East Asian Writing Systems

Chinese Writing: Home >

The Chinese Writing System (4)


Characters are often put forward as a bridge between dialects -- in fact, they are often credited with single-handedly maintaining the unity of the Chinese language.

The ability of characters to act as a 'bridge' between dialects rests on the fact that the modern dialects are genetically related, all being descended from Old Chinese (Classical Chinese). Although the modernday Chinese dialects started drifting apart more than a thousand years ago, their vocabularies preserve large chunks of the older language, disguised by differences in pronunciation. Take the following examples from several different dialects:

Character Mandarin Shanghainese Hokkienese Cantonese
白 'white' bái bak23 beh24 baak6
八 'eight' bak5 bueh1 baat8
日 'sun, day' niek23 lit24 yat6
五 'five' n13 ggoo22 ng5
十 'ten' shí sak23 zap24 sap6

The Chinese are very aware of cognates between dialects. By noting connections between corresponding sounds, speakers can very quickly pick up important common chunks of other dialects. To see what is meant, check out the following sites showing sound changes that have taken place between Classical Chinese and the modern dialects (with reference also to Japanese, Korean, and Vietnamese) and how regular correspondences can be drawn: Chinese Dialects and Chinese Numerals - A Comparison of Readings from China, Korea, Japan, and Vietnam.

The characters formalise the etymological connections among dialects by representing cognate terms with the same character. For instance, (Mandarin), niek23 (Shanghainese), lit24 (Hokkienese), and yat6 (Cantonese) are all descended from the original Old Chinese word for 'day' and are all written with the character 日. In a sense, the characters act as a kind of shorthand, helping make clear which word (or form) is referred to and encapsulating its identity across the many dialects of China.

Because of this, characters can facilitate inter-dialectal borrowing. For example, Cantonese 單車 / 单车 daan3-tse3 'bicycle' has made its way into southern Mandarin as dānchē; Cantonese 的士 dik7-si2 'taxi' has become Mandarin dìshì, etc. Official, academic, scientific, and formal language flows the other way, from the standard language to the dialects, in character form.

(Interestingly the bridging function of the characters has even been extended to Japanese, although because Japanese operates by somewhat different rules, the results can be bizarre, e.g., 大阪 Ōsaka (Osaka) becomes Dàbǎn in Mandarin!)

Bridging function overplayed

However, the similarity among dialects and the role of characters as a bridge can be overplayed. There are very real differences among the dialects, including differences in grammar, sizable chunks of vocabulary that are not shared, differences in the usage of vocabulary that is shared, colloquialisms, etc. Even Chinese dialectologists, led astray by an over-reliance on characters, first concentrated on differences in how characters are read. Since there is a quite a lot of vocabulary for which there are no characters, this fails to convey a complete picture of a dialect.

Even more seriously, the entire concept of characters as a bridge between dialects is being eroded by the new model of Chinese that has been adopted in the modern era. Below I will look at the historical underpinnings of the phenomenon and show how it is being eroded.

Underpinnings of the bridge between dialects

Although it almost certainly had internal dialects, Classical Chinese appears to have been a single language, written and spoken. Starting more than a thousand years ago, the different dialects gradually drifted apart to become virtually separate languages, with different pronunciation, different vocabulary, and different grammar.

Through all this, the formal written language of the literati continued to be Classical Chinese. However, since characters do not directly represent pronunciation, the old Classical pronunciation was gradually forgotten and Classical texts came to spoken in the pronunciation of the different dialects. Educated people thus had a common written language but pronounced it quite differently. Being only loosely tied to pronunciation turned out a plus because the characters could accommodate many different pronunciations. The characters are thus a kind of 'super-script' transcending and uniting the dialects. This may be represented somewhat simplistically as follows:

Standard Written Language (Classical Chinese)
No standard pronunciation
Hokkienese pronunciation
Wu pronunciation
Northern pronunciation
Cantonese pronunciation

This is the original basis for the concept of Chinese as a language unified by its writing system.

