Bad Character: If Chinese Were Phonetic
newyorker.com
newyorker.com
Edit: Apparently, this is part of a series where guest authors are invited to choose something to "uninvent" and explains why they think it has had a negative impact on the world. The essay's abrupt beginning makes much more sense in that context.
For more of the series.
The one I actually agreed with was the high heels one. Both my current and previous girlfriends insist on wearing heels most of the time, to the detriment of their health. My previous girlfriend could essentially not run anymore because of it, and the current girlfriend has chronic foot pain. I've banned (which is weird, I don't like saying that versus "asked", but I gave up on asking) her from wearing them on occasion because she needed to rest her feet/legs. Lasts about a week and then she's back in them. She's even had to go in for PT for a few weeks out of each year because of it.
My ex had grown so used to them that she pretty much always walked on tiptoes (like a kid toe-walking). She found flats and similar shoes very uncomfortable to wear anymore.
A podiatry lecturer that I briefly worked with used to give a superb short illustrated lecture on the importance of your heel and its role in walking and in maintaining blood circulation in the leg. He had side swipes at high-heels in the process. Perhaps there is stuff online?
"Come on, just dance for a little bit, you'll like it!"
"Rollercoasters are fun, trust me, this one is great!"
"Why won't you come eat Ramen with us, we know a great Ramen place! Don't be such a killjoy."
People don't have to be putting stuff in their face to keep you company.
First, literacy isn't completely related to the writing system. Look at Spanish speaking countries, where the alphabet is more phonetic than the English alphabet.
Second, Chinese characters that are more complex, i.e. consists of more than one radical, are usually composed with a semantic component, giving indication to the character's meaning, and a phonetic component, which gives an indication to the sound of the character. Although this isn't a rule, it helps a lot, and it's not like English doesn't have crazy non-phonetic spellings as well (how tf is "through" supposed to be pronounced for a English learner?)
Last, the Chinese language consists of MANY homophones. This isn't necessarily a bad thing, and not one out of design, but something that is the result of being one of the oldest language families in the world. It allows for the concise expression of many things using only single syllables. You might say, but what about the crazy amount of ambiguity if the language has a lot of homophones? Well, ambiguity is a huge problem in all languages and our brains seem to manage. Now, even though homophones aren't a big problem in spoken language, because of intonation and prosody giving a clue to how to analyze sentences, written language is a different story, and it would be very hard to make an easier system to handle it. For all you engineers, the fact that Chinese has characters is essentially a performance trade off. More information density for more ambiguity.
FWIW, I think Ted Chiang (author of Story of Your Life that Arrival was based on) knows a decent amount on linguistics.
At this point, Chinese characters aren't going to go anyhere. But I think it's an interesting thought experiment to puzzle through.
The Sapir-Whorf hypothesis, which underlies Arrival, has been debunked for years in academia, and the movie, if detached from that hypothesis, is more of a thought experiment on the effect of the linearity of time on human thought than a thought experiment on anything related to language itself.
If even the weak form Sapir-Whorf hypothesis (in restricted contexts) is debunked, then I'd like to read up on it. Where can I look into that because a Google by this layperson doesn't turn up anything on it? Thanks in advance.
tell the eskimo seeing X different kinds of snow to any linguist and they will roll their eyes, debating whether or not to explain why the example is probably not only incorrect, but fabricated for the umpteenth time.
Well the "academia" has been debunked for even longer -- especially when it comes to soft sciences, so that doesn't say much.
A national standard, at best (in the PRC). In Taiwan, Zhuyin (Bopomofo) is universal, taught in schools and found on laptop keyboards, and romanisations are mostly following Wade-Giles rules. In Hongkong, Eitel romanisation is used even though their new commie overlords are forcing pinyin and yale canto down their kids' throats.
Using Taiwan as an example isn't really relevant given that it's majority Cantonese speaking, and Cantonese is a completely different spoken language than Mandarin.
Edit: I'm a clown. I was confusing traditional Chinese with Cantonese. Not the same thing.
"Hangul faced opposition by the literary elite, such as Choe Manri and other Korean Confucian scholars in the 1440s, who believed hanja to be the only legitimate writing system, and perhaps saw Hangul as a threat to their status.[9] However, it entered popular culture as Sejong had intended, being used especially by women and writers of popular fiction.[15] It was effective enough at disseminating information among the uneducated that Yeonsangun, the paranoid tenth king, forbade the study or use of Hangul and banned Hangul documents in 1504,[16] and King Jungjong abolished the Ministry of Eonmun (언문청 諺文廳, governmental institution related to Hangul research) in 1506."
