Movable type, using a ceramic font, was also invented in China, by Bi Sheng between 1041 and 1048. Yet the Chinese did not consider movable type particularly useful, partly because the early ceramic type was easily broken, partly because the art of handwriting was so highly valued, and partly because any relatively complete font for Chinese requires thousands of characters.
The Chinese traded paper westward along the silk route, but the technique of papermaking was a jealously guarded secret until 751, when a number of Chinese papermakers in Samarkand were taken as prisoners of war by the Arabs. Samarkand soon became an important Islamic papermaking center. From there, knowledge of the process spread slowly westward, finally arriving in Europe in the twelfth century, a thousand years after its invention in China.
Figure 4.2 The printed Diamond Sutra, the oldest surviving woodblock-printed book in the world, dating to AD 868, during the Tang dynasty. The frontispiece illustration shows the Buddha surrounded by his disciples. Found in Buddhist temple caves at Dunhuang, China. British Library, London. Image copyright © Werner Forman/Art Resource, NY.
Until the early twentieth century, the language of most Chinese books was Classical Chinese, based on the written language described by Lù Fyán’s dictionary of AD 601. The spoken language of that period is now known as Middle Chinese. A great number of changes had already occurred from the language recorded in the first Old Chinese texts. It is also probable that the literary language had already diverged considerably from the spoken language and was written in an artificially terse and abbreviated style. Thanks to this artificial style, and the further changes that occurred in the development of Middle into Modern Chinese, Classical Chinese texts are now unintelligible when read aloud in modern pronunciation.
As Middle Chinese gave way to Modern Chinese, many distinctions of sound were lost, with the result that many of the syllables (and thus morphemes) began to be pronounced the same. The process did not occur uniformly in the different areas of China, leading to significant differences in pronunciation in the different Chinese “dialects” of modern China. Modern Chinese, and especially the Mandarin group of northern dialects, has lost a great number of sound distinctions that older versions of Chinese possessed, especially at the ends of syllables.
In Mandarin, the reduction left only about 1,300 separate syllables, less than a third of what Middle Chinese had had. Spoken language adapts, however. As the number of homophones – and thus the potential for misunderstanding – grew, so did the number of compound words. Rather than calling something by a single, ambiguous syllable, words grew to contain (usually) two syllables, with two morphemes reinforcing each other’s meaning. For example, bàogào, “to report,” is made of two morphemes, literally meaning “to announce–inform.” The addition of the reinforcing morpheme is important, because bào alone (in spoken, not written, form) can also mean “to embrace,” “a leopard,” “cruel,” “an abalone,” “a sudden rain,” or “to explode.” Together, bàogào is unambiguous. The script continues to be written on the syllable/morpheme level, so that many a word that is still ambiguous halfway through its spoken form is already disambiguated in written form by the character for the first morpheme/syllable.
Added to the changes in pronunciation since Middle Chinese are the inevitable grammatical differences between the older and newer forms of language, shifts in word meanings that have occurred, and classical allusions that the modern reader may fail to grasp. Furthermore, the artificially telegraphic classical style meant that only the first syllable would be written in words that may already have been bisyllabic in the spoken language. All in all, Classical Chinese is very difficult to master. Due to its rampant ambiguities, it is meaningless when read aloud or transliterated into Roman letters. Yet written in characters it makes some sense to a literate reader of Modern Chinese – it is the pronunciations that have changed, not the written characters.
Until the fall of the last imperial dynasty, the Qing (1644–1912), Classical Chinese was the standard for written Chinese. It was then replaced by a modern written form, which uses the grammatical patterns and compound words of modern spoken Mandarin, the dialect of Beijing. The same characters remained in use, but the ways of using them in sentences changed. The changes have made written Chinese much more accessible to ordinary people and have encouraged the spread of literacy.
