Inching toward digraphia, with a note on the universality myth

« previous post | next post »

The subject of digraphia in China often comes up in our discussions about the Chinese writing system on Language Log (always be sure to check the comments on the posts, because much good material is often added in them), e.g.:

"Digraphia and intentional miswriting " (3/12/15)

"Substituting Pinyin for unknown Chinese characters " (12/3/13)

"Creeping Romanization in Chinese " (8/30/12)

"Character amnesia and the emergence of digraphia " (9/25/13)

"Dumpling ingredients and character amnesia " (10/18/14)

"Which is worse? " (1/21/16)

The existence of digraphia in China means that a choice does not have to be made between characters and Romanization, but that both can co-exist and simultaneously function in the spheres and roles that are best suited to them.

In my estimation, alphabetical writing could never be promulgated by fiat in China the way it was in Turkey, Vietnam, and elsewhere.  Already in 1951, Mao Zedong had issued the following directive with regard to script reform:

Wénzì bìxū gǎigé, yào zǒu shìjiè wénzì gòngtóng de pīnyīn fāngxiàng 文字必须改革,要走世界文字共同的拼音方向 ("[Our] writing [system] must be reformed; [we] should take the direction of phoneticization in common with the scripts of the world".

Later on, Mao watered down his demands for script reform to half-baked character simplification, which, far from solving the manifold problems of China's antiquated, cumbersome writing system, only exacerbated them.

If mighty Mao could not achieve the phoneticization of the Chinese script, no leader since him, including Xi Jinping today, would be able to bring it about.

Incidentally, pīnyīn 拼音, aside from meaning "spelling", which is how it is normally understood in the expression Hànyǔ pīnyīn 汉语拼音 ("Sinitic spelling"), may also be translated as "phonetic(ization)".

In practice, digraphia (characters + pinyin) is gradually becoming a reality.  Characters, of course, are already here, but in what ways is the Latin alphabet increasingly making its presence felt in China?

1. In advertisements

"Duang " (3/1/15)

"Digraphia and bilingualism in a Nissan ad " (6/3/15)

"A bilingual, biscriptal product designation in Taiwan " (2/7/14)

"A trilingual, triscriptal ad in the Taipei subway " (1/20/140

2. In education, for the phonetic annotation of characters and for other purposes

"Archive for Pedagogy "

3. For computer inputting

"Chinese character inputting " (10/17/15)

4. For movie, TV, and video subtitles

5. In restaurant menus

6. For braille and semaphore

7. For signage

8. For writing various foreign languages

9. For writing Sinitic languages other than Mandarin and a host of non-Sinitic tongue

This particular category puts the lie to the "universality" myth — the idea that the character system magically unifies Mandarin, Cantonese, Hakka, Shanghainese, and a dozen other languages, and allows them all to be written identically.  Nothing could be further from the truth.  We have demonstrated again and again that many of the most common morphemes in Cantonese, Taiwanese, etc. (even some of the most favored Pekingese expressions) cannot be written with characters, their grammar and phonology are different, and so on.  Since there are no characters for essential morphemes, those who wish to write full-blown Cantonese, Taiwanese, etc. have to do one of three things:

a. Borrow another character purely for its sound and without regard to its meaning, thus setting the stage for semantic interference.  For example, máidān 埋单 (lit., "bury the bill"), a Cantonese expression which actually means "bring the bill", the mái 埋 (lit., "bury") being used to represent a Cantonese morpheme for which there is no known, widely accepted Chinese character.  Like many other Cantonese terms, máidān 埋单 (lit., "bury [–> bring] the bill") has become a part of Mandarin vocabulary.  However, since máidān 埋单 ("bury the bill") doesn't make sense to Mandarin speakers, either they will engage in arcane explanations (e.g., when the waiter brings the bill, they will "bury" it under a plate or in a folder) or they will change the character in a vain, desperate attempt to make it mean something logical in Mandarin, thus máidān 埋单 ("bury the bill") –> mǎidān 买单 (lit., "buy the bill"), which doesn't really make sense either, does it?

The Cantonese pronunciation of máidān 埋单 ("bury [–> bring] the bill") is maai4daan1.