It gets really complicated with japanese - there's a many-to-many-to-many <symbolic>-<phonetic>-<semantic> relationship... One character can have many different readings and due to the homophony, any given reading can have divergent meanings. For example - HASHI can mean bridge, "side of something", or chopsticks - and the character for "bridge" can also be read "kyo" depending on the context.
I imagine chinese is much easier for a japanese native speaker to learn than vice versa for those reasons (since often some of the phonetic forms "on-yomi" of the characters are directly borrowed - with some inflective changes).
That makes no sense. If we had a time machine, we'll be able to trace every modern language back to some African tribe that became the ancestor of all modern humans.
Perhaps you meant having a literary tradition that's among the oldest in the world, but even that doesn't necessarily explain why there would be so many homophones. Languages lose words all the time, even one with great literary traditions.
And while of course every language descends from another, and ancient Chinese isn't understandable to a modern Chinese speaker, there is a long tradition within the Chinese language, and change within the language hasn't been shaped by things as drastic as displacement of the linguistic community, or forced political rule under a foreign power, etc.
And the phonetic component often doesn't correspond to anything any modern person would know.
The problem in both cases is the shift of language through China's long history, and its divergence from the original design of the written characters.
The problem with the semantic components is that the meaning is re-used and stretched over and over. E.g. they used to use a 'foot' radical to represent a journey towards a destination. Later that shifted to mean "in a straight line, not veering left or right". Later that took on the meaning of "straight and narrow" or "straight shooter" or "not-deviant". Then it becomes, "not deviating from the right path". So now, the old foot radical typically means "justice" or "correct".
And the image shifts over time. This is the old foot radical now: 正 Does it look like a foot to you?
The problem is with the phonetics is language shifts. In many cases, in ancient Chinese, the characters do have a phonetic component that hints at pronunciation. but, the pronunciation changed over the last 2000-3000 years, so the pronunciation hint that made perfect sense in the Han dynasty is now meaningless because you're speaking a different language.
The result is that in the end the characters end up being arbitrary phonetic symbols with some arbitrary meanings attached.
It works well enough to guess the pronunciation of a character you've never seen. Or maybe a word you know how to say but don't know how to write.
1: https://en.wikipedia.org/wiki/Radical_(Chinese_characters)#S...
That works well because English and German are very similar. If China moved from characters to Pinyin, someone who doesn't know Chinese could sound it out, but it would be mostly unintelligible (wrong initials, no tones).
That's because most people only know about the 2000 Chinese characters used in everyday situations, so they have this misunderstanding; Others who don't know Chinese heard this and keep on parroting it. If you know more about the language (post-seconadary level), you would know Chinese relies heavily on semantic component.
Here is an example characters I saw in Stanley, Hong Kong a few days ago (as part of a 2 lines poem.)
巍峩 vs the normal form 魏我
Normally the character 魏 means "Tower on an emperor's building" and 我 means "I". By adding the component 山 ("Mountain") to the characters, the words 巍峩 now carries a connotation of epicness one associate with mountain range. ie. "The building and I, as impressive as a mountain range."
I am saying _advance_ Chinese writing has a significant semantic component, but most speakers are not knowledgeable enough to realize it.
>That does not sound very useful, especially when it comes to literacy.
Most people can speak English; You probably need an advance English degree to understand all the puns/word plays in Shakespeare's works (or use Cole's Notes.)
The flexible of a language's intricate details has no direct relationship with the literacy of its speakers(or how easy to become proficient at an language to the point where one can communicate compound ideas with it.)
Back to how useful all these are- not really if you stick with the basic. They are as useful as Shakespeare's or Wordsworth's works in the modern world. You could speak/communicate perfectly fine in English even if you are not able to understand advance details of English literature, especially the vocab side.
And with respect to the phonetic shifts, from a linguistic perspective, most changes of the phonetic radical involve one change of the initial sound, i.e. bilabial to interdental, dental to alveolar, voiced to unvoiced. With respect to the phonetic inventory of the language, these shifts are only one feature shifts within an old SPE framework. These small featural differences, whether conscious or not, (usually unconscious because we acquire languages during our infancy) are picked up and used by the speaker of the language to categorize words.
Your detailed and substantive comment works just as well without this uncivil lead-in.
I don't know if the data supports this or not, but it's an interesting thought.
Regarding the example of 正, Wikipedia suggests the foot radical is actually 足/⻊(radical 157) [0].
