The Silk Road: A New History
Page 23
Another Manichaean text even more closely mimics a Chinese text. The prologue of the small scroll sounds exactly like the prologue to the famed Buddhist text of the Diamond Sutra. Yet in this version, Mani—not the Buddha—addresses a disciple: “Good indeed! Good indeed! In order to benefit the innumerable crowds of living beings, you have addressed to me this query, profound and mysterious. You thus show yourself as a good friend to all those living beings of the world who have blindly gone astray, and I will now explain the matter to you in detail, so that the net of doubt in which you are ensnared may be broken forever without recall.”45 Even the title misleads: the text is called The Compendium of the Teachings of Mani the Buddha of Light. The text so closely resembled a Buddhist text that it fooled even an expert like Pelliot, who chose not to take it to Paris, and today it is one of the most important texts in the Beijing Library Collection. Sogdian missionaries prepared this translation in response to an imperial order issued in 731; they hoped to convert the Chinese emperor himself.
The missionaries of each church took different approaches to translation. Whereas the Manichaeans freely adopted Buddhist terms, the Christians from the Church of the East insisted on strict accuracy in translation, no matter how confusing the result might be.46 What was the best way to render “God the Father, the Son, and the Holy Ghost,” into Chinese? The translator of the hymn “Gloria in Excelsis Deo” chose the most literal solution: “Merciful father, Light son, King of the pure wind.” Of the three terms, only “merciful father” would have made sense to a Chinese convert. On the same sheet of paper is a list of holy books entitled “The Book of Honor.” It states that the three bodies of the “father emperor,” the “son emperor,” and the “witness” all form a single body—i.e., the Holy Trinity—another teaching that would certainly have puzzled a Chinese audience.47 A note at the end of this list refers to Jingjing (or Adam), the author of the Christian stele of Ch’ang-an, indicating that it, like the stele, was composed sometime in the late eighth century, a period when the Church of the East was active in China.
The nature of the materials in the cave changed quite noticeably in the mid-eighth century. Before the An Lushan rebellion, almost all of the texts in the cave came from central China and consisted of Buddhist texts. The latest text in the cave from Chang’an is dated 753; after that date, all the texts were locally produced.48 At this time, lay students began to copy a much larger variety of material, including—in addition to Buddhist texts—contracts, charters for lay associations, and literary texts. They even doodled in the margins of texts.49 One hand copy of a market certificate, written sometime between 742 and 758, records the purchase of a thirteen-year-old non-Chinese slave boy for twenty-one bolts of raw silk. Adhering exactly to the detailed regulations of the Tang Code, it lists the names and ages of the seller, slave, and five guarantors, confirming that the Tang enforced the code throughout its realm.50
In 745 the Tang central government sent a payment of fifteen thousand bolts of silk in two installments to a garrison near Dunhuang.51 A bureaucratic document about someone’s salary allows us to see exactly how the Tang state made such payments. The central government deposited two shipments of silk in a commandery in Liangzhou (modern Wuwei, Gansu), about 435 miles (700 km) east of Dunhuang, the location of the regional military headquarters. From there, the silk was shipped to the Dunhuang garrison. As the French scholar Éric Trombert astutely remarks, “One has here a concrete example of two military convoys, each carrying more than 7,000 bolts of silk, that has nothing in common with the images of caravans of private merchants to which we are accustomed.”52 These individual payments of seven thousand bolts of silk are much higher than all the individual transactions recorded in the Turfan documents, which involved at most a few hundred bolts of silk. This document shows just how important the central government’s payments to the military were.
The Tang government had a complex monetary system in which three different currencies—textiles (both hemp and silk), grain, and coins—circulated alongside one another. Confusingly, the central government used a single aggregated unit to represent all three goods. The payment to the Dunhuang garrison included six different types of woven silk and silk floss. Because different localities paid their taxes using locally produced cloth, the Tang authorities simply transferred those textiles to the Dunhuang garrison. Garrison officials converted the tax cloth first into coins and then into grain, some used to feed the soldiers in the garrison, some paid directly to local merchants. This record affords a rare glimpse of payments sent to the military before the An Lushan rebellion: the Tang government injected massive amounts of money—in the form of woven cloth—straight into the Dunhuang economy.
