Is a bad writing system a Good Thing?

« previous post | next post »

In commenting on language hat's link to Victor Mair's post on a "Nontrivial script fail" in Chinese, Vanya wrote:

Do we English or French speakers really suffer compared to Spanish or Turkish speakers because our writing system is far more illogical? Recent economic and cultural history might suggest otherwise. Look at all the waste and nonsense around the German spelling reforms – whose life has improved because of it? China seems to be doing pretty well right now, so what's the issue?

And many of the comments on Victor's post itself expressed similar sentiments; thus Rivers4:

Gosh, how will China ever become the world's second largest economy with such a cumbersome an inefficient writing system?

Oh, wait…

In fact, you could make a case for crappy and hard-to-learn writing systems being positively associated with economic and cultural success. I'd be willing to wager a substantial sum that if you found a way to quantify unmotivated complexity and uneccessary difficulty of learning, the correlation with measures of economic and cultural success would be higher than (say) the correlations of phoneme-inventory size with number of speakers or distance from Africa.

There are some simple factors that will guarantee such a correlation: national languages that were recently reduced to writing tend to have historically shallow and rationally-designed orthographies; and the countries for which this is true tend to be relatively poor and underdeveloped, simply because otherwise their literacy traditions would have started many centuries earlier. In some cases, like Turkish, special circumstances permitted a recent and radical orthographic reform. Spanish seems simply to have lucked out, by having a fairly shallow and transparent system to start with, and then undergoing relatively few orthographically-opaque sound changes. But around the world, countries who came late to the table of literacy tend to have relatively transparent and easy-to-learn orthographies; and the same countries tend, for roughly the same reasons, to remain relatively undeveloped economically, to have relatively little cultural influence outside their borders, and so on.

You could go beyond these trivial historical associations, and make an argument that an unnecessarily complex and hard-to-learn writing system is genuinely and causally a Good Thing from a political and economic point of view.  According to this story, a crappy orthography — in a society where literacy matters — creates a meritocracy based on verbal aptitude and the willingness to work hard at difficult and arbitrary socially-prescribed tasks. Mastering the orthographic system is a necessary (and sometimes even sufficient) condition for economic success, and this tends to offer a path out of poverty to the bright and ambitious children of the masses, and  to create a handicap for the most lazy and stupid children of the elite. You could point to the Mandarin system in Tang-dynasty China, or the English "grammar schools" back in the days when they taught Latin and then a standardized form of English.

I'm skeptical that this argument remains valid, if it was ever valid to start with. For one thing, there are now many gatekeeper subjects that are more intrinsically useful, such as science, history, and math, (And of course we've levelled the global playing field by making it necessary for everyone to learn English, which has the third-worst orthography among major modern languages, after Japanese and Chinese.)

In this context, any possible benefits of a crappy writing system are massively outweighed by the very real social costs. Becoming genuinely literate in Chinese characters takes up a significant portion of primary and secondary education. Accurate statistics on functional literacy in Japan and China (as opposed to self-congratulatory propaganda based on spurious definitions of "literacy") seem not to be available; but the available information suggests that significant fractions of the population never reach the goal, or lose the ability even if they once gained it. (See "Japanese literacy: Back to the future again?", 12/25/2006, for a survey of some of the facts.)

Those who do reach the goal waste thousands of hours on the task. And in the case of Japan, the writing is one of several factors that will make it very difficult for the Japanese to deal with their aging labor force in the most obvious way, by immigration.

In the case of English, the social and individual costs of our crappy writing system (compounded by bad teaching methods) are well documented.  I'd summarize the studies I've read by estimating that nearly two thirds of American children find it difficult to learn to read, and at least 20-30% fall behind to the point that their ability to learn other subjects is harmed.

And in "Ghoti and choughs again", 8/16/2008, I pointed out that Philip H.K. Seymour, Mikko Aro, and Jane M. Erskine, "Foundation literacy acquisition in European orthographies", British Journal of Psychology, 94:143-174, 2003, found that reading error rates among first-grade school children are many times higher in the UK than in other European countries. In their study, only Danish had error rates even close to half as great, and the error rates for correctly decoding words in French, Greek, Italian, German, and Dutch were reported to be less than one-tenth as great as in English.

Seymour et al. suggest that this is because children learning to read and write English face a double problem: complex syllables and opaque letter-to-sound correspondences.

Whatever its causes, the handicap is well documented. It's true that we Americans (along with the British, the Japanese, and the Chinese) have collectively managed to overcome the handicap of our crappy writing system; but this is not evidence that the handicap has paradoxically done us good.

I should add that the Japanese, Chinese, and English writing systems do share the advantage of offering to those who have mastered them certain clues about morphology and etymology that would be missing in a shallower and more phonetically transparent system. The question is whether this is really worth the cost. And the answer, of course, is that question isn't worth answering — the chances of significant orthographic reform are indistinguishable from zero, at least in the case of English, and so the real question is whether or not we'll compound our curse by bad reading instruction.

Finally, there's a partly-separate issue about whether an orthography is lexically consistent, in the sense that the same word is always spelled the same way. This overlaps to some extent with the question of how much importance is given to morphology and etymology, but even a relatively shallow orthography can be more or less lexically consistent. Even in a relatively shallow orthography, however, lexical consistency is not easy for writers to achieve, because of pronunciation variation due to dialect, register, and the phase of the moon; because of the treatment of foreign loans, including foreign names; and because of the frequent cases where people are unclear about which words are which.

In any case, although it's plausible that lexical consistency is a virtue, I've never seen any evidence for this view — the catch-as-catch-can Elizabethan spelling system doesn't seem to have held Shakespeare back.

Share:



73 Comments »

  1. Theodore said,

    May 19, 2011 @ 7:16 am

    I have to question the assumption that elementary education is really about the things ostensibly taught and learned and suggest considering it more like bodybuilding: the weights are back where they were when we started, yet we're somehow "better" for having moved them around. English, Japanese and Chinese might just be heavier weights that somehow work the muscles needed for economic prosperity, or at least make us devote so much time to primary education that we develop that "willingness to work hard at difficult and arbitrary socially-prescribed tasks." I'd bet that willingness is far more important to individual (and maybe collective) economic prosperity than even science or math (and I'm an engineer).

  2. Bill Walderman said,

    May 19, 2011 @ 7:20 am

    "only Danish had error rates even close to half as great,"

    Having tried without complete success to learn Danish, I found this amusing. Danish seems to be almost as orthographically incoherent as English. Complex syllables and opaque letter-to-sound corresponences abound. And it's not just the orthography that poses a challenge to young children, if this is to be believed:

    http://cphpost.dk/classifieds/services/51613.html?task=view

  3. Are Odd Spelling Rules Good for the Economy? | Dr. Platypus said,

    May 19, 2011 @ 7:28 am

    [...] Liberman of Language Log asks, "Is a bad writing system a Good Thing?" He wonders whether there is a correlation between the difficulty of a language's [...]

