate wouldn’t make a significant difference in people’s lives.

  33. LH
    You can’t possibly believe that a significantly improved literacy rate wouldn’t make a significant difference in people’s lives.
    Blaming the writing system is blaming a symptom, not a cause. The examples of Taiwan or Hong Kong show that is perfectly possible with effort to achieve functional literacy for 95% of the population, and that well more than a “tiny elite” can achieve real literacy. Again, it is a question of efficiency – China unquestionably needs to devote more resources to literacy education than other countries do, and on the margins the literacy rate will never be quite as high as in Vietnam or Korea, I grant that. But most of the illiteracy issue in China could be solved by funding schools properly rather than radically overhauling the writing system. Many of you are willing to destroy a keystone of Chinese culture to achieve that marginal improvement and not have to spend those extra yuan. I’m not saying you’re necessarily wrong, I just don’t think most people who take this position have really thought through the consequences of radically changing Chinese society (yet again), including Mair.

  34. LH–It might be true that even a majority of people can’t correctly write (by hand) the ti4 of da3 pen1ti4, but I am sure a very large percentage of those same people would recognize it, especially when they see “da3 pen1” in front of it. They’d also be able to pick it out of the line-up when entering text on a computer.

  35. I don’t claim to understand everything in this discussion — I never went beyond High School English studies. However, I will throw in my 2 cents as a native English speaker, Chinese student, and TESOL teacher.
    1) I’ve said many times that to REALLY learn a language, you must live in it. Language is communication, and inseparable (in its fullest sense) from its creative culture. Hence, I can understand the Taiwanese sadness that their mainland counterparts have switched to a simplified (and perceived as uglier, and less meaningful) character set.
    2) English (alphabetic) and Chinese (what? non-alphabetic) are learnt in 2 very different ways. While English is not 100% phonetic in its spelling, it doesn’t require a large amount of memory (c.f. computer hard disk space). It is largely of putting things together and processing it, e.g. prefixes, suffixes, grammar (c.f. computer RAM and CPU). Chinese, however, takes up much more initial memory, and doesn’t involve so much conscious processing. It’s more simply a matter of image recognition, in terms of words. I find this also correlates (which is cause, which is effect, I don’t know) to culture and methods of teaching.
    In Chinese culture, you traditionally just do what your superiors tell you. You don’t try to process, analyse, synthesize, or much less, question what you are told. In the West, our education is very much about those exact same skills. The upside of the Chinese method is that their skills for simply remembering stuff are much greater than Westerners like myself who aren’t that great at simple memory, but could analyse and process to the most minute details. (And yes, I’ve stereotyped Chinese and Western cultures, before anyone gets too fired up about it.)
    3) Anytime people consciously try to shape and mould a language, they usually screw up. I’d much prefer that either, and in this order, A) people learn 2 languages (in this case, English and Traditional Chinese), or B) just let language and culture pan out how it will, and we’ll just see what we get.
    4) A certain Indian friend of mine has boasted of the great literacy rates of India pre-British colonial days. On his behalf, I propose that we all learn whatever that/those languages was/were, as it/they was/were quite OBVIOUSLY superior to our current situation.

  36. Oh…and if you want to know about Chinese writing and haven’t read this, then you should.
    http://www.pinyin.info/readings/texts/ideographic_myth.html

  37. michael farris says

    A few random observations (partly mine and partly from several uncredited sources!).
    – Theoretical and linguistically sound judgements are totally irrelevant in actually effecting language change or language policy.
    – If Chinese characters really are so crucial to Chinese identity then increased use of pinyin won’t threaten them.
    – The single biggest factor in literacy in minority languages is the existence of authors. Materials by outsiders are of limited use and interest, but if someone writes a book in their language, they’ll pester people into reading it. Interest Chinese people in the literary possibilities of pinyin.
    – People that want to promote pinyin (on its own, not as a ‘replacement’) should use it whenever they can. Write stories, novels, whatever, make web pages in it. At all times, pinyin should be promoted as something worthwhile on its own, irregardless of its wider fortunes.
    – Find niches for pinyin. From everything I’ve heard, Chinese characters do not lend themselves gracefully to certain functions that are often crucial in the modern world, especially those that have to do with sorting. Convince a Chinese languge academic publication that sorting authors in pinyin (with chinese characters following) is quicker and easier than other ways (if this is true), telephone listings and the like might also be something to look at.

  38. “if the Chinese can understand a sentence when they hear it, they can understand the same sentence when they read a phonetic rendition of it.”
    There’s your answer, then. No one can understand most of Chinese literature other than novels from before the 20th century just by listening to it as if it were everyday speech. Transcibing from hanzi to pinyin simply makes this literature unintelligble. It’s a poor kind of literacy that puts the bulk of a culture’s literature off limits.

  39. Michael–I think sorting names, etc., by stroke number works just as well for native Chinese speakers (though personally I have to do the drawing-on-the-hand thing to find what I want). Also, at least in Taiwan, many things (books in bookstores, etc.) are already sorted by zhuyin fuhao order. Pinyin is not the only option for “alphabetical” sorting.

  40. michael farris says

    “Transcibing from hanzi to pinyin simply makes this literature unintelligble. It’s a poor kind of literacy that puts the bulk of a culture’s literature off limits.”
    It’s my understanding that this literature is unintelligible (without special training) already in hanzi, traditional or simplified. I’ve been told by chinese people (university educated) that they can’t make heads or tails of it.
    I think that just transcribing texts into pinyin isn’t the generally best idea, a separate, recongisable written style for pinyin would make the most sense.
    “In Taiwan, many things (books in bookstores, etc.) are already sorted by zhuyin fuhao order. Pinyin is not the only option for “alphabetical” sorting.”
    Cool, I also like zhuyin fuhao and am generally suprised that taiwanese don’t use it more.

  41. The claim is made “if the Chinese can understand a sentence when they hear it, they can understand the same sentence when they read a phonetic rendition of it.”
    I do not accept this claim. Because of the small number of distinct syllables, spoken Mandarin is quite ambiguous, and the apparatus used to disambiguate in spoken discourse is very different from that available in writing. In spoken discourse, context, gesture, intonation, and idiom combine to fulfill functions which would be poorly fulfilled by phonetic transcription, but which are wonderfully fulfilled by the signifying roles of the radicals which make up each hanzi character.
    Moreover, the probability distribution of word juxtapositions is very different in spoken and written forms. In spoken forms, the most frequently used idioms are used much more often, and the least frequently used, are much less so, as compared to written forms. This means that written language is inherently more difficult to disambiguate with high confidence on the basis of phoneme strings alone.
    Even so, the original claim would have much more truth if “Chinese” is understood to mean “Beijing Mandarin”, but in fact “Chinese” is a number of dialects and also distinct languages, and “Beijing Mandarin” is a minority part of that millieu. In the larger context, phonetic writing would be utterly useless. This last point was already made on this page, from several angles, but it is worth repeating in the specific context of rebuttal to the quoted claim.

  42. David Marjanović says

    Generally, China implements a policy of imposing one language (putonghua) for “official” purposes. To the extent that this draconian policy will be successful, an alphabetic code will clearly be the way to go.

    And indeed, Pinyin is being used in PRC schools to teach children Putonghua pronunciation.

    It’s my understanding that this literature is unintelligible (without special training) already in hanzi, traditional or simplified.