As earlier chapters have recounted, in 755 the central government lost control of the northwest. In an attempt to defeat the rebels, the Tang emperor turned to the rulers of the Tibetan Empire for help. The Yarlung dynasty of Tibet was a relative newcomer to Central Asian politics. Before 617, the Tibetan plateau, some 13,000 to 15,000 feet (4,000–5,000 m) above sea level, was home in the north to herders who raised horses in the grasslands and, in the south, to farmers who planted barley in the river valleys.53 With no indigenous writing system, they knotted cords and marked tallies as their means of record keeping. Around 617 the rulers of the Yarlung dynasty, who took their name from the river valley to the southeast of Lhasa, unified Tibet for the first time. They modified the Sanskrit alphabet to form their own writing system, and, at the same time, adopted elements of the Tang legal system.
In 755, after An Lushan rebelled, the Tang emperor wrote to the Tibetans, promising large payments in exchange for their help in suppressing the rebels. The Tibetans were fine horsemen, and the Chinese admired their military equipment. As the official history of the Tang explains, “Their armor is excellent. They clothe their entire body in it, except for their two eyes. Even powerful bows and sharp knives cannot harm them very much.”54 Although ostensibly in the service of the Tang, Tibetan soldiers raided the capital of Chang’an for two weeks in the fall of 763 before retreating. Each autumn until 777 the horsemen returned to plunder Chang’an, and the weakened Tang armies could not keep them out.
During the 760s and 770s, when they were at peak strength, the Tibetans gradually increased the territory under their control and expanded into Gansu. In 781 they conquered the town of Shouchang, south of Dunhuang, and in 786, when the Tang government failed to pay the stipulated amount for their assistance in putting down a rebellion, they seized the prefectural center of Dunhuang. Governing eight former Tang-dynasty prefectures in the Gansu corridor, the Tibetans appointed a council of generals to rule the military districts. The Tibetans immediately established a dual administration headed by a Tibetan military governor and the highest civil official, in Dunhuang, who was often Chinese. Each district was further subdivided into units of one thousand, and these into twenty units of fifty households. The head of the smaller fifty-household units assigned each household tasks so that they could fulfill their labor obligation to the state.55
Some male residents in the Tibetan-occupied territory were conscripted into the army, while others labored in military colonies. In addition to serving as guards, those in the colonies cultivated crops and paid agricultural taxes in grain, which they had to carry to collection points, sometimes several days’ travel away. The Tibetans staffed their army using corvée labor; they did not pay their soldiers with bolts of cloth, grain, and coins as the Tang dynasty had.
The imposition of Tibetan rule in Dunhuang had an immediate effect on the local economy, as contracts written in both Tibetan and Chinese show.56 In 788–790, a few years after the Tibetans took Dunhuang, the financial records of a storehouse mentioned coins; this is the latest known Chinese-language reference to coins.57 It is possible that some Chinese coins, perhaps those minted before 755, circulated in the ninth and tenth centuries, but under Tibetan rule coins largely dropped from use. During the Tibetan period, prices are given in either meas
ures of grain or bolts of cloth.58 A representative contract, dated 803, documents the sale of a cow for a price of twelve piculs of wheat (20–30 bushels, or 720–1080 L) and two of millet (3.5–5 bushels, or 120–180 L). The penalty for breach of contract is also denominated in grain: three piculs of wheat (5–8 bushels, or 180–270 L).59 With only a few mentions of dmar (the Tibetan word for “copper,” which probably indicates bronze coins), the contracts record exchanges almost entirely in grain.60 Sometimes people borrow cloth or paper, but they always pay back their debts using grain.
Earlier analysts saw the Tibetan occupation from 786 to 848 as a brief interlude in Dunhuang’s history, with few lasting repercussions. Sixty years, however, was sufficiently long that the residents of Dunhuang adopted some Tibetan customs. In the early years of Tibetan rule, most Chinese followed Chinese practice and used a family name followed by a given name. But over time the Chinese residents of Dunhuang adopted more and more Tibetan-sounding names. By the second or third generation of Tibetan rule, some even gave up their use of Chinese family names and used only a first name, just as the Tibetans did.
