China’s tyranny of characters | Hacker News Reader

China’s tyranny of characters

economist.com

90 pointskafkaesq93 comments

It is important to note that China had alphabetic systems, like Mongolian, since a long time ago. It was a conscious decision not to use them.

Learning to write and read is much easier in alphabetic so the elite opposed it from the start as they viewed it as a menace to their status.

They were right, when people in Europe could read Calvino's printed Bible and own one themselves it changed the status quo radically, creating lots of problems to the people on top. Before printing it took three years of a worker salary to copy a book.

The same process happened in Korea and Japan, with equivalent systems to alphabetic, the difference is that in China elites won, because it was central planned. It was not easy though, specially at first it faced very strong opposition in these small countries.

"as though Europe had thrown away Latin and decided to enforce French across the continent."

That is exactly what happened with Napoleon. Then the fashion language to speak became German, then after WWII it was English, because of the Americans new world hegemony.

They use the same written language with multiple spoken languages. That does not work with an alphabet.

Signs use the same principle. https://en.wikipedia.org/wiki/Hazard_symbol Though Arabic numerals have become fairly universal.

They use the same written language with multiple spoken languages. That does not work with an alphabet.

It can work if you make it work (and ignore side effects).

As attested by the many, many languages have been (forcibly) switched over to invented or completely foreign alphabets, over the millennia.

You can map a spoken language to a random set of symbols.

  long A sound = vr in language 1.
  Z sound = vr in language 2.

But, unless each language also has a different spelling you can only have one language end up even vaguely phonetic, the other is just going to have random symbols mapping to words. As in "Krithnotrix" -> "Bob" because in L1 "Krithnotrix" translates to a short word.

This means if you have 10 languages then ideograms are easier in 9/10 of them.

Except that mappings between real languages and writing systems don't work that way (being in general far from truly phonetic).

Sure, they are not perfect, but you don't want rainbow and monkey to both be spelled VVKX because they are homonyms in another language so they must be homonyms in all your languages.

What is this "alphabetic system" you speak of and why are you comparing "Chinese", whatever it means, with Mongolian?

Language is powerful in China. Wars were fought in China to unify culture. For example, the Chinese language in modern vernacular is generally not referred as a language in itself but referred as zhongwen, or "Chinese culture." The first emperor to unify China realized how important for maintaining power it was to have the people using a single common language (just as it were with the Roman Empire, compared to after its fall).

The Qin dynasty despite its short reign laid the cultural foundations for generations of China (including and up to now). Before that, China was a collection of several warring states with their own written languages that the Qin dynasty immediately abolished. It allowed them to centralize administration of standards and take power from local lords to the central bureaucracy.

What is happening in Hong Kong is a power play by the mainland authorities, just as it was in many instances in mainland history starting from the first emperor.

My mother's family was from southern China, and relocated to Taiwan when the Communists won, my maternal grandparents' family having worked in the Kuomintang government. While my grandparents' native tongue was Cantonese, the only language my mother knows from her years in Taiwan is Mandarin (the Taiwan authorities had a strict prohibition on non-mandarin Chinese dialects).

Here is a written example of differences between the Cantonese language and the Modern Standard Chinese (Mandarin) language when written in Chinese characters. I've seen many examples of signs or other written language in public places in Hong Kong that are incomprehensible to literate speakers of Mandarin, and some examples of written Taiwanese in Taiwan that are incomprehensible to people from anywhere else. How you might write the conversation

"Does he know how to speak Mandarin?

"No, he doesn't."

他會說普通話嗎？

他不會。

in Modern Standard Chinese characters contrasts with how you would write

"Does he know how to speak Cantonese?

"No, he doesn't."

佢識唔識講廣東話？

佢唔識。

in the Chinese characters used to write Cantonese. If you can see the Chinese characters at all as this is displayed on your screen, you should easily be able to see that many more words than "Mandarin" and "Cantonese" differ between those sentences in Chinese characters.

Well that example is rather radical. As long as you know (or can guess) traditional characters you can read a HK newspaper just fine with a stutter here and there.

That's because newspapers in HK are still written in Modern Standard Chinese most of the time. In less formal contexts such as social media, however, written Cantonese is rather more common.

So the subtitles on HK videos, tv etc are usually sanitized to be standard?

Often yes. For video only the really local content (eg. 100Most) are subtitled in exactly the same characters as the ones actually spoken.