  4. GeorgeW said,

    May 19, 2011 @ 7:42 am

    Arabic orthography is shallow. However, it represents a language that no one speaks. To write, one must learn a different lexicon and syntax.

    [(myl) Degree of diglossia is a separate matter, also highly relevant in China (where a large fraction of the population does not speak Putonghua natively) and for that matter in English.

    But in fact, Arabic orthography is also not so easy to learn to read -- lack of short vowels, solid prepositions and articles, hamza, etc. etc.]

  5. [links] Link salad feels even a bit more like its old self | jlake.com said,

    May 19, 2011 @ 7:55 am

    [...] Is a bad writing system a Good Thing? [...]

  6. Sam said,

    May 19, 2011 @ 7:57 am

    Finnish has a recent and historically shallow writing system but Finland has an advanced technological economy (e.g.., Nokia).

    [(myl) This is social science, man, where a correlation of 0.2 or .3 is considered a big win...]

  7. B.Ma said,

    May 19, 2011 @ 8:12 am

    "Ghoti" isn't very helpful. I'm sure you would agree that English spelling is actually quite regular and there are a number of rules which, once learned, will make it possible to pronounce almost any English word. So there are more rules than Spanish. Well, too bad!

    [(myl) If you still believe that, then you've never tried actually turning those "rules" into an algorithm that works. It's certainly true that there are many letter-to-sound patterns, which learners need to master implicitly or explicitly. But the result, independent of word-specific memorization, will be far from completely successful. Those who would like to try writing some letter-to-sound rules should take a look at this, or try dealing with the million-odd proper names to be found in the U.S.]

    You say that "Those who do reach the goal [of learning their language fully] waste thousands of hours on the task." Why do you think it's a waste of time to learn something of historical and cultural significance which still has applications in everyday life? How about the thousands of person-years spent filling in tax returns?

    http://www.youtube.com/watch?v=omg9rzf4GDM

    How about all the food you eat that is wasted by your cells copying meaningless defunct viruses in your DNA just to keep you alive, or the hours wasted in traffic jams because we can't bulldoze a city and rebuild it with more efficient transport?

    [(myl) And this is all relevant because...? If there are any inefficiencies in life, then all inefficiencies are thereby excused? We could insist that all schoolchildren do all arithmetic problems in Egyptian fractions -- they would thereby learn something of historical and cultural significance, as well as developing their mental muscles. But it would still be a stupid thing to do.]

  8. J.W. Brewer said,

    May 19, 2011 @ 8:22 am

    Can you point to a study that does an intra-European comparison like Seymour et al. for, say, teenagers? It seems rather significant to know whether it just takes longer for Anglophone kids to get over some sort of initial threshold around first grade versus whether and in what ways and to what degree they're still lagging way down the road.

    [(myl) I don't know any such studies, though they might well exist. The problem is that it's very difficult or even impossible to control for other relevant social and linguistic factors -- that was a problem in the first-graders study, as I observed in the post I linked to. But it's clear that in the U.S., at least, imperfect mastery of reading is a serious problem for many students through secondary school, and imperfect mastery of writing (even at the individual word level) is a not-insignificant problem for some students through the college level. How much of this would be fixed by a better writing system is not clear, but the point is moot anyhow, since (as I wrote) the chances of significant reform are effectively zero.]

    I am also skeptical that crappiness-of-orthography (or the less pejorative-sounding "deep" v. "shallow" contrast) is a single dimension that can be uncontroversially measured (which doesn't necessarily mean that English wouldn't rank poorly under any plausible metric). As the post hints at toward the end, an orthography is an attempt to satisfy multiple competing goals simultaneously, and it's hard to evaluate whether particular trade-offs are net positive or net negative without assigning contestable and subjective weights to the importance of different desiderata. For example, it's perfectly plausible to argue that the benefit of increased ability to disambiguate homophones (where English can borrow some of the pro-hanzi/kanji talking points) is not actually worth the associated costs, but coming up with an uncontroversial and objective quantification of the benefit versus the cost seems potentially difficult. I think bopofomo "fits" Mandarin better than pinyin does, but the benefits of using the world's leading character set versus a unique-to-your-language character set are obviously significant but difficult to quantify, so how does one say whether they are worth the costs of decreased fit? (In our own orthographic history, presumably there were benefits and costs back in the Anglo-Saxon days of using Latin script instead of the futhark runes that might perhaps have been a better phonological fit; there were also benefits and costs of the subsequent abandonment of edh/thorn/yogh/etc.)

    [(myl) I don't think that the shallow-vs.-deep distinction is quite what I mean -- I'm more interested in how easy or hard a system is to learn, once you know the spoken language to start with. A deep orthography can be very easy to learn; a shallow one can be hard.

    I certainly agree that there are several dimensions here, as discussed in detail in any of the texts on the subject. But as I noted in response to another comment, we're talking about social science here, and different ways of projecting orthographic systems onto a single "difficulty of learning" dimension are all going to give us the kind of weak correlations we'll be happy with.]

  9. John Lawler said,

    May 19, 2011 @ 8:40 am

    Both Chinese and Japanese have auxiliary phonetic scripts (Pinyin, bopomufu, kana, Rōmaji) which are used widely and helpfully. English has one, too; it's called the English Phonemic Alphabet and it's been widely available since the publication of Kenyon and Knott's 1953 A Pronouncing Dictionary of American English (which Merriam-Webster resolutely refuses to use in preference to Noah Webster's system in its other dictionaries).
    Here is a modest example of it; it can be picked up easily by any American English-speaking child. I always suggest that the sheet be given to some little girls before they learn to read much, telling them that this is a secret writing system that's hard for boys or most grownups to learn. Then you ask them to figure it out and explain it to you.

  10. George said,

    May 19, 2011 @ 8:49 am

    No expert by any means but it seems to me, where crappiness of orthography is concerned, that looking at rates of reading errors is only one side of the story. French, for example, is a doddle to read but quite difficult to write. My personal experience would also suggest that it is more difficult for native speakers of French than for native speakers of more or less related languages (including English) who have learned French as a foreign language to a good standard.

  11. Timothy Martin said,

    May 19, 2011 @ 9:03 am

    Accurate statistics on functional literacy in Japan and China (as opposed to self-congratulatory propaganda based on spurious definitions of "literacy") seem not to be available; but the available information suggests that significant fractions of the population never reach the goal, or lose the ability even if they once gained it.