Learning to read and write Chinese is both easier and harder than English. Each written symbol in Chinese is pronounceable, and children easily learn to read words, without the painful struggle to “sound out” letters that English-speaking children experience. The learning process continues longer, however. To read the first 90 percent of a typical text requires a knowledge of 1,000 characters. To read 99 percent requires 2,400 characters. To reach 99.9 percent takes 3,800; 99.99 percent requires 5,200. An educated reader who is not an expert in Chinese literature or history will know 3,500 to 4,000 characters; a scholar may know 6,000. Popular dictionaries list about 8,000. Unabridged dictionaries, however, tell an even more daunting story: every time a new morpheme enters the language and is written down, a new character is born. Thus dictionaries grow through the ages; recent ones list around 60,000 characters. For some of the more obscure characters, no one today knows either the meaning or the pronunciation. For others, the meaning is known but not the pronunciation.
For all that, there are many Chinese words that cannot be written at all. Colloquial expressions often have no written form, especially in the non-Mandarin dialects. So, for example, a writer from Shanghai will not be able to write using local colloquial vocabulary, but must use the vocabulary for which cognates exist in written Mandarin.
Users of alphabets are quick to see the disadvantages of Chinese characters. The system does not easily absorb new or colloquial expressions. It is hard to use characters to convey information about pronunciation, and they are difficult to organize into a dictionary, or indeed into any searchable database. This last point is becoming more important. These days, a system that does not easily allow the sorting of information is disadvantaged. Telephone books, dictionaries, encyclopedias, library catalogues, book indexes, mailing lists... such compilations of information are becoming more and more common, and more and more essential. As databases become larger, it becomes vital to have a sorting scheme, such as alphabetical order, that even a stupid computer can follow. The alphabet is a small, ordered set; you simply sort by the first letter of a word, then by the second, and so forth.
The latest IT revolution has not made Chinese logograms look good. Besides the difficulty of storing information for easy retrieval, there is the fact that nowadays communication is expected to occur instantaneously, via keystrokes typed into a computer in combinations of around 45 keys. Keyboarding is not easily done in Chinese characters. An earlier invention, telegraphy, was also poorly adapted to characters: telegraphs were sent as sequences of four-digit numbers. Each number represented a character, which had to be looked up in a list.
To overcome these obstacles, the People’s Republic of China has instituted an official Romanization: pnyn (see figure 4.3). School-children are taught pnyn, and then taught the standard Mandarin pronunciation of characters via pnyn. A person doing word processing can type in pnyn and have the computer convert it into characters (as I have done for the characters in this chapter; the Romanized spellings of Chinese words are also in pnyn). Language learners can use pnyn to learn the sounds of Chinese words, and nowadays dictionaries present characters in alphabetical order according to their pnyn spelling.
In Taiwan another script, known as bpmf or guóyn zìm, is used to indicate pronunciation. It is a system of 37 symbols derived from very simplified characters. The symbols are used purely phonologically and indicate a syllable’s initial consonant, its medial semivowel (if any), its ending (which may consist of a vowel, a diphthong, or a vowel followed by [r], [n], or []), and its tone, for a total of three to four symbols per syllable. Yet pnyn and bpmf are used as auxiliary scripts only,
and characters continue to be used by increasing numbers of people as the literacy rate grows.
In an effort to encourage literacy, in 1964 the government of the People’s Republic of China published a list of 2,238 characters which were given new, simplified forms. The result has been that Chinese living in Taiwan, in Hong Kong, and abroad now have trouble reading mainland publications, and vice versa. But the basic system of using characters remains strong.
Why, if they are so cumbersome, are Chinese characters still used? Why not just use pnyn (or bpmf)? One reason which looms large in the consciousness of Chinese people is the large number of homophonous syllables. Because written Chinese represents each syllable/morpheme separately, the Chinese are more oriented to syllables than to complete words as the units of their language. They are therefore very aware of the information that appears to be lost when characters are translated into pnyn. Distinct characters that clearly have different meanings get spelled the same in pnyn. The problem is not intractable, however. After all, the Chinese manage to communicate in speech perfectly well. They rely on context: the context of a morpheme within a word, and of a word within a sentence. Similarly, when a pnyn text is considered on a word-by-word basis, rather than a syllable-by-syllable basis, the ambiguities largely evaporate.