(Edit: actually I just rewatched some 100Most content after I wrote the above and no, sadly even their videos are subtitled in Standard Modern Chinese.)

The Taiwanese equivalent would be:

Does he know how to speak Taiwanese?

他會曉講台語無?

No, he doesn't.

他毋會。

The above is actually somewhat close to the Mandarin version, although many other phrases can vary much more significantly.

This is because we speak Mandarin Chinese in Taiwan, using Traditional, rather than Simplified characters, whereas people in Hong Kong speak Cantonese using Traditional characters. It's a different language entirely.

Similar to how you can use Chinese characters (traditional or simplified, whatever) to write Vietnamese and Korean and Japanese, as used to happen many years ago.

Taiwanese is as different from Mandarin as Cantonese is.

Mandarin is the most commonly spoken language in Taiwan, partly because the Taiwanese language was more or less outlawed during Martial Law rule from 1949 to 1987. It's also the primary language for education.

Taiwanese is still commonly spoken as well though, especially in southern Taiwan, and is actually becoming increasingly common (partly for political reasons, which are a different topic).

For comparison:

他會曉講台語無？ (Taiwanese)

他會講台語嗎？ (Mandarin)

I also intentionally used 講 instead of 說 for the Mandarin version, since both are valid and it makes it more similar.

A bigger difference would be the standard greeting:

你好 (Mandarin)

你食飽未？ (Taiwanese)

The examples you gave are in Traditional Chinese.

Which doesn't really matter in this case. The character for "he" in vernacular Cantonese is "佢", which isn't used at all in Mandarin.

The characters for "don't know how" in vernacular Cantonese are written as "唔識" in traditional Chinese or "唔识" in simplified Chinese, which are used in Mandarin, but never in this context (as the example given, the common way of saying "don't know how" in Mandarin is "不會" in traditional Chinese or "不会" in simplified Chinese").

In a sense, the relationship between various Chinese "dialects" and standard Mandarin Chinese is similar to the Scots language's (https://en.wikipedia.org/wiki/Scots_language) relationship to English.

Cantonese to Mandarin is like Ebonics to English.

https://zh.wikipedia.org/zh-hk/%E7%B2%A4%E8%AF%AD#.E4.BF.9D....

In your example:

> 第一及第二人稱用「我」、「你」，與官話相同，但粵語音「我」（ngo5）更保留了中古漢語唐音（*ngɑ̌ ）之疑母（ng-）。第三人稱不用「他」，而是「渠」（俗寫「佢」；東漢《孔雀東南飛》：「雖與府吏要，渠會永無緣」），跟吳語一樣。複數人稱不用「們」，而是上溯至端系的同源形式 [taʔ] 或 [ti]（現代粵語寫作「哋」，本字為「等」，見聖公會的公禱書）。粵語用「係」而不用「是」來代表正面答覆，「係」是明清兩代常用字。這些字眼在中共接管政權初期，仍然有作書面語用，例如李儼《中算史論叢》第一冊 (1955年版本，第210頁) 提及18世紀數學家 Issac Wolfram 時，就寫為「渠係荷蘭礮隊副隊長」。

TL:DR Cantonese is Middle Age "Mandarin" in disguise, plus some Ancient Vietnamese dialects.

IIRC Hakka is closer to an antiquated mandarin than Cantonese is. Also the Chinese "dialects" are different languages. Calling them dialects is like saying Icelandic and Swedish are different dialects of Norse. They're all closely related but not mutually intelligible.

This is only partially related, but I remember watching the news in the nineties and being dumbstruck on how Académie française policed the usage of language in public life. It was probably about use of Anglicisms in television or by politicians.

I forgot about it for years until I learned from Wikipedia that that wasn't just a reaaction against Internet, but a policy that has been going on for centuries: https://en.wikipedia.org/wiki/Vergonha

So the mandarinisation doesn't seem that odd against such a background. The similar effort bore fruit in France as I think now virtually all people in France speak Parisian French compared to 12% or so at the advent of French Revolution and the following policies.

I learned about the Académie Française in French class (1990 or so). I thought they were guarding against anglicisms creeping in through teen slang -- "c'est too much" and that.

Also through vocabulary. 'Le weekend' (vs 'la fin de semaine') is the classic example, both of what l'academie does and how utterly (though heroically) doomed its efforts are.