    Indeed. Imagine Latin being written in an extremely abbreviated style, with half of all words missing and half of every remaining word missing.

    I do not favour spelling reform to match some spurious standard pronunciation. Which standard?

    For one thing, just such an artificial situation is common to many languages. Take German — or even Mandarin: true Putonghua, despite its name (“ordinary speech”), isn’t anyone’s mother tongue. But so what: “In summary, then… as long as people understand the ways accents vary […], there is no reason to imagine that there are any insurmountable problems here – how many of the people who claim that creating a pandialectal scheme is impossible have ever even tried?” (source)

    I thought Chinese had the problem of having a ridiculous number of homophones that made converting to an alphabetic system quite difficult.

    Classical Chinese is commonly unintelligible when read aloud, yes. Modern Mandarin doesn’t seem to have more homophones than English to me.

    No, it couldn’t. Not under any economic system currently operating on this planet. No more than every literate American, Indonesian or Mexican is somehow magically granted the right to live a life of care-free creativity and prosperity.

    Excuse me? I didn’t claim any such thing. I never said universal literacy automatically brings about heaven on earth. It’s neither sufficient nor apparently necessary for any such goal.
    It clearly helps, though…

    Labeling hanzi a peripheral issue does in fact sound quite like the old arguments about the struggle for gender equality being unnecessary since everbody would be equal once socialism was realized.

    Well said, unfortunately.

    Michael–I think sorting names, etc., by stroke number works just as well for native Chinese speakers

    Even with lots of practice it takes a lot of time. Plus, with the simplified characters, lots and lots of different ordering schemes popped up…

    Also, at least in Taiwan, many things (books in bookstores, etc.) are already sorted by zhuyin fuhao order.

    That is not the case in the PRC.
    BTW, have a look at this example of trigraphia from Taiwan. Yes, it is Mandarin.

  43. David Marjanović says

    Huh? How did my comment get dated back in time? It’s now above 3 comments that were already there before I started to write.

    Moreover, the probability distribution of word juxtapositions is very different in spoken and written forms.

    What are you referring to? Half-classical-half-vernacular literature?

  44. No one can understand most of Chinese literature other than novels from before the 20th century just by listening to it as if it were everyday speech. Transcibing from hanzi to pinyin simply makes this literature unintelligble. It’s a poor kind of literacy that puts the bulk of a culture’s literature off limits.
    What michael farris said. Pre-20C Chinese literature is as intelligible to the vast bulk of Chinese speakers as Chaucer (or maybe Old English, depending on what kind of literature you’re talking about) is to modern English speakers. Defenders of the current system seem to have a fantasy of a country full of people happily reading Tang dynasty classics in their spare time. ‘Taint so. Yes, there will be losses in the (extremely unlikely) event there is a switchover; nobody denies that, and it would be impossible to ignore it, since whenever you bring up the idea of getting rid of the characters the same objections are instantly brought up: what about the literature? what about the unity of China? what about the beauty/tradition/cultural values?
    My answer to all this is twofold: 1) the virtues of the current system in all these areas is exaggerated; 2) it’s not worth the tradeoff.
    But don’t worry, it’s all theoretical. The most we can hope for is increasing use of pinyin on the side (as michael ferris suggests).

  45. People don’t read Tang classics in their spare time, but I have never goten the impression that nayone thinks of Tang poetry as especially obscure or difficult, or if it is, it is where a poem relies on allusions, not because the langugae is obscure. In fact Tang poets like Li Bo make a lot more sense to this English-speaker than verbose rap artists like Shakespeare or Milton, but maybe that’s just a matter of taste.
    If pinyin were such a wonderful solution, it would have swamped hanzi by now. Maybe – certainly – the print media are tightly controlled, but what is left of the chinese blogosphere has opted for hanzi. The users of the language have had decades to make their choice, and maybe they already have.

  46. Well, that’s a good point. I certainly don’t want to be plus chinois que les chinois. If Today’s Youth are happy with hanzi, more power to them; I just have a reflex response to foreigners insisting on the glorious necessity of the characters.

  47. Things I’ve noticed in these comments:
    1) (Seemingly) No fluent Chinese (of any dialect) language persons.
    2) Persons who (seem to) have never studied or lived in a Chinese culture (especially using traditional characters) commenting very openly about the pros and cons of various such issues.
    3) A few people who have a little training in the Chinese language system and culture.
    All in all, I’d say we’re pretty unqualified to say anything. Having said that, I’ll make another comment.
    Pinyin works ok so long as there is a common understanding of the topic of communication. However, it crumbles as soon as that commonality isn’t there. It’s worse still if one wishes to incorporate even slightly subtle ‘word-plays’. All Pinyins are simply pronunciation systems to transcribe the sounds of speech. Using them to replace a writing system per se reverts everyone back to approximately elementary school level language.
    I don’t see too many people advocating the reform of English writing to a pure pronunciation system.
    BTW: As a Chinese student, I became very annoyed with Hanyu Pinyin, i.e. the standard Pinyin. Why? Because it isn’t strictly phonetic. The creators of it decided to mix the idiosyncrasies of true written language, e.g. unwritten sounds & ambiguous pronunciation depending on the following letter, into it. As a person, if I’m going to go to all that effort to build something and make it a standard, I’d make darn sure that it was not as disgraceful as the lumbering oaf called Hanyu Pinyin.

  48. LH: I know what you mean about the hanzi-worshipers, but there are foreigners with a knee-jerk pro-alphabeticization bias as well. I just think it’s worth pointing out that there are other ways of organizing things (and other strategies for representing a character you can’t write).
    DanW: Fluent in Mandarin, 5+ years in full-form-using Taiwan here. Hope that qualifies me at least a little bit. Hanyu Pinyin isn’t so bad once you get the hang of it

  49. I think Mair has some valid points, but it seems to me that the fundamental cause of illiteracy in China is not the characters, but the enormous gulfs (as he mentions) in wealth, status, and opportunities for education. With all the ‘China rising’ hype, we forget that, what, 700-800 million people are still peasants? And are being increasingly, if anything, crapped upon by their urban brethren.

  50. caffeind says

    Much of this has already been said better in http://languagehat.com/archives/002507.php and other previous posts and comments, so I’ll try not to repeat.
    I’m not at all sure that writing in Hanyu Pinyin would be easier for the provincial masses. Producing correctly spelled Hanyu Pinyin requires complete fluency in the standard pronunciation of the national language, including distinctions like the retroflex sounds that are ignored by most speakers. The universal lack of tone marking on Hanyu Pinyin seen in China may also have something do with the lack of consistency in tones in dialects even short distances from Beijing.
    Here’s an example of badly spelled Hanyu Pinyin chat: http://community.livejournal.com/zhongwen/320549.html?#cutid1
    Some of the words were incomprehensible until I saw the character equivalents.
    The most I had to rely on Chinese characters for communication was with an innkeeper in the mountains west of Beijing whose thick dialect I couldn’t understand. I had to keep asking him to write down what he was saying.
    Another example: I didn’t recognize a word that some Sichuanese tea merchants (in Beijing, speaking in Putonghua) were using, asked them to write it, looked it up in the dictionary I was carrying, found that the pinyin was different than the pronunciation they were using, showed it to them and they said they had never heard the dictionary’s pronunciation.
    It is quite possible that as China continues to become more integrated and developed, people’s speech will approach the standard and Hanyu Pinyin will be more usable as a means of writing. But by this point, literacy in the traditional script will likely also approach the levels in Taiwan or Japan.
    I have yet to hear any Chinese say that Hanyu Pinyin is more comfortable to use, or read any article by a Chinese advocating alphabetization that was not written 50 years ago or more. Many of the valid pro-alphabet points in Lü Shuxiang’s essay http://pinyin.info/readings/lyu_shuxiang/ have now been made obsolete by technology.
    More complex writing systems may be more complex to learn and to write, but can sometimes offer advantages in reading speed. A computerized world has mitigated the difficulties of the former, and put even more of a premium on the latter.