From the online CantoDict at the sheik website, we can see why maai4daan1 means what it does in Cantonese and why — even though it has become a very common expression in Mandarin as máidān 埋单 ("bury [–> bring] the bill") — it causes problems (semantic interference) for Mandarin speakers:

MSM mai2 man2

Cantonese maai4


[1] [v] bury; cover up; lay sth underground
[2] [v] conceal; hide; lie low


[3] [] [v] finish up; close out; settle (an account)
[4] [] [v] move closer/nearer to; close in/up
[5] [] [adv] closer together; nearer to
[6] [] [adv] even; too; also; as well

b. Create a new character just for that morpheme

je5 嘢 (n. "thing; things; article; articles; goods; matter; stuff")

mak1 maak1 嘜 (n. 1. "mark; trademark; brand [name]"; transcription of English "mark", 2. n. "mug; metallic container")

di1 dit1 啲 (Wikipedia says "genitive, similar to 's but pluralizing"; CantoDict has di1 [1] [pron] "some; those" [2] [adv] "a few; a little bit" dit1 [adv] "only a little bit; every little bit"); note variants 尐, D

There are hundreds of other such special Cantonese characters, both for content words (nouns, verbs) and grammatical particles and function words (see here and here).

c. Write the morpheme in Roman letters — that's so simple that I don't even have to provide an illustration, but see "D" just above and this post:

"The Roman Alphabet in Cantonese " (3/23/11)

10. On the internet, especially by young people for creative fun, and often to avoid the censors

11. For writing so-called zìmǔ cí 字母词 ("letter words", i.e., words that consist partially or wholly of Roman letters) in Chinese:

"Zhao C: a Man Who Lost His Name " (2/27/09)

Lu Xun and the Zhao family" (1/5/16)

"Creeping Romanization in Chinese " (8/30/12)

"A New Morpheme in Mandarin " (4/26/11)

And so on and so forth.  I could list many more categories of how the alphabet is used in China and could provide additional Language Log posts that provide evidence for each category, but that would be both tiresome and unnecessary.  The point has been adequately made:  the alphabet is here to stay in China.  It is solidly entrenched, and the purposes to which it is applied continue to expand with each passing day.  This is happening despite what anyone may feel in favor of or against it.


  1. Eidolon said,

    January 26, 2016 @ 3:53 pm

    I'm not sure I accept the argument that the Chinese writing system is inadequate for the needs of communication between different Sinitic languages. After all, it did just that for thousands of years, and while I'd agree that this was at the cost of every literate learning a sort of "written koine," and not because the colloquial form of local languages were/are perfectly expressible using characters, the process as such was still easier than it'd have been to learn dozens/hundreds of different mutually unintelligible Sinitic languages.

    To this end, I think logographic writing systems do have certain advantages for communication between similar, but nonetheless mutually unintelligible, languages, but that this advantage is not absolute, which is my take-away from the works of Unger, DeFrancis, etc. They were correct to argue that the Chinese writing system isn't just a series of ideograms, but are too quick to dismiss the possibility that, nonetheless, a subset of the writing system can be used that way.

    For example, it is plain to see that a character such as "火" can be pronounced in a hundred different ways, and still be understood as "fire", while "fuoco" cannot be written as "fire" in any accepted alphabet of Italian/English. Thus, whereas a logograph user could just memorize 火 once, an alphabet user would have to learn a hundred different pronunciations for a hundred different languages. That is, unless you start treating "fire" as a logograph, which defeats the purpose.

    Thus, I am still of the belief that Chinese characters were one of the main enablers of Sinitic linguistic diversity and has historically sustained it. That is to say, had China switched to an alphabet early on, officials from different regions of the country would've had to conform to the same pronunciation of morphemes in order to communicate in writing, and this would have exerted a large amount of political pressure towards achieving elite linguistic uniformity. This pressure could've led to permanent separatism along linguistic lines, as occurred in historical Europe, but could also have led to the development of a spoken lingua franca extending across China, in a situation similar to the European Union & English today.

    Whichever the case, we'd not be in the same situation we are in today, where Cantonese, Hakka, Hokkien, etc. speakers could communicate between themselves and with Mandarin speakers using the written equivalent of a Sinitic koine, while not learning to speak any of these languages or the koine. It is a fact that people can and have used a subset of Chinese characters as ideograms, and I imagine that by doing so, they're able to shift the cognitive load of learning the spoken language to memorizing ideograms. It's not a black and white situation of being able to freely express ideas without any spoken language learning, but shades of grey are still shades of grey:

  2. Victor Mair said,

    January 27, 2016 @ 10:51 am

    "Growing Numbers Of Chinese Teens Are Coming To America For High School"

    NPR- January 26, 2016‎‎‎

RSS feed for comments on this post