David Moser, like many people, Chinese or otherwise, who's argued in the Communist party-line idea that a phonetic written language, or in fact any simplification of Han script helps eliminate illiterates, simply cannot comprehend statistics.

Case in point:

Japanese, a language that mixes 3 different scripts, one of which is Kanji (Han characters), is taught to everyone in Japanese since pre-school. An excellent Japanese reader can comprehend about 3000 Han characters, incidentally, is also the average number required for a Chinese speaker to read newspaper. Japan's literacy rate has been around 99% for so long that the government has basically stopped reporting that statistic.

In the past 30-40 years, Taiwan, Hong Kong, and Macau, all regions that continue to use the "complex" traditional Han script, has achieved 95+% literacy rate. We don't hear many people in those regions complaining about how hard traditional Han characters are.

As a comparison, India, which teaches Hindi and English in school, both of which much simpler scripts than Han, has achieved an average of 74% literacy rate in the most recent census.

All of the above tells us that the complexity of a language's written script has very little to do with literacy. The dominant factor in improving literacy is the introduction of free primary and secondary education across a region. No other factor even comes close to achieving a high literacy rate.

So stop with this non-sense that the Han script is too complex to teach, learn and use now. 100s of millions of people have done it.

Another point of this article is that the Han script is too rigid, which is not true. The 6 ways of constructing new Han characters has been recognized for millennia, it's just over the dynasties, the rulers have been reluctant to invent new ones to cater to new ideas. It's the people who forces down a way to use it that's rigid.

I know about 200 words of spoken chinese and can write maybe 4 (including yi er san ;)). Now, admittedly, I'm just picking it up here and there rather than attending classes and spending hours writing characters over and over, but I think learning the strokes for 3-5,000 characters is undisputably 'harder' than learning to write 26 or so letters and a really consistent pronunciation system like pinyin.

That's not to say you're wrong that literacy is all about education. But harder is still harder.

It's true - Chinese kids age 7 can write less words in Mandarin than American kids age 7 can write in English. However, it doesn't change the fact that the difficulty of the language has nothing to do with literacy rate. China has a low literacy rate because huge swaths of the population don't have access to education.

I learned Japanese and Chinese to a fluent level, and at least from my biased point of view Japanese is a tremendously more difficult written language. Not only do you have just as many "characters" (kanji, hanzi, 汉子, 漢子, whatever) as you do in Mandarin, those characters can have many different pronunciations. You also have katakana (phonetic, 46 + modifications), hiragana(phonetic, 46 + modifications), and just for fun the occasional English (phonetic, 26). And yet, Japan has a 99% literacy rate.

Compare that to the USA's 97.9% literacy rate, a country that teaches English (26 letters, phonetic), and I think the argument that difficulty of the language reflects the literacy rate falls apart.

You aren't biased - just educated in the languages. ;)

To illustrate the difficulty of Japanese for others, I'll provide an example:

    生 - raw (nama)
    生まれる - to be born (u・mareru)
    生きる - to live (i・kiru)
    生活 - living; life (sei・katsu)
    生地 - cloth; fabric (ki・ji)

I'll stop there. There are more readings for that single kanji but they are obscure or an "alternative" to a more commonly used kanji. To my understanding, this problem does not exist in Chinese - which uses a single reading mapped to a single character. The same character, in Mandarin, would be read 'shēng' with many of the same meanings (to be born, life, raw)

The above example is one of many kanji with multiple readings in Japanese. How do you know which reading to use? Context and because you know the word. :) That problem isn't unique to Japanese, English suffers from it as well! Think of homophones like tear and tear. You use context to figure out how it is read. It adds some level of difficulty but isn't impossible.

There are lots of characters that have more than one reading in Cantonese, Mandarin and other Chinese languages -- in fact 生 has 2 readings in Cantonese: https://en.wiktionary.org/wiki/%E7%94%9F#Pronunciation

I stand corrected, thank you. I had Mandarin in mind, although that fact was unclear since I only mentioned it in my example. I'm unfamiliar with Cantonese and lack any knowledge at all of other Chinese dialects. So at least for that character - it has one reading in Mandarin. The two readings in Cantonese are a single (and similar) vowel apart, nothing like the differences between "nama" and "ki". So I'll change my argument: even though Chinese has multiple readings - Japanese takes it a step further in complexity.