  51. LH says I just have a reflex response to foreigners insisting on the glorious necessity of the characters.
    Well I have a reflex response to foreigners insisting the Chinese change their writing system to meet Western notions of utility. The only people I ever seem to hear proposing abolishing or radically reducing the use of characters are foreigners. And people who write like Mair does – sneering, tendentious, and snide – are certainly not good ambassadors for the reform cause.

  52. What a bizarre discussion.

  53. Persons who (seem to) have never studied or lived in a Chinese culture (especially using traditional characters)
    For what it’s worth, I’ve lived in Taiwan and studied Chinese using traditional characters (which I like a lot, hard as they are to learn). And I agree that pinyin is far from an ideal system.

  54. A year in China instilled in me a preference for the simplified characters, for little reason other than that they were the ones I saw most often, and because they were…simple. Relatively, anyway.

  55. Caffeind – Another example for you of needing hanzi to communicate – try interviewing a container-load or worse yet, a boatloas of illegfal aliens form Fuzhou. You need, need hanzi, to get even the most basic information form them, such as name and address. They simply cannot do it in pinyin. They start Mandarin as a foreign language in the sixth grade, and far from everybody gets that far.
    And it is not a merely bureaucratic matter. An immgration judge, formerly immigration lawyer, recounted once how she had gotten all proceedings against a client dismissed because the charging documents had such a confused form of his name that she could argue he hadn’t nearly been identified in them.
    Also, FWITW – the simplified characters are often just cursive versions of traditional characters, or phonetic abbreviations, like ‘nite’ for ‘night’, and of course that can irritate traditionalists. most people are passingly familar or can guess them out even if they are raised using traditional characters.

  56. michael farris says

    “Also, FWITW – the simplified characters are often just cursive versions of traditional characters”
    True, but lots aren’t And lots make no sense (from analyses I’ve read I’m too ignroant of hanzi to say anything definitive).
    IINM the primary concern of the simplifiers was reducing the number of strokes and this overrode all other considerations to an absurd degree.
    A co-worker (not chinese but fluent in mandarin and can read classical chinese) told me there was lots of room for simpification and regularization of traditional characters without greatly altering their interrelations but that the simplified characters as they exist are a largely bungled job.

  57. If there was a move to alphabetic writing and a resurgence of the dialects, you’d have to produce standardized versions of the dialects and decisions as to how many would be formally recognized (by courts, the schools, etc. Within the present political system this is extremely unlikely (in the sense of “impossible”). So you’d end up with alphabetic standard Mandarin as the national language, when it’s a foreign language to at least 30% of the population and a difficult dialect for speakers of non-standard Mandarin.

  58. “if you consider hanzi an indispensible part of Chinese culture
    Which of course they’re not, any more than the Perso-Arabic alphabet was an indispensible part of Turkish culture, though before Atatürk’s reform many people would have said it was.”
    I’m surprised you would say something so ill-informed. Apparently you don’t understand Chinese art, which is based in large part on Chinese Calligraphy. Chinese Calligraphy is not like the Western concept of Calligraphy, which strives for uniformity and is merely a nice flourish, not a centerpiece. Chinese Calligraphy has an extremely rich tradition, being at the root of the derivatives of seal-carving and Chinese painting, as well as being intertwined with Chinese literature, history and religion. It is precisely the meeting of meaning and image which gives Chinese Calligraphy it’s weight and value. It’s not an issue of ‘romanticism’, unless art is too uneccessary for you. It is an issue of culture. Oh, and literacy? Perhaps that has a little more to do with faulty education systems, especially considering that our ‘efficient’ alphabet has yet to spontaneously creat high literacy rates among America’s poorer demographics.

  59. David Marjanović says

    I don’t see too many people advocating the reform of English writing to a pure pronunciation system.

    All puns would still work then, though.
    (Besides, the only reason more people don’t advocate a sane orthography for English is that the current one has a huge Installed User Base. Compare QWERTY and VHS.)

    BTW: As a Chinese student, I became very annoyed with Hanyu Pinyin, i.e. the standard Pinyin. Why? Because it isn’t strictly phonetic. The creators of it decided to mix the idiosyncrasies of true written language, e.g. unwritten sounds & ambiguous pronunciation depending on the following letter, into it. As a person, if I’m going to go to all that effort to build something and make it a standard, I’d make darn sure that it was not as disgraceful as the lumbering oaf called Hanyu Pinyin.

    It’s not one-letter-one-sound, that’s right. But, unlike English, it follows a few clear rules all of the time. There is exactly one way to pronounce any Pinyin syllable. Pinyin simply goes as far as you can with the Latin alphabet without introducing any special characters other than the tone marks and ü; that necessitates the digraphs.
    What happens to e, i, and u, however, can be argued to be phonemic. Depending on the analysis of Mandarin, Pinyin is a perfectly phonemic orthography, plus a few added phonetic features that are intended to help in the main purpose of Pinyin (teaching Standard Mandarin pronunciation to nonnative speakers in China), like why o is written behind b, p, m, f, uo behind l and n, and e elsewhere.

    Producing correctly spelled Hanyu Pinyin requires complete fluency in the standard pronunciation of the national language, including distinctions like the retroflex sounds that are ignored by most speakers.

    That is true.
    However, it’s also required for German, if not more so. What is called German has a dialect diversity that seems at least comparable to that of Mandarin. (Just Mandarin, mind you, not Wu or Cantonese or anything.)
    It’s also nothing that’s new to China. Writing in characters only works for dialects that aren’t too far away from the standard lexically.

    The universal lack of tone marking on Hanyu Pinyin seen in China may also have something do with the lack of consistency in tones in dialects even short distances from Beijing.

    That’s an interesting idea. Sounds convincing.

    It is quite possible that as China continues to become more integrated and developed, people’s speech will approach the standard and Hanyu Pinyin will be more usable as a means of writing. But by this point, literacy in the traditional script will likely also approach the levels in Taiwan or Japan.

    Check the link about Japan again… 🙂

    So you’d end up with alphabetic standard Mandarin as the national language, when it’s a foreign language to at least 30% of the population and a difficult dialect for speakers of non-standard Mandarin.

    That is, of course, already the case.

    told me there was lots of room for simpification and regularization of traditional characters without greatly altering their interrelations but that the simplified characters as they exist are a largely bungled job.

    They make writing easier and reading more difficult, it seems… the reform was implemented, and most likely planned, in steps over 30 years…

    if you consider hanzi an indispensible part of Chinese culture
    Which of course they’re not, any more than the Perso-Arabic alphabet was an indispensible part of Turkish culture, though before Atatürk’s reform many people would have said it was.