Figure 4.3 Chinese pnyn Romanization, with IPA equivalents. By tradition, Chinese syllables are divided into initial consonants and finals (everything after the first consonant), and the system of pronouncing pnyn follows this division. Tone markings are shown at the bottom, illustrated on the letter a.
The concern over homophony applies more aptly to the Chinese classics, which, thanks to their telegraphic style and the subsequent neutralization of sound distinctions, are now unintelligible when read aloud. They would be equally unintelligible in pnyn.
One of the consequences of the technology of writing is the very existence of classical texts. Once created, classical texts give a people a strong incentive to maintain backwards compatibility – to keep writing in an antique style, the better to emulate and to continue appreciating the classics. English is no exception: many of our odd spellings, such as could and might, derive from earlier versions of English, and much of the argument against spelling reform is that it would make older works inaccessible to modern readers.
Yet their writing system has a much stronger hold over the Chinese than their love of the classics. Chinese characters made China China. They fostered a unified Chinese civilization, which, far from being backward in the area of information technology, in fact produced the second great IT revolution (paper) and a significant part of the third (printing). A phonological system such as pnyn favors one form of Chinese (in this case, Mandarin) over the others much more strongly than the current standard of Mandarin-based logographic writing does. A phonological system could lead to the development of regional written languages – perhaps the end of the Han as a unified people.
Chinese writing developed hand in hand with Chinese civilization, and the Han enjoyed a cultural and literary monopoly in Asia for many centuries. When writing began to spread to other Asian peoples, they first wrote in Chinese, just as the Akkadians first wrote in Sumerian. The Koreans, Japanese, and Vietnamese all adopted Chinese characters, to write first in Chinese and later in their own respective languages. The Vietnamese switched to a version of the Roman alphabet in the 1940s, and the North Koreans have likewise abandoned Chinese characters; but the Japanese and South Koreans still use them in combination with their own native forms of writing, presented in later chapters.
Yet Chinese logograms are not particularly adaptable. A Chinese logogram stands for a particular morpheme in Chinese. Another language can in principle use logograms for morphemes of corresponding meaning, but that relies on the language having morphemes that match the meaning of Chinese morphemes (and vice versa). Relatively few languages fit the bill. Chinese has very little inflectional morphology, so that , m, for example, can mean “horse,” or “horses,” depending on context. Languages that add morphemes (like the English -(e)s) to distinguish the two words cannot be properly represented using only Chinese characters.
Rather than try to adapt Chinese characters to their language, some neighbors of the Chinese chose to create their own scripts. Just as Mesopotamian cuneiform inspired the creation of similar-looking scripts such as Old Persian and Ugaritic, so Chinese logograms inspired the creation of Chinese-looking (“siniform”) scripts in Asia. Scripts were invented by various peoples living to the north or northwest of the Han – by the Khitan in AD 920, the Tangut in 1,036, and the Jurchin in 1,120. However, none of these scripts enjoyed a long lifespan. More recently scripts have been invented for some of the minority languages of southwest China, such as Lisu and Yi, but these are not widely used. A syllabic “women’s script” has traditionally been used in part of Hunan province, but has largely gone out of use with the expansion of women’s education in China. Chinese characters still dominate East Asia.
The Chinese script, the most stubbornly logographic of the ancient invented forms of writing, is the only one of those early writing systems still in use today. Used for Chinese as well as (to varying extents) for Japanese and Korean, it encodes the languages of a quarter of the world’s population and continues to hold together the Middle Kingdom.