According to this page [0], most sound differences in Mandarin are a variant opposed to a completely different sound. Variant sounds are easier to memorize because they often "make sense". In Japanese, variant sounds often occur because it is "easier to pronounce". If I had to wager a guess, Chinese variant sounds are done for the same reason - though I could be wrong on that as well. :)

Thank you for the correction.

[0] http://pinyin.info/chinese_characters/

I agree, the multiple readings in Cantonese/Mandarin are nowhere near as different as the different readings for the same Kanji in Japanese. I never meant to imply otherwise, sorry if I sounded I was.

As someone trying to learn Japanese, it is one of the harder parts of the language for me to learn. Conjugations frustrate me more though. :)

I didn't take it as such - I took it as a correction to my misinformation and updated my argument to reflect the new information. My backing point was 'Japanese makes kanji more difficult than Chinese" but my information wasn't correct originally.

Might I ask what you find difficult about conjugations? I might be able to help. With the exception of U-Verbs (五段) I found that they are very consistent and very "mathematical". Nothing else gets as complicated as "house and houses but mouse and mice, goose and geese but moose and moose". While there are still exceptions due to etymology, they are relatively rare. The same 2 verbs are exceptions to most everything and there are only a few common words which are exceptions, such as いい conjugating as よい.

I concede that 五段活用 is a pain in the rear but it is something you get an "ear" for over time. So if that is were you get frustrated my only advice is chug-chug-chug along! Eventually if you conjugate one of the words incorrectly it sounds wrong to you, even if you aren't certain you were wrong or not.

Yes, it is the verb conjugations that are tripping me up (the English Wikipedia page on this [1] has a complicated chart that looks slightly scary), but they are getting easier the more I use them. I do find many parts of Japanese grammar to be very consistent, like you said.

[1] https://en.wikipedia.org/wiki/Japanese_verb_conjugation

If the aim is to improve literacy rate, focusing so much energy into to changing the entire set of characters for a written language is just incredibly stupid. It has a small marginal improvement but an tremendous repercussions. The Chinese simplification process has produced so many characters that completely changed the meaning of the character. The process has also merged so many characters that people's family names are not distinguishable. The Chinese simplification process is one the of the greatest tragedy in the history of social experiments.

What the Communist should do is to revert to traditional Chinese before it's too late, and start investing more in free education.

As to the difficulty in writing the Han script, you can compose the characters just like every other language. You just need to learn the radicals. If you are trying to comprehend written Chinese in Pinyin, which was invented only for Mandarin, you'll find that it's next to impossible as the Pinyin system removes too much context. There are only 4 tones in Mandarin as opposed to Cantonese's 9, so there are a lot of characters that sound exactly the same in Mandarin.

There is an equivalent of Buffalo buffalo Buffalo buffalo buffalo buffalo Buffalo buffalo in Chinese called 施氏食獅史, which is a poem of a story about a person name Shi eating a Lion. 92 characters all sound exactly the same.

Learning the radicals does allow a certain amount of 'chunking' and makes things easier, but it's by no means a consistent system like an alphabet. Sometimes one radical is sound and another is meaning, sometimes neither, sometimes there's like 5 of them crammed in there, and every character has a different little mnemonic story. Alphabets with consistent rules are objectively much less work to learn.

Anyways I certainly wasn't advocating some sort of government-imposed wholesale switchover and banning the characters.. we couldn't even manage a switch to metric in the US, that would be crazy.

As for pinyin being insufficient.. people disambiguate those words from context all the time in spoken conversation, right? Why wouldn't pinyin with tone marks work as a written language? I'm far from fluent or conversational, honestly asking.

Listening is very different from reading. When you are reading phonetic symbols, at least when you first learn the language, you have an extra step of mapping the symbols to the sounds, then the meaning. Familiarity of a language is achieved by removing the step that maps to sound, essentially by mapping the "shape" or the combination of letters in a word to a token of meaning. With pinyin, since the Han script has almost an exact 1-1 mapping from syllable to character, but far less number of commonly used syllables in Mandarin, you simply cannot reconstruct the meaning by looking at the pinyin spelling of a Han character and retrieve the meaning of the character the pinyin spelling maps to. That's what I mean by context.

That's a really interesting point, thanks :)

I'm tempted to try and argue for the sake of it but I'm not even conversational besides 'baby' and 'food' topics and nowhere close to literate. So thanks again for the comment.

