UBC Press Blog

Edwin G. Pulleyblank, 1995, Introduction, p. 3-5.

[Note: Due to technical difficulties, Chinese characters present in the text are not reproduced in this excerpt. We apologize for the inconvenience.]

Chinese was the principal vehicle of culture and civilization for the whole of East Asia for many centuries and today is spoken by more people than any other language. The earliest known examples of written Chinese are the so-called ‘oracle bones,’ records of divination from the last capital of the Shang dynasty at Anyang. They date from approximately -1300 to -1050. From the following centuries, after the founding of the Zhōu dynasty, come inscriptions on bronze vessels recording royal donations and other such events. The earliest of the Chinese classics — parts of the Book of Changes (Yjiīng), the Book of Documents (Shūjīng) and the Book of Odes (Shījīng) — also date from the early centuries of the Zhou dynasty. All these texts are written in an archaic form of Chinese referred to as preclassical.

The classical period proper begins with Confucius (-551 to -479) and continues through the Warring States period to the unification and founding of the empire by Qín in -221.  This was the period of the major philosophers and also of the first works of narrative history. Though all the productions of the period are in Classical Chinese, there is considerable linguistic diversity among them. This is, no doubt, partly the result of the geographical disunity and decentralization of the country, which allowed various regional dialects to become the vehicles of literature in their own areas. It is also the result of historical evolution. Exhaustive studies of these differences have yet to be made, but one can distinguish at least the following: (a) a rather archaic form of literary language, showing features in common with the Shījīng and probably based on a central dialect, used in historical texts such as the Zuŏzhuàn and Guóyù ; (b) a Lŭ   dialect used in the Confucian Analects (Lúnyŭ; more archaic) and Mencius (Mèngzĭ; more evolved); (c) a Chǔ dialect used in the Lí Sāo  and other early poems of the Chǔcí; and (d) a third-century dialect found in texts such as Zhuāngzĭ, Xúnzĭ  and Hān Féizĭ, showing an evolution towards a common literary standard but still with marked differences between different texts.
With the imperial unification under Qín and Hàn towards a common literary standard was accelerated, not only by the centralization of the government, but also by the increasing tendency towards imitation of classical models in preference to the living spoken language. An important influence in this respect was the triumph of Confucianism which made the Confucian classics the basis for education and for advancement in government service. In a comparatively early text like the Records of the Historian (Shĭjì) one can still detect influence from the spoken language, but as time went on Literary Chinese (wén yán) became increasingly a dead language, playing a role like that of Latin in Western Europe, from which the current spoken language increasingly diverged.
Literary Chinese was never completely static and uniform. Different styles were fashioned by successive literary movements and for special purposes such as government documents or Buddhist writings. There was no development of a prescriptive grammar and people learned to write by imitating earlier models rather than by obeying explicit rules as in the case of Latin. The spoken language always had some influence even in belles lettres and poetry, and still more in writings of a more practical nature. The result is that even those well versed in classical texts may have difficulty when they first encounter later material, such as official documents of the Qīng dynasty.

Chinese characters are sometimes referred to as if they directly represent ideas. This is a fallacy. Even though many of them are pictorial or otherwise iconic in origin, in their use as a system of writing they are conventional symbols for particular spoken words. Thus synonyms (words that are the same in meaning but different in sound) are normally written with different characters, while homophones (words that are the same in sound but different in meaning) may be written with the same character. For example, quăn ‘dog’ is written  , based on a pictogram for ‘dog,’ but gŏu,  which also means ‘dog,’ is written  , with a distorted form of   + gŏu   ‘hook’ to represent the sound. On the other hand, ān ‘how? where?’ and ān ‘peace’ are both written   .

Since in Chinese, as in every other language, the spoken form is primary, it is desirable to get back, as closely as possible, to the actual sounds that underlie the characters. Unfortunately, since the characters represent whole syllables and give no direct phonetic information, and since the sounds have changed greatly over the centuries, this is only possible through a difficult process of reconstruction. The most widely used system of reconstruction is that of Bernhard Karlgren as published in Grammata Serica Recensa (1957). This gives two reconstructions, one for what he calls Ancient Chinese, based on the Qièyùn, a rhyme dictionary of +602, and one for what he calls Archaic Chinese, based on the rhymes of the Shījīng, relevant to a period terminating around -600.
A revised system of reconstruction for the Qièyùn, called Early Middle Chinese (EMC), together with a reconstruction for Late Middle Chinese (LMC) of the Táng period, which together replace Karlgren’s Ancient Chinese, is published in Pulleyblank, Lexicon of Reconstructed Pronunciation in Early Middle Chinese, Late Middle Chinese and Early Mandarin (1991), which also contains a new reconstruction of Early Mandarin (EM) of the Yuán period.
The reconstruction of stages earlier than EMC is a much more difficult problem since the available evidence is more fragmentary. While the rhyme patterns of the Shījīng, worked out by scholars of the Qīng period, and the rhyming of poets at various periods between then and the Qièyùn provide evidence for the evolution of the finals, that is the rhyming parts of syllables, comparable systematic evidence for the non-rhyming parts, the initial consonants or groups of consonants, is lacking. Anything that purports to be a complete reconstruction of Old Chinese (OC), such as Karlgren’s Archaic Chinese, is bound to be somewhat illusory at the present time. In this Outline, reconstructed readings in EMC or LMC will be given from time to time for illustrative purposes. Tentative reconstructions in OC will also sometimes be given, marked with an asterisk *.
Apart from systems of reconstruction which propose actual phonetic values, there are some traditional methods used by commentators for indicating how characters should be read that readers of classical texts should be aware of. These are the traditional spelling system known fănqiè and the system of indicating the four ancient tones by small circles at the four corners of characters.

About the Author

Edwin G. Pulleyblank is Professor Emeritus of the Department of Asian Studies at the University of British Columbia.