    But the practice makes them so good at drawing! :D

    Also, a question. Are there not cognitive benefits of learning a complicated orthography (aside from "practice at doing arbitrary tasks")? As someone who's always been a good speller and reader, whatever my brain has learned to enable me to do these things must be a type of cognitive ability. I'm sure an amount of rote learning is involved, but also it seems that English has more and more complicated spelling patterns than other languages, and learning to recognize them is an ability that one must hone. If so, we should ask if this ability is transferable to other things, and how much benefit we get for the price we pay in dealing with English orthography.

  12. J.W. Brewer said,

    May 19, 2011 @ 9:40 am

    Prof. Lawler: Do you have another cheatsheet showing how your 11 vowels + 3 dipthongs are deployed to cover all 24 of the Wells lexical sets? Or is this script intended for a dialect with enough different mergers to get the total number down to 14? (I've never stopped to count exactly how many I'd need personally – I have trap/bath merger and horse/hoarse, but not cot/caught, for example.)

  13. Andrew Dowd said,

    May 19, 2011 @ 9:50 am

    If historically convoluted orthographic complexity as a barrier to literacy was causally implicated in economic success, Irish speakers would rule the world.

  14. George said,

    May 19, 2011 @ 9:59 am

    @Andrew Dowd. Irish orthography isn't actually very convoluted. In fact, it's consistency is well illustrated by Myles na Gopaleen's ability to write perfectly comprehensible English using it. (Perfectly comprehensible to anyone who knows the system, that is!)

  15. George said,

    May 19, 2011 @ 10:00 am

    Regular commenter Maidhc is another good example.

  16. George said,

    May 19, 2011 @ 10:05 am

    Actually, the name Dowd might suggest that you know all that already. I'll get me coat…

  17. Kyle Gorman said,

    May 19, 2011 @ 11:06 am

    I happen to have had the luxury of hearing Richard Sproat, and Eugene Buckley lecture elegantly on this subject.

    No discussion would be complete w/o a mention of the research on the Vai, or Hannas' "Writing on the wall" thesis (http://www.amazon.com/Writing-Wall-Orthography-Creativity-Encounters/dp/0812237110).

  18. Jean-Sébastien Girard said,

    May 19, 2011 @ 11:06 am

    "If you still believe that, then you've never tried actually turning those "rules" into an algorithm that works."

    Someone has: "What I hope to have shown, however, is that beneath all the pitfalls, there's a rather clever and fairly regular mechanism at work, and one which still gets the vast majority of words pretty much correct. It's not to modern tastes, but by no means as broken as people think. "

    [(myl) Many people have worked on this problem, including me. See e.g. Mark Liberman and Ken Church, "Text Analysis and Word Pronunciation in Text-to-Speech Synthesis," pp. 791-832 in Furui and Sondhi, Eds., Advances in Speech Technology, Marcel Dekker, 1992. Any system that has a decent chance of pronouncing a whole English sentence without errors includes lots of word-specific information. ]

  19. Laura said,

    May 19, 2011 @ 11:31 am

    This post reminded me of something I'd heard a few months ago about harder-to-read fonts attributing to how much learning is retained:

    http://www.bbc.co.uk/news/world-11573666

    Could there be a correlation between that and complex languages? Or do you think they're unrelated?

  20. Tidbits of Inspiration: Bad Writing System « The Undiscovered Author said,

    May 19, 2011 @ 11:42 am

    [...] Here's the article. [...]

  21. Chandra said,

    May 19, 2011 @ 11:59 am

    Mastering the orthographic system is a necessary (and sometimes even sufficient) condition for economic success, and this tends to offer a path out of poverty to the bright and ambitious children of the masses, and to create a handicap for the most lazy and stupid children of the elite.

    *wince*

    Whether you intended it or not, this paragraph perpetuates the myth that ability to master reading and writing correlates with intelligence and ambition. I work as a literacy program coordinator and fundamental English instructor to adults, and I can assure you that it very much does not.

    [(myl) Of course you're entirely right -- I've known several accomplished genius-level dyslexics. I expressed myself carelessly in laying a parody of a position that (you'll note) I prompted disagreed with; but my description had some presuppositions that were not effectively cancelled.]

  22. Nick Lamb said,

    May 19, 2011 @ 12:17 pm

    Chandra, you can try to assure us, but you will do much better to provide data.

  23. army1987 said,

    May 19, 2011 @ 12:18 pm

    @myl (in the comment to Jean-Sébastien Girard's post):

    Once, someone said that the English pronunciation is actually predictable from the spelling, but people don't know the rules well enough. I pointed him out an example such as bury, and he answered “That's perfectly regular.” I asked “How comes it doesn't rhyme with fury or jury?” and they answered Because the initial consonant is different.”

  24. army1987 said,

    May 19, 2011 @ 12:21 pm

    As for Irish orthography, not only it is way better than the English one, but I dare say that (at least after 1947) it's not even as bad as the French one.

  25. Daniel von Brighoff said,

    May 19, 2011 @ 1:13 pm

    English, which has the third-worst orthography among modern languages, after Japanese and Chinese

    Not even close. Let's revisit this claim after Seymour et al. have spent some time surveying South and Southeast Asia, shall we?

    Irish orthography isn't actually very convoluted. In fact, it's consistency is well illustrated by Myles na Gopaleen's ability to write perfectly comprehensible English using it.

    No to mention a previous language log discussion on the subject (http://languagelog.ldc.upenn.edu/nll/?p=499>).

  26. Riccardo said,

    May 19, 2011 @ 1:20 pm

    "In fact, you could make a case for crappy and hard-to-learn writing systems being positively associated with economic and cultural success"

    So… should a country bent on economic expansion adopt Egyptian hieroglyphs as a writing system?

  27. Chandra said,

    May 19, 2011 @ 1:34 pm

    @Nick Lamb: A quick scan of any reputable book, article or website dealing with the subject of dyslexia and other learning disabilities will tell you quite explicitly that intelligence is not a factor. In fact, scoring in the average-or-above range on a test of global intellectual functioning is one of the required criteria in establishing a clinical diagnosis of learning disability.

  28. cameron said,

    May 19, 2011 @ 3:00 pm

    There's no way English has "the third-worst orthography among modern languages". All of the non-Semitic languages that use variants of the Arabic script, most prominently Persian and Urdu, are far worse than English in that regard.

  29. Bill Walderman said,

    May 19, 2011 @ 3:43 pm

    "should a country bent on economic expansion adopt Egyptian hieroglyphs as a writing system?"

    Linear B would be an even better choice.

  30. Rubrick said,

    May 19, 2011 @ 4:38 pm

    @Chandra: "In fact, scoring in the average-or-above range on a test of global intellectual functioning is one of the required criteria in establishing a clinical diagnosis of learning disability."

    Surely that's because such a diagnosis requires a downward deviation from an average-or-above baseline, not because those with below-average intelligence never have difficulty learning!