Hmm, characters are square and have fixed width. The fonts are automatically fixed width when printed or even handwritten (well enough). With pinyin, you know, like any other phonetic languages, even Japanese, you can hardly match line width with syllables so perfectly. And that's why the Chinese classic poetry, in my biased opinion, is a level higher than, say, its English counterpart, as it requires more effort to not only rhyme the lines, but also match the length of each line.

But it's not just 26 letters, it's also the 26^n combinations of letters that one has to learn. You might argue that it's not really 26^n because there are rules about vowels etc; but then there are also rules and patterns regarding how the strokes combine together to form Chinese characters, and if there's one thing the human brain is really good at it's recognizing patterns.

//As a comparison, India, which teaches Hindi and English in school, both of which much simpler scripts than Han, has achieved an average of 74% literacy rate in the most recent census.// The reason for lower literacy rates in India may not anything to do with how complex their scripts are. Elites just passed on their Sanskrit scriptures orally among themselves to exclude others from learning them. Their laws also denied education to the working class. In modern times, many children just can't go to school for many reasons. There are no schools accessible or affordable, they may need to fetch firewood, water, may need to take care of cattle, family members, and so on. (My mom curse their parents to this day, because they stopped her school in her 4th grade, to take care of her baby sister). In south India, there are free meal schemes for school children to alleviate these problems to some extent.

The above post repeats some widely believed but inaccurate statements about literacy in Japan. The frequently mentioned 9x% literacy rate is a myth. In the premodern period, it was much less than 1% of the population. In 1948, the rate of full literacy was reported to be 6.2% of the population. In 1955, a literacy census reported that functional illiteracy ran at 60% of the adult population. See Rubinger[1] for the premodern period and Unger[2] for post-WW2. Unger also maintains a bibliography on Japanese literate on his web site.[3]

[1] Richard Rubinger. 2007. Popular Literacy in Early Modern Japan.

[2] J. Marchall Unger. 1996. Literacy and Script Reform in Occupation Japan.

[3] https://u.osu.edu/unger.26/books/literacy-and-script-reform-...

I find it extremely difficult to believe Japan was able to become a major technological, economical and military world power before WW2 with an essentially illiterate population. I'll keep these works in mind, thanks.

In Hong Kong, I sometimes see peoples screens when I'm using public transport. WhatsApp audio recordings are very popular, sometimes I can see the entire conversation is just audio recordings back and forth. Other times it's Chinese (I can't differentiate which), sometimes English, often a mix. I've only seen someone drawing characters with his finger once.

Keep an eye out for either Pinyin or Zhuyin. I believe they are somewhat common for texting purposes. You input this stuff and your device starts displaying characters for you to pick from avoiding the slower process of actually drawing characters.

Links for those unfamiliar with those two things.

https://en.wikipedia.org/wiki/Pinyin

https://en.wikipedia.org/wiki/Bopomofo

Although in the case of Hong Kong, neither Pinyin input nor Zhuyin input are likely to be used, as both are based on the phonetics of standard Mandarin Chinese. There are alternative input methods that are shape-based (https://en.wikipedia.org/wiki/Chinese_input_methods_for_comp...) such as the Cangjie, and Cantonese-only input methods.

I actually have not come across Cangjie before and it looks really interesting. Thank you for the link :)

I guess I've just been assuming any Roman characters are English, some definitely might be Pinyin. I generally try not to stare at peoples screens for too long haha. I did an intro to Chinese (Mandarin) class a few years ago so I'm aware of Pinyin but have forgotten almost everything (except some of the bits I found particularly interesting like word order for times and places, how numbers are represented and a few other things).

This is a useful reference for hanyu pinyin/tones: http://lost-theory.org/chinese/phonetics/

> drawing characters

It's only drawing if one doesn't write the language. If I have forgotten how to write a character and have to refer to a dictionary, I'll still be writing the character, not drawing it. Compare/contrast with my trying to write Thai, Tamil, Arabic, etc - then I'll be drawing.

Personally I write Chinese on the iPhone/iPad using my index finger much faster than I enter pinyin. I find that iOS's character recognition is fairly good - as long as I get the stroke order right, even though what I have written looks like random scribbling on screen, most of the time it presents the right character.

> gaining so much regional identity and independence that they want to do a Brexit of their own.

Yeah, I'm pretty sure they've wanted their own Brexit from day 1. At least Hong Kong does, and even Taiwan probably too. Especially Taiwan. I feel sorry for Taiwan, watching stuff like this (http://reason.com/blog/2014/09/30/hong-kong-student-begs-for...) from Hong Kong and knowing that's a real possibility for them. At least the pro-China people in Taiwan (yes, they exist) will have to reevaluate their opinions after seeing how China has handled the Hong Kong return...