    Of course it was, and that’s a large part of why Atatürk made the reform at all (and why he made it overnight, without a transition period AFAIK), together with a thoroughly botched reform of the language itself that was aimed at eliminating the countless Arabic and Persian loans, and of course with the general Westernizing-by-force of the country. If it were possible to impose this kind of thing on a billion people, guess what the Culture Revolution would have looked like!
    However, Arabic calligraphy still exists in Turkey, and AFAIK much of the classic literature has been transcribed and reprinted. Nobody suggests forbidding Chinese characters overnight; the Turkish culture-break scenario will not happen in China no matter what else happens.

    I have yet to hear any Chinese say that Hanyu Pinyin is more comfortable to use, or read any article by a Chinese advocating alphabetization that was not written 50 years ago or more.

    On the latter point, though, I don’t expect any to come out of the PRC… they would be against the party line…

    Perhaps that has a little more to do with faulty education systems, especially considering that our ‘efficient’ alphabet has yet to spontaneously creat high literacy rates among America’s poorer demographics.

    I agree on the education system, but don’t forget that English has the worst orthography of any language that uses an alphabet or syllabary, worse than Irish, worse than Mongolian-in-the-Mongolian-script, worse than Tibetan, worse than French, worse than everything. This goes so far that learning to spell in English (or French, BTW) is easier for me than for a native speaker, because I usually learn pronunciation, meaning, and spelling at the same time, or if anything, spelling first, while native speakers learn the spelling last for most words.
    (Of course, this works the same with Chinese characters, except they don’t consist of a selection of limited size of 26 x 2 letters in an arbitrary linear arrangement, but of a selection of highly variable size of 12 strokes in an arbitrary two-dimensional arrangement. Even words like Leicester or Tucson or through or read/read/read are much easier to learn than most Chinese characters.)

  60. Noetica says

    Useful detailed analysis, DM.

  61. I’m surprised you would say something so ill-informed. Apparently you don’t understand Chinese art, which is based in large part on Chinese Calligraphy.
    Let’s try to read a little more carefully and make fewer insulting assumptions, shall we? I didn’t say characters weren’t important, or valuable, or much loved. I said they weren’t indispensable. Which they’re not, any more than the Perso-Arabic alphabet was an indispensable part of Turkish culture, though before Atatürk’s reform many people would have said it was. Perhaps you’re not aware that Islamic art is traditionally based to a large extent on calligraphy? Or perhaps you just think Islamic art is somehow less important than Chinese art? I never said, nor do I believe, that switching from characters to another system would not involve significant loss; that would be an idiotic thing to say or believe. I said (more than once, but I’ll say it again for your benefit) that the tradeoff is too great. If you find that hard to understand, let me know and I’ll try to make it clearer for you.
    Useful detailed analysis, DM.
    Agreed!

  62. Noetica says

    Let’s try to read a little more carefully and make fewer insulting assumptions, shall we?
    Agreed!

  63. michael farris says

    “together with a thoroughly botched reform of the language itself (Turkish: maf) that was aimed at eliminating the countless Arabic and Persian loans”
    But can that be laid at Ataturk’s feet? There was a significant problem with Ottoman Turkish in that the majority of Turks couldn’t use or understand it (making modernization difficult). Eliminating Persian/Arabic grammar from Ottoman and replacing an awkard misleading script with something that is pretty far from great but is quite good enough were absolutely good things (I agree that the drive to ‘purify’ the lexicon of words that everybody knew was not constructive or wise).
    “Nobody suggests forbidding Chinese characters overnight”
    I wasn’t aware of anyone who was in favor of ‘forbidding’ or eliminating Chinese characters at all. My own personal view is that they are not an optimal daily script for a modern state language and that parallel script(s) would do no harm and would possibly benefit people. I also think pinyin (or bopomofo) if given a chance would eventually take over most of the daily functions of characters but that characters would still have a place (especially in some types of literature, calligraphy, ceremonial purposes etc)
    “This goes so far that learning to spell in English (or French, BTW) is easier for me than for a native speaker, because I usually learn pronunciation, meaning, and spelling at the same time, or if anything, spelling first, while native speakers learn the spelling last for most words.”
    I think the situation for native speakers is a little more complex than that (nb. the following concerns everyday literate native speakers of English and not the super-literates that populate the comments section here).
    What I’ve noticed is that a person’s spoken and written vocabularies are often somewhat disconnected. That is, there are words they’re used to hearing (but don’t necessarily come across in written form) and words that they’re used to reading (but don’t necessarily hear often). If you listen to individual English speakers long enough, almost every native speaker has a list (of varying length) of words they consistently pronounce in odd ways. Mostly this is from imagining pronunciations of words they know first in written form.
    Sometimes people have problems connecting the written and spoken forms of uncommon words though they may know both. When I was learning to read and write, for a long time I thought the spoken word ‘island’ and the written word ‘island’ were separate words, but that one was spoken more often (and probably written ‘iland’ or ‘ighland’) and the other was printed more often (and probably pronounced izland)… This kind of thing happens a lot and often lasts into adulthood.
    In relation to this and the concept of script reform, I also think that it would be fairly painless to simplify some spellings (affecting about 10 % [wild guess] of the word stock) in a way that would eliminate 80-90 % of the worst difficulties in learning to spell English (without having it devolve into unintelligible national varieties, though I’m generally in favor of that as well).

  64. … together with a thoroughly botched reform of the language itself that was aimed at eliminating the countless Arabic and Persian loans,

    If you have JSTOR access, this article (‘Language Reform in Turkey and Iran’, John R. Perry) suggests that was the orginal idea, but that it was intentionally not taken to extremes:

    ‘… even the Turkish Language Society’s truculent purism had somewhat abated by November 1935, when it published its selection of the best “substitute vocabulary” so far collected in two pocket dictionaries, the Osmanlıcadan—Türkçeye Cep Kilavuzu and vice versa. Its stated goals—and the actual results, insofar as could be judged—were not not so much to maintain the racial purity of the language as to eliminate diglossia—i.e., to bring the written language into concordance with the spoken, for the practical purpose of expanding education at all levels. Thus many assimilated loanwords had, sensibly, been retained; this was unashamedly rationalized in the Society’s journal with the claim that etymological research had proved many words formerly thought to be foreign loans to be really Turkish in origin. This attitude, crystallized after the Third Language Congress of August 1936 in the Güneş-Dil Teorisi or “Sun-Language Theory,” had the salutory result that it was “no longer necessary to sacrifice any word needed in our language and familiar to the people.” Some Turkish writers have surmised that Atatürk himself launched this ultranationalistic and blatantly unscientific piece of hokum, tongue in cheek, specifically to pull the rug from under the extremists of the purist movement. […] Whatever its origins, the plan worked. Atatürk himself set the tone by rehabilitating Arabic loanwords such as millet “nation” in his speeches from late 1935 on, having previously used the purist replacement ulus.’

    ‘Successfully and intentionally eliminating diglossia’ is not any sort of ‘thoroughly botched’. 🙂 . The rest of your comment is relevant, well-articulated and correct to my knowledge.