  31. John Cowan said,

    May 19, 2011 @ 4:41 pm

    We should just all switch to Blissymbols, which aren't even related to a particular language. The entire planet would at once go into an orgy of development, with the total elimination of poverty to be expected within the decade.

    myl: Technically, nobody speaks Putonghua natively; it's a modified version of an older version of Beijing dialect with accommodations to other dialects (the same might be said of RP English, reading London for Beijing). But about 70% of the Chinese population speaks some variety of Mandarin natively, which is a pretty good figure for the sole official language, as countries go.

  32. Lugubert said,

    May 19, 2011 @ 4:56 pm

    cameron "All of the non-Semitic languages that use variants of the Arabic script, most prominently Persian and Urdu, are far worse than English in that regard."

    Agreed, and how many ways is Thai [t] written? But my personal favourite is Tibetan. What rules would help the reader realize that Dbus-gtsang is pronounced Ü-Tsang?

  33. Ken Brown said,

    May 19, 2011 @ 5:52 pm

    I went through a typical English education that failed to teach me French. Its hard for me to accept that French spelling is really that much more transparent or consistent than ours!

    @Riccardo: heiroglyphs were probably quite easy for anyone who actually spoke early Egyptian. Simpler to get started on than the alphabet (of which it is both a superset and a precursor). The problem for us is that no-one speaks early Egyptian any more. The problem for the Egyptians is that they used the same system through ovet two millenia of language change.

  34. sprachnomade said,

    May 19, 2011 @ 6:34 pm

    My personal impression is that orthographies that do not correspond well to pronunciation tend to have other merits. Take the notorious Thai and Tibetan orthographies that were already mentioned. For me it is clearly a plus that loanwords (and that includes the many borrowings from Pali and/or Sanskrit) are written in original orthography, although pronunciation has moved on to a point where you wouldn't even know a word was once borrowed without the clues given by orthography.

    This kind of conservative orthography can maintain cultural links that would otherwise be broken. As difficult as the orthography of, say, Urdu, might seem to a learner without any background in other languages of the region, for someone who knows some Persian or Arabic, the conservative orthography for loanwords provides many entry points.

    And this doesn't only apply to the academic learner. I consider it a vast advantage that the word "information" is still written in pretty much the same way in virtually every European country, making it a lot easier for passengers to make their way through an airport than if it were spelled, say, ĩformasjõ, infirmayshn, informazjohn, etc. in different countries.

  35. GeorgeW said,

    May 19, 2011 @ 7:27 pm

    sprachnomade: When I was in Pakistan a number of years ago, I discovered there are many Arabic and English loans in Urdu. And, the script was close enough to Arabic that I could make relatively good sense of signs and the like and even figure out what some newspaper articles were about.

  36. Chandra said,

    May 19, 2011 @ 8:29 pm

    @Rubrick: Where did I state that those with below-average intelligence never have difficulty learning?

    The assumption that people make is that those who can't learn to read and write easily are stupid. Although certainly there will be cases where general cognitive impairment impacts learning ability, for the vastly largest percentage of native speakers in developed countries who haven't mastered literacy skills by adulthood, the cause will be some kind of specific LD. And no, I don't have exact figures and percentages on hand in front of me, but this is common knowledge in my field and there are studies to support it. I interact with a large network of other literacy service providers in my part of the country and I guarantee their personal experience with students in the classroom would bear this out just as mine does.

    The laziness argument is almost even more cringe-worthy. Does anyone really believe that those who can't read and write well are coasting easily through life?

  37. J. Goard said,

    May 19, 2011 @ 10:26 pm

    It's remarkable to me how most people seem to focus on the directness of mapping between form and sound/sense, while entirely neglecting the issue of how well the script exploits the visual space in order to facilitate fast and accurate visual perception. Korean is my main second language, and I'm also a beginning student of Vietnamese. The former's use of tiny lines crammed into syllable blocks (and often parallel to other tiny lines) to differentiate minimal pairs, finds a way to frustrate me on a nearly daily basis, whenever I want to read difficult texts in less than 14-pt font and/or perfect lighting. (And I'm only 34.) The latter's use of many tiny, similar diacritics for tone and vowel quality will no doubt similarly frustrate me once I get beyond big fonts and kindergarten vocabulary.

    To forestall the obvious response: yes, of course literate Koreans can read small hangeul just fine, unless their eyes are much worse than mine. This by no means shows that they don't suffer from visual processing deficiencies resulting from the script; rather, it illustrates the role of top-down processing, which is fundamental to literacy in any script.

    My larger point here is that the right way to evaluate a script is to consider the various psychological factors involved in reading under various real-world conditions, not merely learning.

  38. John Walden said,

    May 20, 2011 @ 1:17 am

    Do eccentric spelling systems lead to more puns and word-play?

    My gut-feeling is that they may well.

    Is this a large part of the "verbal aptitude" which leads to a rich culture and perhaps economic success?

    My gut-reaction says that it doesn't know but rather hopes so.

  39. YankeeTranslator said,

    May 20, 2011 @ 3:13 am

    Here is my argument in favor of keeping English writing as is: for those of us interested in historical evolution and etymology, it much more effectively preserves such information than if we were to change it.

    Also, I disagree with Dr. Liberman's critique of Arabic orthography as hard: there is a distinct advantage to not writing the short vowels. Between the dialects and the classical language, the basic consonantal structure is often the same, however, the vowels slightly change, so that even educated Arabs often confuse the voweling when trying to speak fusha. (For instance, even educated Arabs will often say tajruba instead of the classical tajriba (experience)). But since you don't need to write the vowels, this doesn't matter.

  40. Nick Lamb said,

    May 20, 2011 @ 3:48 am

    Chandra, ok, so you don't have any data

    The existence of a group of outliers is not proof that a trend doesn't exist in the general population. If you want to establish that it doesn't exist you will need more than "here is this group of people who, if they were considered on their own defy the correlation". A team of women's basketball players could, if we accepted your shoddy methods, prove that human height isn't correlated to sex

    You wrote "The assumption that people make is that those who can't learn to read and write easily are stupid"

    Nobody on this blog made that assumption, you're the one who introduced it. If "people make" this assumption you should take your complaints to their blog, not clutter this one.

  41. maidhc said,

    May 20, 2011 @ 4:25 am

    John Cowan: I've long felt that Blissymbols were unfairly neglected. There is after all a growing collection of international hieroglyphics having to do with mostly traffic management and finding your way around airports, among other topics. Why not extend the system?

    I don't know about Bliss's knowledge of Chinese, but some aspects of the system seem to relate to what little I know about Vietnamese. The idea is that a symbol represents a conceptual area. If you need it make it more specific, you add more symbols. Like in Vietnamese the word "puppy" would be "animal dog offspring". If you don't need to be so specific you can omit some of the words.