Some Chinese living under Tibetan rule made an even bigger change: they stopped writing in Chinese and adopted the Tibetan alphabet. Immediately following the Tibetan conquest, local scribes learned the language in order to draft government documents for officials or contracts for Tibetan speakers. Between 815 and 841 the Tibetan military governor launched a large-scale initiative to copy Buddhist texts. The project employed over one thousand scribes, many of them Chinese.61 As they copied texts, these scribes grew more proficient in writing Tibetan. They realized that it was easier to use a phonetic alphabet than to memorize thousands of Chinese characters.
As the rulers sponsored the mass copying of Buddhist sutras in the hope of generating merit, so too did they finance the construction of new caves. The sixty-six caves built in the Tibetan period have several distinguishing characteristics: they often feature mandalas, or diagrams of the cosmos, among other elements from esoteric Buddhism. Cave paintings from this period also grant greater prominence to the donors, especially the Tibetan emperor.62
During the Tibetan period, Dunhuang artists painted Mount Wutai, and they continued to do so in the tenth century, when Dunhuang came under Cao-family rule. One of the most magnificent caves at Dunhuang, cave 61, dates to just around 950.63 The upper section of the entire western wall of the cave, a section measuring 11.5 feet (3.5 m) tall by 51 feet (15.5 m) wide, shows the Mount Wutai pilgrimage site in Shanxi Province. The top of the painting shows a heavenly assembly, while the middle shows ninety different buildings at Mount Wutai, each bearing a label, and the bottom shows travelers on their way to the mountain. The painting is not an accurate map of the pilgrimage site; it probably was intended as a guide for viewers unable to visit the site themselves. The cave’s patrons included Cao Yuanzhong, the ruler of Dunhuang from 944 to 974, and his wives, one of whom was from Khotan.
Despite occasional outbreaks of armed conflict, the rulers of Dunhuang maintained contact with both the Tang and India during the Tibetan period. Monks and envoys, dispatched by the rulers of Tibet, China, and India, traveled between Tibet and China, and they often stopped at Dunhuang during their travels. The lack of circulating currency did not prevent envoys and monks from proceeding from one oasis to the next; much as they had in earlier periods, rulers provided emissaries with escorts, transport, and food.
In 848 a Chinese regime reestablished control at Dunhuang, but the use of Tibetan persisted. Earlier generations of scholars assumed that the Tibetan-language materials in cave 17 must have been written before 848; more recently scholars have come to realize that Tibetan remained in use as a lingua franca even after 848.64 Under Tibetan rule, the pilgrimage route from Tibet to Mount Wutai, which passed through Dunhuang, saw an increase in traffic. Also from the library cave are copies of five letters of introduction, in Tibetan, for a Chinese monk traveling to Tibet; they, too, date to the period after 848, when Chinese armies expelled the Tibetans from Dunhuang.65 The letters explain that the monk is going to India to study at the great Buddhist center at Nalanda and to obtain relics. Beginning his journey at Mount Wutai, the monk visited several towns on the way to Dunhuang, where he left the letters behind, presumably because he did not need them in Tibet.
Another Tibetan-language manuscript was dictated by an Indian monk to a Tibetan disciple who could understand some Sanskrit but made many spelling mistakes. The document explains that in 977 (or possibly 965), the Indian monk Devaputra traveled from India to Tibet and then to Mount Wutai; on his way back, near Dunhuang, he bestowed his teaching on a disciple. The text gives many technical terms in Tibetan, followed by the approximate Sanskrit spelling.66 Tibetan monks encouraged the study of Sanskrit, possibly because their own alphabet was modeled on Sanskrit, making it more accessible.67 At least some monks must have spoken Sanskrit to communicate, most likely with other learned monks, in monasteries just as Xuanzang did on his way to India.