  65. Suppose that the great and good have made up their mind and decided that China should adopt some alphabetic script or other. Now consider the difficulty of the transition.
    One of the reasons the transition in Turkey was so successful is that literacy in Turkey was only 9% at the time (http://home.vicnet.net.au/~ozideas/wturk.htm). The slate was to a large extent blank. I understand this isn’t so in China. There is a large part of the population with a stake in the current writing system, and for them the transition, for all its advantages, can be very difficult.
    I know that although I don’t know one word in Chinese. Here in Israel one hears from time to time about the great advantages of “latinazation”. In fact some very knowledgable people have advocated such reform. A few years ago I read an article with examples of Latinized Hebrew text. I could read this text, for sure, but it wasn’t nearly as easy as reading “normal” Hebrew. Now what reason do I have to practise the new writing system when whatever information I need is available in the age old (and to some sacred) Hebrew characters.
    I try to imagine how newspapers, book publishers, schools, offices large and small and everybody else in a mostly literate society will make such a transition and my conclusion is that this is possible when another Mao comes along and kills a few million people in the process.

  66. michael farris says

    “A few years ago I read an article with examples of Latinized Hebrew text. I could read this text, for sure, but it wasn’t nearly as easy as reading “normal” Hebrew.”
    First I’m not suggesting orthographic reform of Hebrew, okay? But if I were … surely the question is not “Which is easier for a super-literate already acquainted with the old system?” but “Which would be easier and quicker to learn for children and easier to remember and use for adults?” And the answer to that is hard to figure out because the research that could bring more or less definitive answers (if such answers there be) is mostly unethical.
    If I were to try to do research on literacy in Hebrew I might use Maltese speaking children as a control (the same way I think Vietnamese is a good control for Chinese).
    Assuming research is done and the choice is made to switch (or give people the opportunity to switch should they so desire) implementation is another question, but i think top-down decisions are unrealistic compared to gradual introduction (give children equal access to both and see which they prefer).

  67. David Cowhig says

    Victor Mair’s proposal of digraphia for the Chinese would make their system more flexible than it is now, although it wouldn’t replace the characters. The Japanese have digraphia (maybe trigraphia now). Japanese can write something in kana if they don’t know a particular character. So I think what the proposal is to mix characters and romanization or some other phonetic script with easier characters to make literacy easier.
    The causes of illiteracy go beyond the writing system although a more flexible writing system would help certainly. I remember a missionary friend in Taiwan told me twenty-five years ago that the Presbyterian Church in Taiwan some time before had published the Bible and church newletters in Taiwan (minnan) dialect since romanization illiterate Taiwanese could be taught to read and write their own dialect in romanization in just a few weeks.
    The poor quality of schools and poor funding for education, especially in the countryside is a large part of the problem. A few months ago the Chinese government announced a USD 10 billion annual subsidy to bring down the cost of text books for poor students and to make the books free in some cases. How much of the money reaches the students and how much gets pocketed by officials for other purposes as it moves from the center to the localities will be something to watch!

  68. michael farris says

    “So I think what the proposal is to mix characters and romanization or some other phonetic script with easier characters to make literacy easier.”
    I think Zhuyin Fuhao would mix with characters much more gracefully than pinyin…

  69. There is simply no such thing as an indispensable feature for any culture. It would be unthinkable for French culture to include a monarchy today; 250 years ago, it would be unthinkable for it not to. Except when cultures are artificially preserved under glass, they are continually changing, adding new features, discarding old ones.

  70. Taking a quick look through this discussion, it’s pretty obvious that most people that are speaking here have little to no knowledge of Chinese. I can’t claim that I’m an expert by any means, but I have enough experience with the language to know that simply using pinyin would be an utter disaster. Beyond the most basic sentences, reading hanzi is much easier than reading in pinyin. Because Chinese has so many homophones, sentences written simply in pinyin can be quite unclear. Also, if you see the pinyin for a word you don’t know, there is no way to guess its meaning. Once you understand enough about hanzi, on the other hand, you can sometimes make an educated guess of the meaning of character (or combination of characters) and also make an educated guess about its pronunciation (though you still can’t make an educated guess about the tone of the word). On another note, learning to read hanzi is much easier than learning to write hanzi. Literacy can be achieved without learning to write the characters by hand.
    While I know next to nothing about Vietnamese, because it is also a tonal language, it would be interesting to compare it’s romanized system with pinyin. Comparisons of Chinese with Korean and Japanese are much less valid.

  71. caffeind says

    Check the link about Japan again… 🙂

    Ok, let’s go over its points…

    • He couldn’t find the 99% figure on the Education Ministry’s website. The page also says “Last updated: 2000-08-09”.
    • A German study published in 1924 found that military recruits, who at that time would have been mostly impoverished farm boys, had forgotten (probably meaning they couldn’t remember it on an exam with no context) half the kanji they had learned in school, between leaving school after 6th grade and entering the army, which should be 5 years or more.
    • Furigana have been and are used, even on every kanji in some unpretentious texts. This only shows that furigana are considered useful enough to continue to include them. Since Japanese publishing has always been set up for furigana and pervasive furigana use was normal in the Edo and prewar periods, this is a pretty low threshold. It is more likely that furigana are excluded from more prestigious texts to avoid the impression of insulting the reader’s intelligence. English texts that want to be understood by the whole population and are not afraid of talking down often also include pronunciation or definition for difficult words. And, don’t forget that in postwar Japan, the Self-Defense Force is considered the bottom of the barrel.
    • “A Chinese who forgot the character for a particular word would have basically no way to write it down.” Well, this shows he’s no follower of DeFrancis, who vigorously disproved this point, showing that when needed, Chinese script functions as a large syllabary in non-prescriptive contexts. For that matter, the Japanese on-yomi also provide a rough reminder of pronunciation for phonetic-semantic compound characters (which is what most obscure kanji which get forgotten will be), which is like seeing a difficult English word, getting the pronunciation vaguely right from the spelling, and remembering a spoken word that matches. It’s only Japanese kun-yomi that have no phonetic value.
  72. David Marjanović says

    But can that be laid at Ataturk’s feet? There was a significant problem with Ottoman Turkish in that the majority of Turks couldn’t use or understand it (making modernization difficult). Eliminating Persian/Arabic grammar from Ottoman and replacing an awkard misleading script with something that is pretty far from great but is quite good enough were absolutely good things

    I agree.
    (Except, the only thing that’s IMHO not great about the Turkish alphabet is how J is wasted on a sound that seems to be completely restricted to French loans.)

    (I agree that the drive to ‘purify’ the lexicon of words that everybody knew was not constructive or wise).

    We had the anecdote about how Atatürk himself didn’t understand a speech he was supposed to give, and how only the Sun Language Theory managed to stop the replacements of loans with cranky neologisms… it came to fighting crankery with crankery! :o)

    I also think pinyin (or bopomofo) if given a chance would eventually take over most of the daily functions of characters but that characters would still have a place (especially in some types of literature, calligraphy, ceremonial purposes etc)

    Agreed.

    In relation to this and the concept of script reform, I also think that it would be fairly painless to simplify some spellings (affecting about 10 % [wild guess] of the word stock) in a way that would eliminate 80-90 % of the worst difficulties in learning to spell English

    Agreed again.

    a missionary friend in Taiwan told me twenty-five years ago that the Presbyterian Church in Taiwan some time before had published the Bible and church newletters in Taiwan (minnan)

    Yep. There’s a whole Wikipedia in this spelling, and it’s also used in the dissertation abstract mentioned above (way above). However, almost nobody outside the Presbyterian Church knows this spelling. Funny how the centuries-old connection between script and religion just never goes away…

    I think Zhuyin Fuhao would mix with characters much more gracefully than pinyin…

    Obviously. It was even designed for that purpose.