    If we regularized English spelling, we could use the time saved to teach children Blissymbols.

    I was thinking about commenting on the Hat, but I decided I was too outclassed. But maybe here someone can enlighten me.

    Both Italian and Spanish, renowned for their spelling regularity, have "ps" at the beginning of their word for "psychology". But they have "s" for their word for "psychosis". How so?

  42. minus273 said,

    May 20, 2011 @ 4:33 am

    @What rules would help the reader realize that Dbus-gtsang is pronounced Ü-Tsang?
    It's very simple. With a minuscule amount of training, you will know that "db" is pronounced the same as the letter "w", "s" makes a front version of the preceding vowel (like a Finnish umlaut) with a falling tone, and "g" before a consonant doesn't count.

  43. army1987 said,

    May 20, 2011 @ 5:57 am

    @J. Goard:
    Maybe that's true when you're learning to read or when reading unfamiliar words, but I think quickly skimming through a text about a familiar topic in a familiar language takes shorter (page size and margins and font size being equal) if it's two pages than if it's three. (Which is why I'd prefer if Irish orthography still used the dot-above rather than the H for lenited consonants.)

    @minus273:
    Italian does have a P in psicosi! (And if Spanish doesn't, I strongly suspect it's because the pronunciation of the word for ‘psychosis’ doesn't.)

  44. John F said,

    May 20, 2011 @ 6:47 am

    J.Goard's comment reminded me that most English readers only need to see the shape of the word to recognise it, i.e. the first and last letters and the approximate length, so most spelling errors inside the word are recoverable.

  45. Daniel von Brighoff said,

    May 20, 2011 @ 9:51 am

    @J. Goard: Though it's less of a consideration in the computer age, my Korean friends used to complain that their fingers hurt more after writing a page of Korean than a page of English because it's a less flowing script.

    Just a couple days ago, I was talking with some friends about the relative difficulty of reading certain scripts at the same font size. I don't see how anyone can read 10-point Thai with the same ease as 10-point Latin; they singled out Perso-Arabic and traditional Chinese. (The more complicated characters can turn into almost solid blocks of ink at smaller reproduction sizes.)

  46. Dominik Lukes said,

    May 20, 2011 @ 9:56 am

    This also depends on the largely untested assumption that there's a larger societal benefit to "full" literacy. Historically, literacy levels have always been measured by methods (like signatures in marriage registries or later postcard sending) that do not indicate the actual level of facility with reading and writing. The modern notion of 'functional literacy' is not something that can be compared with that kind of literacy. But more importantly the current Western political vision of literacy is not supported by good ethnographic understanding (looking forward to: http://www.amazon.co.uk/Language-Ethnography-Education-Bridging-Literacy/dp/0415872499).

    I doubt that there is is any economic benefit for an industrial society in pushing for literacy over 90% which only means you create a lot of crypto-illiterate people. Illiterates in highly literate environments are extremely good at devising strategies to hide their issue and do without (the same goes for numeracy).

    I remember reading a book on the downsides of the Japanese writing system as a form of corporate communication. But it doesn't seem to have slowed the Japanese corporations down. I spent a lot of the 90s dealing with Cyrillic in email and e-documents and I saw a huge amount of compensatory strategies develop until technology caught up. I know it's not quite the same but the principle remains. Knowing what people do with their literacy (whatever it may be) is more important than knowing the underlying principles of their writing system or the education system.

  47. ronbo said,

    May 20, 2011 @ 10:17 am

    I hadn't really realized how nasty French orthography can be until I started helping my teenaged son, who is trying to fit in some French between Spanish and Arabic.

    I felt like I was quoting "Chinatown" half the time: "Because it's French". Still, I haven't taken him up on his offer to teach me Arabic in exchange, and I don't think I will.

    BTW, Le Monde has an occasional usage blog that is often very interesting. Recently it covered the "correct" pronunciation of place names in France. You can't imagine the smile I was wearing all day over the difficulties native speakers were having with the place names I had mangled in the past!

  48. army1987 said,

    May 20, 2011 @ 10:22 am

    @Daniel von Brighoff:
    But then a text which needs about 10000 characters in English will only need about 3000 in Chinese, so you can use a larger font to fit the same text in the same amount of space.

  49. R. Wright said,

    May 20, 2011 @ 11:03 am

    I'd be willing to wager a substantial sum that if you found a way to quantify unmotivated complexity and uneccessary difficulty of learning, the correlation with measures of economic and cultural success would be high…

    As a math and statistics teacher with a deep interest in language, I agree and would love to see this demonstrated.

    Timothy Martin said,
    Are there not cognitive benefits of learning a complicated orthography (aside from "practice at doing arbitrary tasks")?

    And Laura said,
    This post reminded me of something I'd heard a few months ago about harder-to-read fonts attributing to how much learning is retained…

    I actually came into the comments to bring this up. I noticed in my youth that nations with difficult writing systems seemed to have more "smart people" (remember that I was just a youth) than those with easier systems. My assumption has always been that a difficult writing system provided a source of deep cognitive training. After reading the paper Laura's article references, I am somewhat more convinced. Incidentally, the paper itself is available here.

  50. Chandra said,

    May 20, 2011 @ 11:14 am

    @Nick Lamb: Excuse me, I responded directly to a paragraph in this blog post, and to comments directed at me, with perfectly valid points. You are the one now "cluttering up" this blog with insulting remarks, and a lack of evidence to back up the counter-argument you seem to be trying to defend. I have nothing more to say to someone who chooses to express their dissent in this way; feel free to have the petty last word, as I'm sure you will.

    [(myl) Bad feelings aside, this discussion aises an important point.

    Reading researchers know very well that there's an overall correlation between learning to read and other sorts of learning. As a result, simply looking at "good readers" and "poor readers", as measured by some test instrument, is not a good way to get at the causes of reading-specific disabilities, since a (perhaps substantial) portion of the reading performance will be due to more general differences in skills and attitudes. Instead, it's necessary to compare samples with high or low reading scores who are also equated in other ways. This issue also plays a role in remediation practice, and there seems to be some controversy about how cleanly people with reading disabilities separate into what are called "IQ discrepant" and "IQ consistent" sets. (See for example Karla Stuebing et al., "Validity of IQ-Discrepancy Classifications of Reading Disabilities: A Meta-Analysis", American Educational Research Journal 39(2) 2002.)

    But it would be a terrible mistake to conclude from this that anyone with reading difficulties will necessarily score poorly on tests of other skills, or lacks motivation, or whatever.

    People seem to have a remarkably hard time keeping straight the difference between distributions and individuals.]