It did not end there. Although Classical Chinese continued as the standard written language until the 20th century, a second written language gradually emerged alongside it. This second language was known as baihua ('plain language'). Baihua was closer to everyday spoken language, mainly reflecting the vocabulary and grammar of the spoken dialects of northern China. It became the vehicle for a body of prose literature, notably the celebrated Dream of Red Mansions (Dream of the Red Chamber). Despite its popularity, baihua never really matched the prestige of Classical Chinese, which remained the language of serious literature and government affairs.

Although based on the northern spoken dialects, baihua was accepted in all parts of the country. This included the south, where it was in many ways a foreign dialect due to differences in grammar and vocabulary. Significantly, none of the southern dialects developed its own written language.

In the 20th century, when China decided it wanted a standard national language, Classical Chinese was passed over in favour of baihua as the basis for the new standard. Initially, the new standard (known as guoyu or putonghua, often called Mandarin by English speakers) was in a similar position to Classical Chinese. People learnt how to write the standard language but pronounced it in their own local dialect. It is still possible to find older dialect speakers in China who can read and write standard Chinese perfectly but have difficulty speaking Mandarin. This situation can be summarised as follows:

Standard Written Language (based on baihua)
Standard pronunciation exists but in theory only
Hokkienese pronunciation
Wu pronunciation
Northern pronunciation
Cantonese pronunciation

This is still very much the situation in Hong Kong and Macao, where the written language is standard Chinese based on northern baihua but pronounced in Cantonese. Hong Kong news broadcasts represent this most graphically as they mostly use standard grammar and vocabulary (with the intrusion of some Cantonese-isms) spoken entirely with Cantonese pronunciation. This phenomenon is much less evident on the Mainland, where most TV and radio broadcasts are in Mandarin.

Perhaps not coincidentally, Cantonese speakers seem to be among the most vigorous proponents of the idea that Chinese is a language unified by its writing system. This underlies the often expressed view that the written language used in Beijing, Hong Kong, and Taipei is equally valid as a form of Chinese, regardless of how it is pronounced.

Unfortunately, this is not the model of Chinese that is envisioned by the Chinese government. The model that China is aiming for is this:

Standard Written and Spoken Language
Pronunciation based on Beijing phonology
Vocabulary based on northern dialects
Grammar based on baihua

There is no room for dialect pronunciations in this schema. There is only a single unified national language with a single standard pronunciation, based on Beijing.

The disregard for dialects in national language planning is manifest in the simplification of characters. Some characters have been unified on the basis of a common Beijing pronunciation even though the original characters are pronounced differently in dialect. An example is the character 里, which brings together three different characters. The pronunciation and meanings in putonghua and Cantonese are as follows:

Traditional Characters
Putonghua pronunciation
Cantonese pronunciation
Simplified Character
'village, neighbourhood, mile'
裏 (variant: 裡)

In amalgamating the three traditional characters into one simplified character, it is clear that only the Beijing pronunciation was taken into account. The distinction in dialects such as Cantonese was completely ignored.

This new concept of a standard spoken and written language has profound implications for the role of characters as a bridge between dialects. The old tradition of education using dialect pronunciation is still found in Hong Kong, where many dictionaries show the Cantonese pronunciation of characters and students are taught to read with Cantonese pronunciations. In the past this applied to other dialects as well. Some dialects (e.g., Hokkienese) had quite complex systems of dual literary and colloquial readings. Students were thus taught to explicitly link characters with dialect pronunciations.

The new concept of a standard language based on Beijing implies that only the Beijing-based pronunciation should be taught. If official policy is properly implemented, the role of characters as a unifying factor must logically come to an end. Students learning only the Beijing pronunciation will not learn the dialect pronunciations -- in other words, dialects will no longer be formally associated with characters.

The reality is that teachers continue to use the local dialect to a greater or lesser extent in the classroom. Moreover, connections will still be drawn between dialect words and characters. Speakers will always be able to make the analogies required owing to the large amounts of shared vocabulary and the many correspondences with dialect pronunciations. Shanghainese speakers will always notice, for instance that 日 ('day') in Mandarin is usually equivalent to niek23 in their own dialect. But it will be difficult to argue that this 'bridging function' is due to the role of characters.



Back to Top