    Beyond the most basic sentences, reading hanzi is much easier than reading in pinyin. Because Chinese has so many homophones, sentences written simply in pinyin can be quite unclear.

    If the tone marks are missing, yes, at least if there’s not enough context…

    Also, if you see the pinyin for a word you don’t know, there is no way to guess its meaning.

    Except if you already know words that are pronounced that way (e. g. if you’re a native speaker). When there are several homophones, context will sort them out, as it does in the spoken language. I don’t think confusion between “lesson” and “quarter of an hour” is likely, even though both are ”kè”.

    Once you understand enough about hanzi, on the other hand, you can sometimes make an educated guess of the meaning of character (or combination of characters) and also make an educated guess about its pronunciation (though you still can’t make an educated guess about the tone of the word).

    Yes — enough. Under 1000 characters don’t bother trying.

    On another note, learning to read hanzi is much easier than learning to write hanzi.

    Agreed…

    Literacy can be achieved without learning to write the characters by hand.

    If so, I would have felt three times as literate when I was in China (for a scientific congress). I came across lots of characters where I felt at first I ought to recognize them, and then I noticed what I recognized was just a part of the character; there are just too many characters that are quite similar but not identical, and these you can only tell apart if you really know the entire character by heart = if you can write it.

    While I know next to nothing about Vietnamese, because it is also a tonal language, it would be interesting to compare it’s romanized system with pinyin.

    Well, similar, except that the Vietnamese alphabet was developed in the 16th century, and therefore preserves features that have since disappeared from the language and contains oddities like the absence of ”k” except in front of ”e” and ”i”. Oh, and one more thing — in Vietnamese practically all syllables, even in loans that are unanalyzable in the original and in Vietnamese, are written as separate words, a leftover from the Chinese characters; that’s not done in pinyin.

    “A Chinese who forgot the character for a particular word would have basically no way to write it down.” Well, this shows he’s no follower of DeFrancis, who vigorously disproved this point, showing that when needed, Chinese script functions as a large syllabary in non-prescriptive contexts.

    Yes, a large, ambiguous syllabary.

  73. David Marjanović says

    I also think pinyin (or bopomofo) if given a chance would eventually take over most of the daily functions of characters but that characters would still have a place (especially in some types of literature, calligraphy, ceremonial purposes etc)

    On this point again: I’m still fascinated by what I once saw on TV: the tombstone or what it was of a certain Khan Bilge. It’s in Old Turkic, in Orkhon runes (one of the most important documents in that script) — except for the headline, which consists of three masterfully chiseled Chinese characters. 🙂

  74. As a theoretical matter, I think allowing digraphia would be a good step in alleviating illiteracy.
    Also, I don’t buy the argument that allowing for digraphia would cause the downfall of characters. The 高考 (college entrance exam), and other tests, will always contain characters, and therefore, millions of kids around China will always learn to read characters. Characters, despite their disadvantages, fit the language pretty well.
    However, from a practical and political point of view, the poor have no “voice” in the way China is governed. Also, I seriously doubt that anyone in the Ministry of Education spends any amount of time contemplating the merits of digraphia. If these officials who are obsessed with centralized power do think about it, they might realize that allowing for alphabets could produce more than one standardized and popularized alphabet in order to fit the various regionalects. This could possibly lead towards stronger regional identities, which the central government generally discourages. Thus, any patriotic official would squash heretical alphabetization, in the name of the Motherland. (Of course I’m just speculating).
    Also, as some other posters pointed out, a popularized alphabet could only succeed if it were widely used and if the materials made using it were interesting and relevant to the lives of the target audiences (i.e. formerly-illiterate farmers). Hmmmm…..who has benefited the least from the unequal economic development of the last decade? Farmers. Who gets screwed by the system when they enter the cities? Farmers. Who have been screwed over by the hukou system? Farmers. Where are there thousands of riots and disturbances every year? Almost solely in the countryside. Are these disturbances linked nationwide, or localized? Localized. So, then, from the government’s point of view, would it be a good idea to adopt measures that might lead towards greater political consciousness among farmers through literacy? Absolutely not.
    Digraphia- DOA.

  75. michael farris says

    “Except, the only thing that’s IMHO not great about the Turkish alphabet is how J is wasted on a sound that seems to be completely restricted to French loans”
    Yeah, I wish that j had been used instead for c or y (either one).
    I also think the word-separation rules could have been better chosen, I think bili yorum and bil miyorum make more sense than biliyorum bilmiyorum (not to mention that it would be nice to distinguish git me and gitme or maybe even olacag im and olcagim).
    There’s also the unclear and unstandardized status of the circumflexes.
    But those are mostly quibbles as opposed to serious complaints.

  76. I’m no expert on any of the Chinese stuff here. But I know a bit about another “Ch-” language where issues like these were debated very early on in the development of its writing system (likely debated within the head of just one person actually…) The result was a resounding success; take from it what you may.
    Sequoyah, a completely illiterate Cherokee, devised a syllabic writing system for his people, but not without a lot of trial and error. He first tried a system of one symbol per word or concept and found it extraordinarily unwieldy. It was then that he came up with the syllabic system still in (albeit limited) use today.
    While the print-font that he developed to supplant his original handwritten syllabary contains a lot of obvious borrowings from Roman and Cyrillic/Greek alphabets, this does not detract from his genius. (a) There’s no correspondence of any of those borrowings to any of the alphabets “borrowed from”, showing a clear creative stamp on his part but, more importantly, (b) The true genius lies in the fact that coming from a standpoint of total illiteracy he was able to discern on his own that a syllabic system would be the best system; he likely hardly knew what a “syllable” even was (in the formal sense) to begin with. It’s a highly organic and creative writing system to say the least.
    Anyway… one could wonder what would happen if (in theory) the Chinese had no written language for centuries and a Chinese “Sequoyah type” would one day feel moved to devise a system to try to get “leaves to talk”. Unencumbered by any standing tradition or presumption of his/her own literacy, might he/she be moved to go the syllabic route as well?
    Yes, I know there’s a lot about Chinese writing that’s already phonetic/syllabic and I won’t purport to know any more about that than I already do, which is basically nothing, but the question is, would someone “starting from scratch” like Sequoyah see some merit in a _purely syllabic system for Chinese…? Or would some other system “rise to the top” as the best choice…?
    That may provide one of the keys to this ongoing debate… The question may simply be thus: Which system for writing Chinese _inherently would be most advantageous with all extenuating factors being equal*? (* – which of course I know they’re not in reality…)

  77. michael farris says

    “it would be nice to distinguish git me and gitme”
    Oops! I meant “git mem” (I don’t go (aorist)) and “gitmem” “my going”

  78. David Marjanović says

    On literacy in Japan
    Lots more information about the 99 % figure and related matters. It’s an entire book chapter. The rest of the book is not online.
    On Cherokee:
    – That the meanings of the symbols have nothing to do with those of the letters they are derived from is not due to some extraordinary creativity; it’s just because Sequoyah had a book (in English presumably) and knew what that was but, as you write, was illiterate.
    – Cherokee consists mostly of CV (consonant-vowel) syllables, AFAIK, and at that time it had no tones (the Oklahoma dialects have developed a lot of tones). That makes a syllabary workable. Having a lot more consonants and diphthongs and triphthongs, Standard Mandarin has, I’m told, 412 syllables not counting the tones; with the tones the number lies about 3 times as high, if not more. (It’s not 4 times because not every syllable exists in all 4 tones, but 1200 is a large enough number!)
    – I find syllables much easier to understand than isolated sounds. Especially plosives that can’t even be pronounced in isolation. I’ve read Japanese children learn hiragana from their mothers before entering kindergarten; wouldn’t surprise me. (How many hiragana are there in modern Japanese, 47?)
    – Finally, a syllabary for Chinese exists. It’s called Nüshu and was developed by women who were not allowed to learn to read but had nevertheless learned a respectable number of characters and used them for their sound values. Nüshu never got very far. Even geographically it has stayed a very restricted phenomenon. There are clearly lots of political and social reasons for that, but the large number of characters it has certainly played a role.