  51. Norwegian Guy said,

    May 20, 2011 @ 11:20 am

    "I consider it a vast advantage that the word "information" is still written in pretty much the same way in virtually every European country"

    A quick look at Wikipedia makes me doubt that this is true. As far as I can see, the only European languages that use the spelling 'information' is German, French, English, Swedish and Danish.

  52. J. W. Brewer said,

    May 20, 2011 @ 12:26 pm

    myl's interjection to Chandra's last comment made me think it would be interesting if there were comparable data for students with reading difficulties in the schools of Spain or Italy (pick your own country – the idea is one with at least rough economic-status similarities to the US but with a language widely thought to have a more easily-learnable orthography). If the ratio of IQ-discrepent to IQ-consistent (accepting that it's not necessarily a clean, binary division) within the poor readers varied in a way that seemed to relate to the comparative difficulty of the orthography, that seems like it might tell us something interesting.

  53. RF said,

    May 20, 2011 @ 2:03 pm

    China has a large economy simply because it has a lot of people. On a per capita basis, it's not even close to the developed world.

    In the phrase "to deal with their aging labor force in the most obvious way, by immigration", "by" was a poor choice of preposition, as it suggests that the aging labor force will simply be shipped to another country.

    "Arise" is not a transitive verb.

    In your sentence "But it would be a terrible mistake to conclude from this that anyone with reading difficulties will necessarily score poorly on tests of other skills, or lacks motivation, or whatever", you should have written "everyone", rather than "anyone".

  54. RF said,

    May 20, 2011 @ 2:06 pm

    Chandra said,
    "A quick scan of any reputable book, article or website dealing with the subject of dyslexia and other learning disabilities will tell you quite explicitly that intelligence is not a factor."
    Of course not. Because, as YOU ACKNOWLEDGE, the term "dyslexia" is DEFINED as "a reading/writing problem not resulting from low intelligence".

    "In fact, scoring in the average-or-above range on a test of global intellectual functioning is one of the required criteria in establishing a clinical diagnosis of learning disability."
    If you really understood what you are saying, you would understand that that actually CONFLICTS with your position. The very fact that IQ discrepant students must be separated from IQ consistent students shows that the latter exist, just as the fact that separate evaluation methods for students not having English as their native language reflects the fact that having English as a native language correlates with English mastery.

    "@Rubrick: Where did I state that those with below-average intelligence never have difficulty learning?
    The assumption that people make is that those who can't learn to read and write easily are stupid."
    No, your claim was "Whether you intended it or not, this paragraph perpetuates the myth that ability to master reading and writing correlates with intelligence and ambition." Saying that there is a CORRELATION between intelleigence and reading ability is COMPLETELY DIFFERENT from saying "those who can't learn to read and write easily are stupid". There's also a correlation between being white and reading ability. Am I saying that black people can't read? Of course not.

    "Excuse me, I responded directly to a paragraph in this blog post, and to comments directed at me, with perfectly valid points."
    No, you did not respond with "perfectly valid points". You have responded with assertions that there are studies that support your position while refusing to provide such studies, and then switched to a completely different position, either from a deliberate attempt to obfuscate the issue or due to you simply having no understanding of what the word "correlation" means. You, either through dishonesty or a lack of clarity of thought, claimed that you were arguing against a position when no one here has taken that position. When Nick Lamb pointed that out, you accused him of "insulting" you, and declared that you would not engage in any further discussion. If anyone should be insulted, it should be Nick Lamb and Rubrick, for having you imply that they hold a position they never expressed. Perhaps instead of stomping off in a huff when someone points out that you are wasting their time with nonsense, you should at least entertain the possibility that you don't know what you're talking about.

  55. John Cowan said,

    May 20, 2011 @ 3:19 pm

    Yankee Translator: I should rather say that because short vowels aren't written, even educated Arabs are vague on what the modern standard Arabic verb is.

    John F.: That idea (that the first and last letters are all you need) has been repeatedly debunked here on Language Log and elsewhere.

  56. GeorgeW said,

    May 20, 2011 @ 8:00 pm

    @John Cowan: "I should rather say that because short vowels aren't written, even educated Arabs are vague on what the modern standard Arabic verb is."

    Short vowels are written with diacritics and are used in some contexts (principally religious material and children's books). However, in many contexts they are superfluous and only used when there is ambiguity. I don't think this is an obstacle to literacy.

  57. YankeeTranslator said,

    May 20, 2011 @ 10:59 pm

    @John Cowan: The only types of verbs where the vowelization is variable is for present-tense class-one verbs. Of course Arabs often mix-up these vowels when speaking Fusha, but this does not in any way alter the intelligibility of the word or their ability to understand it on the page. Again, the fact that the writing system exempts them from having to really know these nuanced differences is actually a huge relief for them.

  58. a George said,

    May 21, 2011 @ 7:29 am

    I know that life has moved on, and that nobody will read comments on old posts. They only come up in searches. So, only for the record, there are quite a number of English words that are extremely similar yet have vastly different meanings and even pronunciations. An example is ‘ally’ and ‘alliance’ – a very fundamental concept that ensures your survival in a hostile world. However, all to no avail, if your allies dally and display ‘dalliance’. I tend to agree with John Walden who stressed the opportunities for word-play.

    Like those who master the various characters of the Far East along with having a europeanised mother tongue must feel some pride caused by intellectual superiority, so I too feel a pride in recognizing and knowing the deviations from fixed rules made available to us who know English and the languages that have inspired it. It is for us that the term Shavian was created ……… [who would have guessed it concerned a bearded man?].

  59. C said,

    May 21, 2011 @ 8:12 am

    "Accurate statistics on functional literacy in Japan and China (as opposed to self-congratulatory propaganda based on spurious definitions of "literacy") seem not to be available; but the available information suggests that significant fractions of the population never reach the goal, or lose the ability even if they once gained it."

    As anyone who has recently attended an American high school knows, functional literacy in the United States is much lower than 99%, and only decreases after people leave school. Anyone who doubts that Japan has a high literacy rate only needs to look at how popular reading is there in comparison. Does America have serialized novels for teens appearing in popular monthly or semi-monthly literary magazines?

    "Functional literacy" mostly results from low vocabulary levels, a poor understanding of grammatical functions, and a complete lack of interest in reading. None of those factors have anything to do with the writing system in use. High-level vocabulary and technical terms are actually easier to learn in languages that use Chinese characters, since you don't have to learn to recognize a large number of Latin and Greek root words.

    Especially in Japan, where most published material uses furigana to indicate the proper reading of any character outside the ones learned at school, I have a hard time believing that characters are keeping anyone from becoming more literate. As someone mentioned in another thread, after moving away from the use of Chinese characters, South Korea has problems with young people not understanding the etymology of Chinese loanwords. That is exactly the sort of thing that can increase rates of functional illiteracy.

  60. C said,

    May 21, 2011 @ 8:31 am

    Oops, I mean "functional illiteracy" in the third paragraph.