In 842, when the confederation of peoples who had supported the rulers suddenly broke apart, the Yarlung dynasty collapsed, and Tibetan control over Dunhuang weakened. In 848 the Chinese general Zhang Yichao organized an army that expelled the remaining Tibetans.68 The Tang dynasty was much weaker than it had been before the An Lushan rebellion. Even within central China (the area often called “China proper,” which includes the Yangzi, Yellow, and Pearl River valleys), the Tang had ceded political control to military commanders who collected taxes and financed their own armies, sometimes sending some revenues to the center. Like these commanders, Zhang Yichao received the title of military governor from the Tang court in 851. Zhang pledged allegiance to the Tang dynasty but in fact governed Dunhuang as an independent kingdom. Under Zhang family rule, Dunhuang sent envoys to the Tang capital at Chang’an to present tribute to the Tang emperor, much as other independent Central Asian rulers did.
Zhang did not gain complete control in 848; his armies fought those of the Tibetans again in 856, as a literary text entitled “Zhang Yichao Transformation Text” relates. Of all the literary genres represented in cave 17, transformation texts, which alternate passages of prose with poetry, are the most distinctive. Transformation texts combine stretches of sung poetry and recited prose (this literary genre existed in Kuchean, too). The library cave preserved thirty or so examples of Chinese-language transformation texts; they survive nowhere else.69 Most broadly, the term “transformation” refers to the different transformations of creation; monk storytellers performed these tales to help their audience members escape from the cycle of life, death, and rebirth from which all Buddhist teachings offered escape. Transformation texts contain a tell-tale formula: “Please look at the place where [a specific event] happens; how does it go?”70 As they narrated these tales, storytellers pointed to scenes in scroll paintings so that the audience could picture the events they described.
The Zhang Yichao transformation text, which describes several battles in 856 between Zhang’s army and the tribes fighting on the Tibetan side, sets the scene using words:
The bandits [the Tibetan coalition] had not expected that the Chinese troops would arrive so suddenly and were totally unprepared. Our armies proceeded to line up in a “black-cloud formation,” swiftly striking from all four sides. The barbarian bandits were panic-stricken. Like stars they splintered, north and south. The Chinese armies having gained the advantage, they pursued them, pressing close at their backs. Within fifteen miles, they caught up with them.
Then, just as the narrator points to a scene showing the army, he says, “This is the place where their slain corpses were strewn everywhere across the plain.”71 None of these scroll paintings survive, but a cave painting done in 861 does show Zhang’s army on parade.72
This cave was completed in 865. Four years earlier, General Zhang Huaishen, the nephew of the Chinese ruler Zhang Yichao, began work on the cave, the first to be financed by a member of the ruling Zhang family. The inscription exp
lains that Zhang Huaishen
had a strong desire to carve a cave. He looked around the whole area, but there was no place at all, except for a single cliff, where cutting was possible. Undaunted by the enormity of the work to be accomplished, his spirit was concentrated to the point where it could pierce stone, his purpose strong enough to move the mountain.
ZHANG YICHAO’S ARMY
Zhang Yichao’s troops carry fluttering banners. Some figures wear plain robes favored by the Chinese, while others are dressed in brightly patterned fabrics often worn by Uighurs and other non-Chinese peoples. The painting offers a glimpse of the diverse peoples who supported Zhang-family rule. Drawing by Amelia Sargent.
Then he prayed to the heavenly spirits above, gave thanks to the earth spirits below, divined to find an auspicious time, and calculated the day for the work to start. The cutting and chiseling had hardly begun, when the mountain split of its own accord; not many days had elapsed, when the cracks opened to a hole. With further prayers and incense, the sands began to fly, and early in the night, suddenly and furiously, with a fearful rush, there was the sound of thunder, splitting the rock wall, and the cliff was cut away.73
The author offers a step-by-step account of how to make a cave: workmen started by digging a crack, which they gradually expanded until it could hold wall paintings and statues. The process of digging a cave was labor-intensive but did not require the use of expensive materials. Local artists lived on site, in the northern cave complex, where archeologists have found numerous artists’ workshops, some complete with pots of paint.74 In the ninth century most artists belonged to local workshops, and by the mid-tenth century the local government had established a painting academy staffed by artist-officials.75