  79. DM,
    Thank you for the comments. I think w.r.t. the “creativity” of some of the letters I meant to convey something else – that, by accident or by design, Sequoyah introduced Roman/Greek/Cyrillic letters (and a limited no. of that) that were not beholden to European equivalents… and thus not “Eurocentric”… I know it’s a stretch but I take some pride in the thought that the syllabary is “Americo-centric”… ;^)
    That said, one can still note the great number of letters in the Cherokee print font (i.e., the non-handwritten syllabary that’s long been in wide use) that clearly trace to Sequoyah’s pure creative effort; the more “squiggly” the letter, the more likely it was definitely “his”… ;^)
    I appreciate the insight on Chinese tones, etc., and the no. of possible syllables. Correct, Cherokee is largely made of CV syllables. The only exceptions in the syllabary are: pure “vowel” syllables (of course); the “s” character which is, simply, just “s”; and a couple of symbols that convey an “intrusive h” mixed in w/ the syllable (e.g., “nah” as compared to “na”).
    The current 85-character syllabary represents a compromise between economy and “coverage”. There are many symbols that could be represented but aren’t (e.g., “to” vs. “do”, “tu” vs. “du”); for now, a single symbol covers each of many such “pairs”. Also largely not covered is the “intrusive h” in many contexts. But… it certainly conveys things phonetically much better than English or French.
    Re: tones, I thought they were always present in Cherokee, though I understand they’re more prevalent in OK Cherokee vs. NC Cherokee. That said, I also understand that even though there are a decent no. of minimal pairs – and tones aren’t marked in the syllabary – there likely won’t be much confuusion between similarly spelled words – “homographs?” – especially in context.
    Why wouldn’t Nushu be a workable alternative for Chinese? Or put another way… if political/cultural baggage is put aside, would Nushu be a viable system in theory…? (i.e., the kind of thing a “Chinese Sequoyah” might come up with in some parallel universe…?)

  80. caffeind says

    allowing for alphabets could produce more than one standardized and popularized alphabet in order to fit the various regionalects. This could possibly lead towards stronger regional identities, which the central government generally discourages.
    The most distinct regionalects are in the coastal provinces, which under the current system are doing much better than the interior. It’s quite possible that literacy in Mandarin with hanzi is higher in the non-Northern dialect areas than in the Northern dialect areas.
    It’s also notable how almost no linguistic nationalism and separatism has been evident, even among the Cantonese who would be best equipped for it. The only place we see those phenomena is in Taiwan which is already independent, and even there word is that the younger generation are tending to happily enjoy Mandarin media and use Mandarin without the anti-mainlander grudges of their Taiwanese elders.
    Also, I don’t buy the argument that allowing for digraphia would cause the downfall of characters. The 高考 (college entrance exam), and other tests, will always contain characters, and therefore, millions of kids around China will always learn to read characters.
    Well, that pretty well describes the current situation. In five years even the poorest farmers will have mobile phones and be keying in text messages in pinyin. Digraphia is already there; it’s just that no preference for leaving output in pinyin has emerged.
    Yes, a large, ambiguous syllabary.
    Ambiguity doesn’t prevent it from functioning. This is a desire for a neat system, not a system that works.
    Why wouldn’t Nushu be a workable alternative for Chinese?
    http://en.wikipedia.org/wiki/Nushu notes “It is not clear whether Chengguan (the dialect represented by Nushu) is Mien of the Hmong-Mien languages, a Kadai language, or a local dialect of Xiang Chinese, all of which are spoken by people officially classified as Yao.” And, Nushu is no smaller than the estimated (DeFrancis) 700 or so characters of hanzi already functioning as a syllabary.
    what would happen if (in theory) the Chinese had no written language for centuries and a Chinese “Sequoyah type” would one day feel moved to devise a system to try to get “leaves to talk”. Unencumbered by any standing tradition or presumption of his/her own literacy, might he/she be moved to go the syllabic route as well?
    Almost all writing systems invented from scratch or of unknown origin seem to be syllabic with some logographic. The logographic ranges from a little to a lot, but is always far from enough to write most words in the language. No system is completely logographic, including Chinese. The alphabet apparently evolved once, in Semitic where the lack of vowel representation means that each consonant letter actually represented a syllable, and spread elsewhere by stimulus diffusion or actual copying. It is not clear that separating the consonants and vowels in a syllable into distinct written letters, some of which may be unpronounceable in isolation, is a very natural act. It may have been as difficult to grasp for the ancients as it is for us to grasp writing down tone or other features not represented in the script we are used to.

  81. caffeind says

    Korean hangul were designed to take analysis further and have shapes that correspond to phonetic features below the level of English letters.
    http://en.wikipedia.org/wiki/Hangul#Jamo_design
    This is not comprehensible to us without study. Yet, for someone brought up on a featural script, the arbitrary shapes of the Roman letters might seem like a pointless impediment to literacy.

  82. caffeind says

    In either Chinese or English, the way to make learning to write easier would be to just relax prescriptivism and stop punishing people for writing any way that seems natural to them. This has its merits, though personally I’d prefer the standard script which I can read much faster.

  83. caffeind says

    Also, the Koreans refused to abandon grouping their letters into syllabic blocks in favor of writing letters in line.
    http://en.wikipedia.org/wiki/Hangul#Linear_Hangul
    This is another feature of Latin script which makes it harder to read without long practice, and would be modified in a script designed for maximum ease of learning.

  84. caffeind says

    (in contrast to the issue of writing syllables in horizontal rows instead of vertical columns, where Chinese, Japanese and Koreans all found change easy.)

  85. “It’s also notable how almost no linguistic nationalism and separatism has been evident, even among the Cantonese who would be best equipped for it. The only place we see those phenomena is in Taiwan which is already independent, and even there word is that the younger generation are tending to happily enjoy Mandarin media and use Mandarin without the anti-mainlander grudges of their Taiwanese elders.”
    That is in line with everything that I’ve read about regionalects. Even though the idea that popularization of dialects would lead towards “separatism” seems ludicrous, the government would probably still like to err on the side of centralized control. The Wen/Hu government has, if anything, tightened restrictions on local media (newspapers, and TV stations), and has strengthened the central government’s control.
    Again, this may seem irrelevant, but it goes to the heart of the issue: why popularized digraphia is politically unlikely to happen right now.