  61. David J. Littleboy said,

    May 21, 2011 @ 10:49 am

    Certainly the impression one gets living in Japan is that the Japanese are far more involved with their language than generic Americans are, and thus it feels like a far "more literate" space. But a lot of the writing is dreck (even the famous authors, e.g. Kawabata and Mishima, cranked out stuff that went a lot further downhill for being high-volume serializations than Trollope or Dickens ever did). Still, I find it hard to believe there's _any_ functional illiteracy here. Having been a nerd all my life, I've finally taken up a sport (bowling) that's rather working class, and no one has any trouble whatsoever with the language.

  62. army1987 said,

    May 21, 2011 @ 11:06 am

    @RF
    "By immigration" in that context means what it's supposed to; your meaning would be "by emigration". And "arising" is likely a trivial typo for "raising".

  63. Matthew Stephen Stuckwisch said,

    May 21, 2011 @ 6:30 pm

    maidhc: the ps- initial consonant can always be written either ps- or s-, bother are perfectly accepted by the Royal Spanish Academy. From the Diccionario Panhispánico de Dudas, under the entry "p":

    3. El grupo consonántico ps, resultado de la transcripción de la letra griega psi, aparece en posición inicial de palabra en numerosas voces cultas formadas sobre raíces o palabras griegas que comienzan por esa letra (psyché ‘alma’, pseudo- ‘falso’, psittakós ‘papagayo’, etc.). En todos los casos se admite en la escritura la reducción del grupo ps- a s-, grafía que refleja mejor la pronunciación normal de las palabras que contienen este grupo inicial, en las que la p- no suele articularse: sicología, sicosis, siquiatra, sitacismo, seudoprofeta, etc. No obstante, el uso culto sigue prefiriendo las grafías con ps-: psicología, psicosis, psiquiatra, psitacismo, pseudoprofeta, etc., salvo en las palabras seudónimo y seudópodo, que se escriben normalmente sin p-.
    3. The consonant group ps, caused by the transcription of the Greek letter Psi, appears at the beginning of a many educated words formed by Greek words or roots that begin with the letter (psyché 'soul', psuedo- 'false', psittakós 'parrots', etc). In all of these cases, a reduction from the group ps- to -s is permissible in writing, since the latter better reflects the normal pronounciation of the words that have the initial grouping, when the p- is not normally articulated: sicología, sicosis, siquiatra, sitacismo, seudoprofeta, etc. However, educated use still prefers writing with ps-: psicología, psicosis, psiquiatra, psitacismo, psuedoprofeta, etc., except for the words seudónimo and seudópodo, which are normally written without p-.

    The same goes for Pt, except that the P is more commonly not written.

  64. Peter T said,

    May 22, 2011 @ 3:46 am

    Is the short form of the argument that if everything were very easy, everybody would would be able to do everything better?

    [(myl) No. The basic argument is that if the time required to learn basic word-reading and word-writing skills is reduced from (say) 250 weeks to 5 weeks, schoolchildren and their teachers gain 245 weeks to study other things.]

  65. Ben Hemmens said,

    May 22, 2011 @ 12:14 pm

    German speakers may have an easier time spelling words correctly.

    The downside is that a good many of them seem to think there isn't anything else to language.

  66. C said,

    May 22, 2011 @ 6:38 pm

    "No. The basic argument is that if the time required to learn basic word-reading and word-writing skills is reduced from (say) 250 weeks to 5 weeks, schoolchildren and their teachers gain 245 weeks to study other things."

    Do schools really devote that much class time to teaching characters? Most of the work kids have to do to memorize them comes out of their free time, so I can say exactly what they would be doing if they didn't have to learn to write this way: watching more television and playing more video games. We have all seen how well that's worked out for America's youth.

    If the kids in China, Taiwan, and Japan were falling behind in their other school subjects because they found the task of learning how to read and write too difficult, then that argument would make more sense. Most kids learn characters before they are taught in school, either from their parents or from reading on their own. I don't know how it works for Chinese kids, but Japanese kids can and do easily teach themselves to read many kanji just by reading manga with furigana. They do have to spend a significant amount of time practicing writing in order to develop the muscle memory necessary to write quickly, but again, I see absolutely no evidence that the time they spend on this would be spent any more productively if they had a simpler writing system.

    By the way, schools could teach the kids all the characters in the first year if they wanted to. The problem is that it makes no sense to learn a character that isn't used in any of the vocabulary you know at that age, so the characters are spread out to be learned gradually as the children expand their vocabularies. That's why simple characters that happen to have complicated meanings are often taught years after complicated characters with simple, basic meanings.

  67. C said,

    May 22, 2011 @ 6:48 pm

    Also, I think that it's important to make the distinction between a system that is easy to learn and one that is easy and quick for someone who has already learned how to read. Chinese and Japanese are both much easier to read when written with hanzi/kanji, and I think that makes up for the learning process being longer.

    [(myl) There's no basis for making such a judgment, since no one has ever put in the kind of practice required to acquire skilled reading ability in an alternative system for those languages.

    Similarly, reading English in a more transparent orthography is much harder -- for readers skilled in the current system -- than reading the way they're used to. Similarly, I can read English transcoded into Cyrillic or Greek or Hebrew characters, but it's really slow and difficult. It would take a lot of practice -- maybe years of practice -- before I acquired the same skill in such a system, even though the information content is identical.

    As a result, arguments of this kind have no force at all, in my opinion.]

  68. Rob said,

    May 22, 2011 @ 8:39 pm

    I work in a Japanese High School, and was having a discussion with one of my colleagues about school sports back in the UK (such as they are). I mentioned that the playing fields (again, such as they are) were all grass, as opposed to the mixture of sand, grit, and broken glass seemingly favoured over here.

    "But don't you have to pay someone to cut the grass?" She asked.

    "Yes, but we all know the alphabet by the end of elementary school, so we don't have to pay calligraphy teachers."

    I also occasionally get interrupted during my study by my less diplomatic colleagues, remarking that, "Kanji are difficult for foreigners." To which my reply is always "Yes, but they're difficult for everyone. It's just you got to study them for a decade or so in school." This isn't always well received.

    I offer these up as illustrative anecdotes, though I'm not exactly sure what they illustrate…

    And a final, tentative (given the way the conversation is going) query. Can anyone point me the way of anything regarding dyslexia in Japanese speakers? I'm really curious to know how it manifests, but haven't really been able to get any satisfactory details by asking over here.

  69. C said,

    May 23, 2011 @ 10:56 am

    @Rob
    I already mentioned that the only reason they spend so long learning the characters is because it makes no sense for the school curriculum to teach characters that children don't have the vocabulary for yet. You may have known the alphabet a few weeks into elementary school, but you certainly did not have the vocabulary of a high school student. Kanji isn't an alphabet, no matter how many times I've heard it referred to as "the hardest of Japan's THREE alphabets." Ugh.