It suggests 正 is not a radical but rather derives from radical 77, "stop (止)."
When you add the top line to form 正, the image meaning becomes "stop in the middle," which seems reasonably aligned with the meaning of "right/justice/normal."
Not sure who's right: Wikipedia or you, so please clarify if Wikipedia is wrong. Thanks!
someone could prove/disprove this by analyzing the 3000 most common chars and indicating what percentage contain a semantic component (even if obfuscated).
do you know if wikipedia or the parent is correct about 正 as the radical for foot?
The world's languages, of course, developed nicely without intervention from linguistics, a latter-day mostly retrospective discipline.
Spanish adults read at a 5th grade level, English adults read at a 6th grade level, Japanese adults read at a 9th grade level - this isn't a reflection of education it is a reflection on how difficult the written language is to learn.
As a bad speller with a passing knowledge of Spanish I'm jealous: when I hear a Spanish word I can spell it, with English I have no clue.
Perhaps more to the point, Japan, which uses Chinese characters for nouns, adjectives, adverbs, and verb-roots, has a very high literacy rate. [1]*
[1] https://nces.ed.gov/pubs2014/2014008.pdf
* Not trying to cherry-pick for Japan being at the top, but Japan's not listed in the 2015 UNESCO report that's referenced all over the place. This US DoE report from 2013 seems legitimate enough.
While literacy might be high, there's been a stream of articles over the years in Japanese media noting that literacy in kanji (the component of Japanese based on Chinese characters) in particular has been steadily declining, even among the highly educated; e.g. http://www.japantimes.co.jp/news/2013/07/03/national/kanji-w...
Anecdotally, I feel like Japanese picks a better point in the tradeoff space than other languages - Chinese characters are used where they get some leverage, with phonetics for everything else, even with a bunch of half-kanji words. Still, I'd imagine some of the less-commonly used ones will get dropped over time.
On the other hand, English+Emoji+Txt simplification seems to be approaching the same point from the other direction.
English is probably the worst case among alphabetic languages, but you can still usually guess the exact pronunciation of an unfamiliar word.
> the Chinese language consists of MANY homophones.
But still Chinese speakers make themselves understood without difficulty, even over the telephone. Most Chinese words are polysyllabic. Taking that into account, homophones are rarer than you might think.
There's a language+, Dungan (https://en.wikipedia.org/wiki/Dungan_language), which is to some extent mutually intelligible with Mandarin. It's written entirely in the Cyrillic alphabet without diacritics to indicate tone. Sample text: http://www.omniglot.com/babel/dungan.htm
+ You could call it a dialect, but it has its own script.
A rough 1:1 phonetic correspondence between oral and written languages intuitively (to me at least) would be more fit for the creation and propagation of news words, or am I missing some other natural form of conveyance and propagation that exists in the Chinese writing system?
mobile phone: 手机 （hand machine)
programming: 编程 （compose orders)
QR code: 二维码 （2d code)
hacker: 黑客 （black guest, this also sounds phonetically like "hacker")
For brands and other names, there is a system of transliterating sounds using characters, or simply a literal translation of the name, e.g.
Android: 安卓 (an zhuo)
Microsoft: 微软 （small soft)
Apple: 苹果 （literally apple)
If the author can be taken this seriously given his self-professed ignorance, I am probably much more qualified than him to speak about Chinese characters.
There are two points worth disambiguating here. One is whether Chinese characters hinder the literacy of native speakers and the other is if it hinders that for learners. Neither I nor the author have any authority to speak about the former.
As a heritage speaker, I actually only recently learned Chinese to a level where I consider myself literate. My experience was that the characters were not an obstacle, like the author suggests, but an indispensable tool for rapidly learning the language.
To learn any language, it is unavoidable that you need to memorize thousands of new words. Memorizing a Chinese character is not much harder than memorizing a word. However, the magic of Chinese characters comes when you combine them to form actual words.
The vast majority of Chinese words consists of 2 characters, but because each character also encodes meaning, you can more often than not guess the meaning of a word you have never seen before.
Although you can take advantage of common roots for words in other languages, the scale is simply incomparable. For other languages, you pretty much have to memorize every new word you see.
My experience has been that learning Chinese characters may be a higher upfront cost (although I disagree), once you learn enough, you rapidly understand way more vocabulary because characters themselves encode semantic information.
Perhaps it's simply the lack of any cognates between English and Chinese, and a tinge of cultural supremacy, that make people, like the author, think it's Chinese characters that's the root of everything that makes the language difficult to learn.