5
Maya Glyphs: Calendars of Kings
The Earth’s geography played a strong supporting role in giving the world a script so distinctive and visually complex that its nature as a true writing system was widely misunderstood until the late twentieth century. The Bering land bridge connecting Siberia and Alaska had allowed human migration to the New World in Paleolithic times. But with the end of the last ice age the land bridge was submerged, leaving today’s Bering Strait, and the New World was cut off from the Old World until the voyages of Leif Erikson and Christopher Columbus. In the absence of any outside stimulus, the New World fostered its own civilizations and its own scripts. While the civilizations of Peru used nongraphical record-keeping systems, such as the Inca’s quipu system of knotted cords, those of Mesoamerica (a cultural area stretching from southern Mexico to northwest Costa Rica) developed a number of graphical recording systems. Some of these systems were developed to the point of true writing while others remained conventionalized pictographic systems that were used to record certain types of information but not the full language of the speakers.
Of these Mesoamerican scripts and proto-scripts, the most advanced – and the most amply preserved for modern epigraphers – was the hieroglyphic system of the Maya. Carved on monuments in the cities of the Classic Maya and painted on their fine polychrome vases, the script was also used to make books through Post-Classic times until the Spanish conquest. It was then neglected, actively repressed, and ultimately forgotten. But the carved monuments remained, increasingly hidden by the encroaching jungle, to enthrall later generations of explorers, art historians, and decipherers.
The ultimate origin of the Maya script lies with the cultural predecessors of the Maya, the Olmec. The Olmec were the first civilization to arise in Mesoamerica, flourishing between 1200 and 400 BC (see map in the appendix, figure A.5). A number of distinctive aspects of common Mesoamerican culture had their roots in Olmec traditions: all the successors of the Olmec used a distinctive 52-year calendar (with local variations), and many of them developed writing or advanced pictographic systems that verged on true writing. The various Mesoamerican scripts are quite diverse, but they share a bar-and-dot system of numerals and the signs for the names of the calendar days. As the very earliest surviving inscriptions record almost exclusively dates, it seems likely that all the Mesoamerican scripts evolved from a calendrical system of the Olmec period.
In fact, the use of a complex calendar seems to have been the intellectual stimulus for developing writing in Mesoamerica. The 52-year Calendar Round consisted of two pieces, the Sacred Round and the Vague Year. The Sacred Round was in use before 600 BC and is still used among the Maya
today. Its 260-day ritual cycle pairs 13 numbers with 20 named days. Each day that dawns has a number and a name associated with it; the following day will have the next day number in the sequence, coupled with the next name. Thus 1 Imix is followed by 2 Ik, which is followed by 3 Akbal – not unlike the way Monday the 1st (of some month) is followed by Tuesday the 2nd and Wednesday the 3rd. The fourteenth day of the sequence has the number 1 again, but not the name Imix, as the list of 20 day names will not yet be exhausted. Rather, it will be 1 Ix, followed by 2 Men, and not reaching Imix until the eighth day. It takes 260 days to run the cycle all the way around from 1 Imix through to the last day, 13 Ajaw, arriving back at 1 Imix on the two-hundred-and-sixty-first day.
Running concurrently with the Sacred Round was the calendar of the Vague Year, so called because it contained 365 days, nearly a true solar year. The Vague Year contained 18 named months of 20 numbered days each (progressing in a way more familiar to us – Pop 1 decorously followed by Pop 2, Pop 3, etc.), with an extra inauspicious five days at the end. Just as it took several individual cycles of the names and numbers of the Sacred Round to complete a full cycle from one 1 Imix around to the next 1 Imix, so the combination of the Sacred Round and the Vague Year – the time it took for a date in the Vague Year to correspond to the same number and day in the Sacred Round again – took many turns to complete, a total of 52 years. A Calendar Round date will thus repeat after 52 years, and inscribed dates using only the Calendar Round must be interpreted with the help of other archaeological evidence to determine which 52-year cycle they occur in.
The Writing Revolution Page 10