At least in Hong Kong, "new" characters for spoken Cantonese are often derived from existing Chinese characters that sound similar, despite those characters having a completely different meaning. Wikipedia has a fascinating entry [1] with many examples of this.

[1] https://en.wikipedia.org/wiki/Written_Cantonese#Cantonese_ch...

> But central authorities are also now worried about any regional languages (which it insists on calling dialects) among the Han majority

It's not as clear cut that variations of Chinese are languages and not dialects as the economist makes it out. This is very similar to arabic. Wikipedia for instance also calls them dialects. This is very different from Uighur and Tibetan which are pretty clear cut separate languages, but which are intermingled here in this article.

Wikipedia calls them “varieties” because to call them “languages” would result in a giant political flamewar. https://en.wikipedia.org/wiki/Varieties_of_Chinese, also see https://en.wikipedia.org/wiki/Talk:Varieties_of_Chinese

That doesn’t mean they aren’t distinct languages (and not mere “dialects”) by any definition a modern linguist would typically use.

Calling them “dialects” would be similar to calling English, German, and Dutch “dialects” of “Germanic”.

English is pretty far from continental western germanic (German, Dutch, ...).

Continental western germanic is actually a good example of a group where the distinction between dialect and language is not clear. There's a dialect continuum in the area [1], so you often can't draw clear cut lines what belongs to one language and one to another.

- With Low German the question is wether it is an own language or just a dialect of German.

- With the three frisian varieties the question is wether they are different languages or just different dialects of Frisian.

- Luxemburgish can be considered just a dialect, like the German dialects spoken just behind the border that are essentially the same, or an own language.

- Dutch belongs to a dialect group that also spans over wide parts of Germany and Luxembourg [2]

- And there are many more things to think about like Afrikaans or Yiddish

[1]: https://en.wikipedia.org/wiki/Dialect_continuum#Continental_...

[2]: https://en.wikipedia.org/wiki/Franconian_languages

A better, more neutral term would be 'topolect'. It's a better translation of the term 'fāngyán' (which is typically translated as 'dialect'), and blurs the line between the English terms 'dialect' and 'language'.

"A language is a dialect with an army and a navy"

To me, any non-mutually-intelligible dialect is in position to be called a language.

> To me, any non-mutually-intelligible dialect is in position to be called a language.

The problem is that there is often a continuum of dialects where dialects that are near each other in this continuum are mutually intelligible, but those that are further apart are not. Where do you set the boundary between languages and dialects there.

P.S.: I'm not a linguist but because of this fact I would consider to define languages and dialects in a quite different way (the following concepts and description is my personal invention and not taken from some linguistics literature).

First we define the concept "speak-kind" as a generalization of language, dialect etc. (we use this word so that we aren't burdened by the fact that generally a dialect is defined to be some variety of a language etc.).

Now consider the graph where the vertices are the "speak-kind"s. This graph has an edge if two "speak-kind"s are mutually intelligible. Now we can define the concept of the "cont[inuum]-language": a "cont-language" is defined to be a connected component in this graph.

In this sense Bavarian and Dutch would be of the same "cont-language" (exercise: How would the connecting path in the graph probably look like?), though probably not mutually intelligible, but Hungarian and English would belong to a different "cont-language".

Conversely, the opposite happens as well, where people have higher mutual intelligibility of different language families than other dialects within their own language. I've heard of this in places like India, where "Hindi" is a standard language of a massive dialect chain that is spoken next to various Dravidian languages. A Hindi speaker might be better able to understand a Dravidian speaker from the same city than a Hindi speaker from a distant city.

I believe it's no longer so these days when >3/4 of people live in cities. Cities tend to normalize and homogenize languages. So you will ask: "Is the language spoken in city X is a dialect of a language in a city Y? Is it mutually intelligible?"

The answer is often "no" or "not really".

It's no longer a continuum rather a graph.

Weird. On Firefox, with ad-blocker and PrivacyBadger disabled, I am stuck on a page with only navigation menus and a title that reads 'Explicit cookie consent'.

In Chrome I get the expected I-agree-to-your-cookies consent form.

Anyone else getting this?

I've had this for a while on this domain, which Chrome and Ghostery. I haven't been able to read their articles for months at least.