Your reaction would be reasonable if someone had claimed that a particular spoken language was hard to learn. Spoken languages are all equally easy, because they are all learned by children in a few years.
But written languages can be arbitrarily difficult, and in Chinese, people with post-graduate educations will often forget how to write common words. This is not cultural supremacy. This is linguistics.
This article takes as a given some assumptions that I don't understand at all. China being resistant to change? There are few cohesive societies that I can think of that have experienced more change than China since the end of WWII. And it would be an understatement to say that the rise of literacy rates are more strongly correlated with industrialization than adoption of phonetic alphabets.
I think it's telling that the author grew up in a Chinese-American community. Expat communities tend to lag their mother cultures in terms of social progress. An experience growing up in a community could explain this "steeped in tradition" characteristic that is attributed to Chinese culture here.
Disclaimer: I'm a white American. My wife is Chinese and emigrated herself from China as a young adult; I get a lot of my perspective from her. Would be interested to hear what Chinese members of HN think of the article.
It's a lot harder to claim that some old druids or even the anglo-saxons are the same society as modern british-descended english speakers.
I would also argue that China's "5000 years of uninterrupted continuous society" is less continuous and uninterrupted than might be implied here. The concept of "China" is more or less a modern, retroactive, political invention, somewhat comparable to identifying the Holy Roman Empire and the Roman Empire as a continuous entity. There is a shared heritage, writing system, and some overlapping territory -- just like Chinese dynasties. If the Habsbugs did a better job of consolidating and expanding their control of the continent in the sixteenth century then maybe we would talk about the continuous, unbroken Roman society.
The metaphor is far from perfect, but I hope my point is clear. The difference between a continuous civilization and whatever the alternative would be called is, in my opinion, just a matter of our perception and a few historical happenstances.
I wasn't trying to be too prescriptive or racist, like "the nature of this people is this, therefore that". But through accidents of history, that constant of the characters has helped to build the illusion you describe.
What? First of all, it doesn't appear that Chinese was the author's native language, given that he was "forced to attend Saturday morning Chinese school" as a child, and did poorly. He was born in New York state and currently lives near Seattle.
Second, and more importantly, Chiang lists several advantages of phonetic systems over purely-character based systems, which hold equally whether we're talking about Chinese or any other written languages. So even if Chinese were his native language, the differences he notes between phonetic and character-based writing systems are general differences that would exist in any language that had both systems of writing.
It's not clear to me, however, that he thinks the differences make phonetic systems "superior". The character-based system has its own advantages it gains from not being tied to pronunciation, e.g., easy readability of classic texts. And Chiang expressly says that "he has no idea if he would be better off in a world" where there had never been a Chinese character-based writing system.
Regarding the rate of change in China, you say that there has been rapid change "since World War II". It seems to me that this rapid change has been possible precisely because China was so backward before then, resistant to change for centuries and millenia before WW II. I suspect -- but have no idea myself -- that the recent rapid change has been enabled in large part by heavier reliance on a phonetic writing system, which as Chiang correctly points out is much more suitable for use with computers.
I read this article as promoting phonetic languages. The entire premise of the article series is things that should never have been invented, so I don't think that's a leap here, even if he does hedge at the end with the "he has no idea if he would be better off in a world" thing.
Did you even read the article you linked to? The author makes so such argument. The point of the article is that it's extremely difficult for non-native speakers to learn written Chinese.
Author is saying it's unnecessarily hard for everybody.
> David Moser also arguing that the Chinese writing system is objectively inferior
1. This essay is part of a series, where guest authors are explicitly asked to "uninvent" something. So there's a level of built-in hot take to this.
2. The author is Ted Chiang, writer of Story of Your Life (and the work that the movie Arrival was based on). He's explored how language affects how we think before (Story of Your Life/Arrival is explicitly about this). So this kind of falls in line with that.
Anyway, I think this is a fascinating thought experiment to work through. What would China be like with a phonetic system? What would change? How much of the culture is derived from the method of language and how much from other factors?
But... they are? They're ideograms and there is at least some reasoning to many of them. Things written with a part of "fire" tend to have a relation to..well...fire and heat.
火, 烧 炊. You might see the relation 秋 shares with fire. But don't fall into the trap that it is 100% consistent because you'd be wrong about 約.