  86. David Marjanović says

    That said, one can still note the great number of letters in the Cherokee print font (i.e., the non-handwritten syllabary that’s long been in wide use) that clearly trace to Sequoyah’s pure creative effort; the more “squiggly” the letter, the more likely it was definitely “his”… ;^)

    Yes, sorry!

    http://en.wikipedia.org/wiki/Nushu notes “It is not clear whether Chengguan (the dialect represented by Nushu) is Mien of the Hmong-Mien languages, a Kadai language, or a local dialect of Xiang Chinese, all of which are spoken by people officially classified as Yao.”

    Yes, but — apart from all the sources asserting it’s “Chinese”, and two-script texts with normal Chinese characters where many characters in both scripts are identical — Omniglot has a few examples of how some characters are pronounced; they look very similar to Mandarin to me (even though they are clearly not Mandarin).

    And, Nushu is no smaller than the estimated (DeFrancis) 700 or so characters of hanzi already functioning as a syllabary.

    Wikipedia says 700 or 800 (and that digraphs are used to make up for the rest of the syllables), Omniglot says 1000 to 1500. That’s rather hard to learn, though still easier than a real-world Chinese newspaper.

    Yet, for someone brought up on a featural script, the arbitrary shapes of the Roman letters might seem like a pointless impediment to literacy.

    Indeed they are. It’s just that the degree is so small that I don’t think it matters. Hiragana and Katakana cover Japanese completely with 47 syllables each, and the three-year-olds can reportedly read Hiragana (I’m not surprised) — there are phonemic alphabets with more letters than that.

    in Semitic where the lack of vowel representation means that each consonant letter actually represented a syllable

    Only once you ignore all the Ablaut — that’s a major abstraction, don’t you think?

    Also, the Koreans refused to abandon grouping their letters into syllabic blocks in favor of writing letters in line.

    The grouping was intended to make each syllable fit into a square the size of a Chinese character (which is also pronounced as a syllable), so both scripts can be used in the same text. In South Korea the Chinese characters are still not completely gone.

    This is another feature of Latin script which makes it harder to read without long practice, and would be modified in a script designed for maximum ease of learning.

    Eh, that may well be true for a language like Korean which everyone can segment into syllables. But it doesn’t work everywhere. In German, for example, syllable boundaries are not obvious, they are an abstraction; they tend to run through a consonant — and if you pronounce such syllables with a pause between them, you risk finding yourself in the awkward situation of having to pronounce an utterance-initial [ŋ]. Or try French. Does rien have one or two syllables? I’d say it’s in between, with a genuine semivowel (something quite rare cross-linguistically, IMHO).
    This must be why orthographic syllable separation is so diverse and arbitrary in Europe, while doing it in Pinyin is dead obvious and entirely objective. In German, the hyphen goes before an available single consonant and in a fixed but arbitrary place through a consonant cluster (including purely orthographic ones like ng. In English, it tends to go behind it. In French… I should know that, but I don’t… I think it’s more like in German. Serbocr… BCSM is apparently unique in allowing two out of three possibilities for consonant clusters: in that language it doesn’t matter if you separate зе-мља ze-mlja or зем-ља zem-lja. (I know that from having had a random look into a random book where this rule happened to be explained.)
    In short, nobody said “maximum ease of learning” (…and Hiragana clearly is very easy for native speakers of Japanese). We say “reasonable ease of learning”. 85 characters is reasonable, if perhaps pushing it (note that Sequoyah gave up instead of creating more glyphs). 3000 is not.

  87. caffeind says

    Omniglot has a few examples of how some characters are pronounced; they look very similar to Mandarin to me
    The South Chinese non-Chinese languages are phonologically similar to the South Chinese provincialects (to coin yet another word). So transporting a syllabary from one to another is probably not that hard. For that matter, Hangul could be used for Chinese, and even originally had extra letters used only for reproducing the distinctions of Middle Chinese. China could have done worse than adopting Hangul as its phonetic script.
    Omniglot says 1000 to 1500
    Is that just an estimate of the total number of syllables, or of an actual syllabary? I think DeFrancis attempted the latter, and understood that some ambiguity was tolerated.
    Only once you ignore all the Ablaut — that’s a major abstraction, don’t you think?
    Certainly, but not as big a step as inventing alphabetization of consonants, not to mention consonants and vowels, from scratch. At least they could have started with syllabics from Egyptian or wherever, and then noticed that the ones starting with the same consonant alternated when writing different forms from one Semitic root; once they were found to be interchangeable in this way, then reduction to a single symbol could occur.
    It’s just that the degree is so small that I don’t think it matters. Hiragana and Katakana cover Japanese completely with 47 syllables each, and the three-year-olds can reportedly read Hiragana
    I agree kana are well suited to Japanese, and that it’s not surprising Japanese people find them comfortable, but by “featural script” I meant Hangul. Kana are a syllabic script.
    85 characters is reasonable, if perhaps pushing it. 3000 is not.
    3000 is a pain, but that alone does not demonstrate that it is a major determinant of literacy level. At the least, other factors are more decisive.
    As far as making Latin script less boring and linear, spaces, punctuation, and lower case forms that are more distinct and extend above or below the lines, are all innovative features that aid visual comprehension and were probably accepted for that reason. Given that the Romans wrote in all caps with no spaces, it’s not as surprising that they thought reading could only be done aloud.
    I’m not clear on what if anything could be profitably added to Latin script, but it is clear that more linearity and homogeneity are not necessarily virtues. It’s also illuminating to examine other existing scripts by these criteria – Cyrillic and Hebrew and Thai are probably a bit more homogeneous (and Thai still lacks spaces!), Arabic probably more heterogeneous, at least if the dots are written clearly.

  88. marie-lucie says

    About syllabics, nobody has mentioned the Cree syllabics, first developed in Canada by a missionary, and eagerly adopted by Cree speakers. Cree (an Algonkian language spoken in the Canadian Prairies) has a very regular CVCVCV etc structure, with a fairly small number of consonants, and only 4 basic vowels (which can be long or short). Each character shape (for instance a triangle) can be oriented right, left, up or down, depending on the following vowel. A very simple and straightforward system, with only one drawback that I can see: dyslexics would have a problem recognizing the vowels if they were confused about the orientation.
    At a later date this writing system was adopted for Inuktitut (the main “Eskimoan” language in Canada), although it is less well adapted to that language (which has uvulars as well as velars, consonant clusters, and only 3 basic vowels). To my knowledge, very little has been published in Cree syllabics, but there are many Canadian publications in French and/or English + Inuktitut, the latter printed in syllabics, so that at least in Canada many people have seen samples of this writing system.

  89. caffeind says

    http://en.wikipedia.org/wiki/Canadian_syllabics
    They have a documented history of being preferred by users even in the face of outside pressure to use the Roman alphabet. The obvious explanation would be that the simple and distinct geometric shapes, and perhaps even the lack of a need for vowel letters that would make words much longer, made for superior ease of use.
    The most surprising thing in the Wikipedia article was at the end: resemblance between specific Canadian Syllabics and Devanagari letters.

  90. marie-lucie says

    Cree and Devanagari scripts: The ressemblance might not be fortuitous: many of the missionaries were very erudite and might have taken their inspiration from other scripts.
    Cherokee syllabics: Earlier, someone commented that Sequoyah had had no one to discuss his ideas with: apparently he was helped in his efforts by his granddaughter.

Speak Your Mind

*