"the use of a standard language is undeniably helpful in educating the poorest and helping them engage with broader development trends in the country and across the world."

This is what the powerful say every time they're trying to forcibly strip a community of its language and culture. But is it even remotely true? The major varieties of Chinese have as many speakers as major European nations. Is it really to be believed that local varieties of Chinese or minority languages can have no recognition in school?

Crikey, 'duang' is a much better word than 'boing'.

https://en.wikipedia.org/wiki/Taiwanese_kana

Hokkien Taiwanese is no longer particularly prominent in Taiwan, let alone the Taiwanese kana (according to that link they're not even all in Unicode).

https://en.wikipedia.org/wiki/Zhuyin is more relevant, as it's still moderately widely used today in Taiwan.

Out of curiosity, how do writers deal with dialog with Chinese characters? It seems like characters would give the meaning but not the voice.

There are some complications when Chinese characters are used to write non-Chinese languages, but in general Chinese characters correspond to sounds (one character per syllable). You might like this article, "The Ideographic Myth". http://www.pinyin.info/readings/texts/ideographic_myth.html

That was a really fascinating read.

> It seems like characters would give the meaning but not the voice.

You'd be surprised. It's easy enough for a writer to distinguish dialog spoken by a country bumpkin vs a sophisticated urbanite or between a ruler and a peasant because they have a different set of vocab and there are Chinese characters for most of the vocab that the different speakers will use.

Can you elaborate, because I don't understand what you mean. Is it because you think they're ideographs that have no corresponding phonemes? They do. Every character is a syllable complete with a tone.

It's fairly common in English for authors to give phonetic spellings in dialog, to make things seem more "real" or to indicate accents. To pick a random line from my favorite book, "What for? You caint do nothin for it." I'm not sure if that's what OP is asking, but I'm actually curious how you would indicate the same effect (essentially, encoding nonstandard pronunciation in a comprehensible way).

As far as we can tell, all writing systems with independent origins started out as logographs that where stylized representations of the words, and then they started to change by adapting the rebus principle--basically, puns based on homonyms rather than their abstract representation. Egyptian, Sumerian, Mayan, and Japanese all made translations thence into syllabaries, whereas Chinese remained primarily logographic. Egyptian was borrowed into Phoenician as an abjad (like an alphabet, but no vowels), whence the Greeks added vowels as separate letters, and the Indians readded vowels by modifying the base characters (known as an abiguda), and further complicated the system by stacking the consonant clusters on top of each other.

Chinese can represent phonetic systems by a similar rebus principle. One crucial difference is that syllabaries and abigudas tend to work on having one character per consonant-vowel pair, which works poorly on languages like English where consonant clusters can get atrocious (strangsts: 1 vowel, 8 consonants). Many languages tend to limit themselves to CV or CVC (which can be handled by special rules like C₁V₁ + C₂V₁ really means CVC, e.g., Mayan), so it's not so much a problem unless you're trying to transliterate completely different languages. Of course, Latin script has its own problems in transliteration: it's atonal, it relies on voiced/voiceless distinctions (Chinese uses aspirated/unaspirated primarily!), and good luck if you have languages that have four sounds we'd analyze as "h."

I'll point out that Chinese script has been used to write other languages--most notably Japanese, Korean, and Mongolian, none of which are related to the Chinese languages.

Another problem with Chinese is that, because you can write the same sound many different ways, you can make mere phonetic transcription a political deal. A transcription might happen upon the character for "small" or "stupid" or "backwards," which might not go over well if it's someone's personal name.

Mayan is strictly CVC and CV? I always assumed that Chinese remained logographic because of the phonology's quick shift to CVC within a thousand years or so of the script's invention. Reading a lot of Baxter and Sagart's papers and monographs I've come to remember many of the Old Chinese reconstructions, but it would have been a nightmare for a farmer to learn.

My recollection is that Chinese remained logographic in large part because its language had a lot of homophones.

We usually substitute with a different sounding character which results in a non-sensible string of characters, yet still understandable (to native speakers at least) given sufficient context. As an example, we often joke about how the staff of a certain local noodle soup chain often speak in heavily accented Cantonese by writing their speech as "勿演", "懶肉", "生小辣" ("don't act", "lazy meat", "raw little spicy") instead of the correct "墨丸", "腩肉", "三小辣" ("cuttlefish balls", "pork belly", "3rd little spicy").