And the last statement isn't even (technically) wrong. I'm not sure if the Chinese word is the same [0] but for Japanese it is 危機. With individual readings of "danger" and "opportunity" although a more reasonable 1-to-1 equation would be "danger" and "occasion". A crisis is a dangerous occasion. 機 happens to have several meanings and one of those is "opportunity". It is probably intentionally misleading but not incorrect to say it is danger + occasion.
You can't have your cake and eat it too. For most compound words the individual meanings of the hanzi/kanji are very relevant - maybe not always relevant but more often than not.
[0] Google Translate tells me it is, so my confidence is >0% but not by much.
There are some exceptions. For example, in Japanese, today (kyou) is written as "今日”, even though it is a single morpheme. In cases such as this the kanji are used for their meanings. In cases such as this, is is not clear how much the kanji actually help, because the sementic information provided is precisly the information that you would get from knowing the roots, and you (or rather a native speaker) does not have to be taught roots (or at least those roots that still have semantic meaning).
「自転車」, 「図書館」, 「作家」, 「日光」,「棋士」. You even have witty ones like 「親子丼」. Just have to be careful about ones like 「多少」.
The reason I would say not to rely on such literal meanings to define words is because some require some stretch of the imagination like「子宮」 and my rule of thumb is if it it requires a stretch of the imagination it's because it doesn't actually mean that. But literal translations can still be useful as mnemonics to learn new words/kanji. Like a pair of training wheels that eventually get removed.
This works with an alphabetic system: in Iceland kids read the sagas in school.
> Because the speech of this country is different from that of China, it [the spoken language] does not match the [Chinese] letters.
[1] https://en.m.wikipedia.org/wiki/Moros_y_cristianos [2] http://www.spain.info/en_US/que-quieres/agenda/fiestas/murci...
Yes, a lousy script isn't a fatal impediment. You can produce beautiful literature even with a poorly-designed script. And I suspect script complexity is only weakly correlated to literacy. Learning to read involves a lot more than recognizing characters and words. But a bad script definitely adds a layer of confusion that is completely unnecessary and ends up creating an elitist and cumbersome language.
The standard counterpoint to complaints about scripts is some form of whataboutism involving English. Yes, English is not remotely phonetic, but the modern form of the Roman script for English is actually pretty decent. It has a lot of visual differentiation between different letters. Compare this to Persian and Arabic where the placement of a dot or three makes all the difference between wildly different sounds.
The Roman script also has relatively few letters. This comes at the cost of ambiguity, and we need to make up sequences for certain sounds (th, sh, ch), but also means the learning curve is a lot less steep. Persian and Arabic have the huge mess involving initial, medial, final and standalone forms of a letter -- it serves no purpose really. Brahmi-based scripts are all abugidas which means vowels are folded into adjoining consonants and so you need to learn about 4X as many symbols as English.
You do have to guess what a letter sounds like in English (g,j,k,c). But this is also pretty common. Tamil doesn't distinguish between voiced and unvoiced consonants. This is an unnecessary annoyance. A much worse example is of Persian, Arabic and Urdu where the script allows one to drop a lot of vowels. This is especially bad for Urdu. For example, the Hindustani words for here (idhar) and there (udhar) are written identically in Urdu, so you need to "backtrack" to fill-in vowels appropriately based on the rest of the sentence.
Yes, English isn't phonetic, and is unstructured and unsystematic. But it is still a pretty decent script compared to some of the others out there. From an Indian perspective, I wish we could move to one script -- perhaps some variant of Devanagari for all our languages. There's really little purpose in differentiating between Gurmukhi and Devanagari. And we should just forget about Urdu all together. It made sense when the goal was to teach people one script that gave them access to both Hindustani and the official court languages of Persian/Arabic. But for today, Devanagari is vastly superior for writing any dialect of Hindustani. Similarly, Kannada has an almost bijective mapping to Devanagari, but each symbol in Kannada is just so much more elaborate and a pain to write. Life would be much better with one script.
Bring up changing scripts though and people -- and this appears to be a global phenomenon -- just go nuts! People need to realize that changing scripts isn't a terrible thing and has occurred many times for many languages. We're not going to forget our languages/culture just because we choose to change the symbols we use in writing.
All YMMV of course, I'm not an expert. I try to avoid being too euro-centric as it is easy for me to claim alphabets are superior because that's what I know, but it does seem like memorizing abstract rules and a small set of letters is easier than memorizing thousands of characters. It is certainly easier to deal with by machine (whether that be a typewriter, a terminal, or designing a modern OpenType font).
This isn't actually true. Handwriting recognition is much more popular than any other means of text input in Hong Kong.