    And besides, most of the actual instruction, especially in higher grades, is related to writing them properly or in a beautiful style, because the kids can read even if they've never taken a calligraphy class. That might be why your "You got to study them for a decade" quip isn't always well received.

    About dyslexia in Japanese, I remember reading an article about a child who was raised bilingually in English and Japanese, and he had dyslexia, but only when reading English. I recall that some researchers concluded that this meant that whatever causes dyslexia in English is different from what causes dyslexia in Japanese, or that maybe Japanese is easier for dyslexics to read since it is written in syllables and you don't have to sound out individual consonant and vowel sounds when learning to read. At any rate it isn't as common as it is in English, so I'm not surprised you haven't been able to get much information from asking people over there.

  70. Rob said,

    May 23, 2011 @ 6:33 pm

    @C
    Thanks for the reply. It probably wasn't a good idea to come over all faux-naif in my first post, as I'm completely sure what (I meant) those anecdotes to illustrate. The first was to make a point about how the education system here prioritises teaching writing at the expense (both in money and time) of other subjects. Shodo classes don't take up a huge amount of the curriculum, it's true, but it's still time which could be used for studying something else. I (genuinely) don't know if it's through necessity or not, though I suspect not, which prompted the second point.

    In all honesty, the reason that quip isn't always well received is that most people are considerate and/or aware enough not to say to my (or another foreigner's) face what amounts to 'foreigners are stupid'. Those that do usually display fairly strong nihonjinron tendencies in other areas as well, and my reply is a slightly passive-aggressive way of pointing this out. Probably not big or clever, I know, but the point I was making here was about the (misplaced?) pride and prestige which can be attached to the mastery of skills which are culturally and socially valued, but practically unnecessary. In language or any other field. Which is of a part with spending (unnecessary?) time learning to write, however beautifully. And I've used too many (brackets).

    None of this negates your points, of course. And the information on dyslexia is very welcome, and what I was really asking about. I remember reading somewhere that native English speaking dyslexics found kanji easier to learn than other new languages' writing systems, but can't remember where. Does anyone have anything concrete on dyslexia and languages which use ideograms (and I know they're not really…)?

  71. Jen said,

    May 23, 2011 @ 9:59 pm

    I'd just like to point out that myl wrote "aises" instead of "raises", RH thought myl meant "arises" as a transitive verb and John Cowan then pointed out that the idea of first and last letters being all you need has been debunked previously on here and elsewhere–

    It was also just illustrated.

  72. Apollo Wu said,

    May 25, 2011 @ 2:04 am

    The correlation between language difficulty and economic strength is an interesting but hard to measure. Language is a communication tool with numerous applications in many field. As a productive force, language contributes to the development of national power. Hence, we should setup independent criteria to compare different languages and should bear in mind that their demerits and merits often are linked together. In general, we should consider the learning cost as well as the usage value. As far as learning goes, it is absurd to assert that the more difficult the better. From the point of view of mental gymnastic, one can choose from many more interesting activities. Because of the need to concentrate in Hanzi learning, most Chinese has narrower scope of knowledge and experience. The lack of ethnic dances among the Han people in China seems to illustrate this situation. Chinese parents often strongly discouraged their children to spend time in dances and plays, as they are considered as distractions to their children’s learning efforts. However, learning character scripts by rote makes one feel helpless and unhappy and maybe detrimental to creativity, as language learning is not a creative process, but a cultural replication process.
    Even more important is the utility of the writing system. Non-alphabetic Chinese and other alphabetic languages are analogous to cold weapon and hot weapon systems. Alphabetic languages simply have more fire-powers. Alphabet is a closed system suitable for machanical manipulations and computer processing. Alphabet can be arranged in sequential order. Non-alphabetic Hanzi system requires alphanumeric encoding (e.g. Pinyin), and it poor phonetic information leads to difficulty in verbal communication. Chinese routinely have to spend a lot of time to find what they want to get. The lack of order in Chinese script contributes to the chaos in daily live and the low per-capita production. Historically, Chinese society did have a early start in early age, because the lack of printing press minimized writing system disparity. All ancient great Chinese inventions were visual in nature, with no symbolic content. Since the Renaissance, the European societies benefited greatly from the use of the three great symbol systems, namely the alphabetic, numeric and the musical systems, which leads to great advances in all fields. On the other hand, Chinese society stagnated for more than a thousand years, and only recently developed under western influences. Chinese should find out why America only need two hundred years to develop from a primitive country into a highly developed superpower.

  73. Bob said,

    June 6, 2012 @ 2:01 pm

    I live in a country that could be considered a case-study in the switch from a writing system with spotty correspondence to pronunciation, to a nearly phonetic one: Turkey. Much is said about the problems with the Ottoman script and they're valid of course, but to hold the language reform up as the reason for increased literacy in Turkey is pure propaganda. The real reason for that change is the unprecedented availability of education to people who had no access to it before. (China arguably has one of the most complex writing systems in the world but nearly 100% literacy after all, and for the same reasons.)

    The reason for the change was much more political in nature, designed to create identification with the west. This is not without precedent; witness the significance of Latin and Cyrillic script in Serbian and Croatian, which are essentially the same language. Arabic script is more than a "writing system," it also has deep religious significance to Muslims.

    The flip side of the story is that even though the modern Turkish writing system is nearly phonetic people still managed to write it incorrectly; even educated people make basic orthographic errors (Turkish counterparts to "I should of gone", or "Buy me some apple's").
    Dyslexic people aside, those who have access to education, who read, and consider good writing important will learn to spell; those who don't care about it, won't, no matter how "difficult" or "easy" the writing system.

    One could even make an argument that, as we tend to read more in "chunks" and context rather than individual letters, there is some merit to a system that cuts down on the amount of letters necessary to render a word, especially in languages like Turkish where one "word" may correspond to many English words and one letter may make all the difference. A word like "işitemediklerimizden" (from the things that we couldn't hear) differs from "işitemediklerinizden" (from the things that you couldn't hear) by just one letter; the difference would be easier to spot in the Ottoman rendition. Also because all the vowels after the root "işit-" are regular and predictable, why bother writing them all when you can infer them? Some basic reforms could have handled the biggest problem (the irregularity of actual Turkish words as opposed to those of Arabic and Persian origin).

    In any case, we now have a nation of people who are cut off from their own literary past, unable to read anything written before 1927, unless it has been chosen for transliteration; even letters written by their own grandparents. A powerful tool in the hands of social engineers, and many would say that it was worth it. I'll withhold judgment on that particular subject.

RSS feed for comments on this post · TrackBack URI

Leave a Comment