In otherwords, Chinese is "misspelled" for dialect... Just like any other writing system.

Maybe you and the original commenter mean to refer to "dialect" rather than "dialog"?

https://en.wikipedia.org/wiki/Eye_dialect

You can convey provincial language through different diction, and a different frequency of usage of certain words, e.g. https://en.wikipedia.org/wiki/Sentence-final_particle

Although that is fairly mild, I've seen authors that butcher the language to indicate accents, making their work unreadable. Introduce the character, mention their accent, and move on with your life.

I don't know if anyone does it, but one idea might be to use different fonts to indicate accents.

Other responses have shown why characters do the job just fine, and they're not wrong, but there's something to be said for phonetic literature, which is why there is Pinyin Literature contest [0]. See [1] for some more explanation.

[0]: http://languagelog.ldc.upenn.edu/nll/?p=26503

[1]: http://languagelog.ldc.upenn.edu/nll/?p=26546

At least in Mandarin, given enough context, each Chinese character has one correct pronunciation. It's frequently the "context" that trips up non-native speakers.

Generally representing dialog is not a problem. In the case that they want to represent a word/sound that does not have a character assigned to it there are a number of strategies to deal with it: romanization (pinyin), zhuyin fuhao (phonetic notation), non-standard character or pick a character with similar meaning or sound. Generally context is sufficient for native speakers to figure out what they are trying to convey.

"It seems like characters would give the meaning but not the voice."

Isn't this true for any general writing system? The textual recording of one's speech is always lossy. That's why the writers, when they deem it to be important enough, add descriptions here and there about the speaker, to capture all kind of verbal or non verbal details.

>The inflexibility of the Chinese script has always reinforced the inflexibility of the Chinese state.

The democratization of Taiwan and Hong Kong must be pretty frustrating to writers who want people to take these kinds of baseless yet simplifying statements seriously.

> They do not accept a canon of unchanging characters handed down from antiquity. They want to make up new ones and use Roman letters

I've never heard such claims. Why do we need to make up new characters when there are more strange characters than necessary? Even the youngest can enjoy the so called Martian language[1] (the Chinese version of the leetspeak) with valid Chinese characters. And you can actually make up new characters by writing components as individual characters, and the outcome is just a larger and funnier character.

As for wishing for a Roman alphabet... I think it's more of a fancy of those already using such an alphabet. If you grow up seeing Chinese characters everywhere but Roman letters not so often if ever, the characters will be a habit. Also, pinyin may not be very easy for many parts of the country, as the inhabitants will have difficulty in making the precise pronunciation. Their words can be easily understood, but it will be a hard time for them to pin down the correct pinyin. To input characters, the input method is very forgiving, allowing missing or redundant g's in ng or h's in sh, ch and zh. Some people even need to guess between r and l, h and f. Characters are a much better, semantic way of communication.

"reinforced", not "caused".

"Movable goalposts", not "fixed goalposts".

"the boxy prison of Chinese characters"

What does it even mean? We are all "prisoners" of our language, because we can't easily think and express anything that cannot be molded in this language. But Chinese characters may actually give more "freedom" to their users, in that they convey directly their meaning, without the necessity of a sound. See 凹 and 凸, they mean concave and convex. Avoiding the articulation over the pronounciation is a blessing in many respects. It allows more people from distant cultures to converge with the writen language, and share much more.

Except that's not what "concave" and "convex" look like.

It also means that you need to agree with everyone, what the meaning of every symbol is. For instance symbol of "convex" you give, might also be a street intersection, so you can't really just rely on the shape of a character, and you need to spend many years memorizing each one of them. If you look at reality, your assertion that

> people from distant cultures [...] converge with the writen language

is really not true. Chinese writing systems aren't universal in a way that makes them easy to pick up to express yourself in. Quite the contrary.

Only a tiny minority of Chinese characters are pictographic, and most characters have a phonetic element. You'll have a better chance guessing the pronunciation of a random Chinese character than guessing the meaning of a random Chinese character.

The value of iconocity in the more pictographic characters is more of a mnemonic tool. It's much easier to remember the meaning of the more pictographic ones.

I guess it relates to this: "It is one of many slang or dialect words and phrases that you can say but not write using Chinese characters."

This is not true, Chinese people are very able to find Chinese characters to express slang or dialect words, they do it all the time, and the case of M. Chan's hair is a rare exception.

Reply on news.ycombinator.com