The Silk Road: A New History

by Valerie Hansen

  Yoshida Yutaka, a professor of linguistics at Kyoto University, has painstakingly identified four different places where different Khotanese-language documents were found. Two were at Dandan Uiliq; one contains documents dated between 777 and 788 that mention Sidaka.39 As this and other documents from the same spot show, Khotan had a Chinese-run government between 777 and 788, the years in which documents with Sidaka’s name appear. During these decades, the Tibetans seized on the weakness of the Tang central government and aggressively expanded into Central Asia. After conquering Dunhuang in 786 they fought the Uighurs in Turfan for three years starting in 789 before defeating them in 792, and they conquered Khotan before 796.40 Historians of the Western Regions have known that the Tibetan Empire collapsed from within during the 840s; during the same decade, the Kirgiz defeated the Uighur Empire based in Karabalgasun, in today’s Mongolia, and many Uighurs fled south to modern Xinjiang. The newly discovered documents from Dandan Uiliq make it possible to know which oases fell to the Tibetans or to the Uighurs in precisely which years.

  Mazar Tagh was a military fortress on a strategically important route between Kucha and Khotan, which lay 90 miles (150 km) to the south of the fort. It also happened to be in an uninhabited part of the desert; Khotanese cooks and guards served there on fixed rotations.41 With the conquest of Khotan, the Tibetans gained control of the outpost, which had originally housed Chinese soldiers. One document dated 798 urges the official receiving the document to evacuate the men and cattle in the fortress to a nearby town. The document does not name the enemy, but it is probably the Uighur Kaghanate, which had occupied Kucha around 800.42

  The Tibetans left much of the previous administration in place; several named individuals continued in office before and after the imposition of Tibetan rule. They issued orders in both Khotanese and Chinese, an indication of how deeply Chinese bureaucratic practices had influenced the Khotanese and then the Tibetans.43 A few officials continued to use individual Chinese characters as their signatures. Scribes drafted contracts that translated multiple Chinese phrases literally into Tibetan. Those contracts, although never used within Tibet, established models for Tibetan-language contracts at Dunhuang.44 The Tibetans ruled Khotan indirectly; whenever they needed anything, the top official, the commissioner, made a request to his Khotanese counterpart, who then issued the order to the appropriate local officials.45

  As rich as they are, the tax materials from Dandan Uiliq do not reveal who was traveling along the Silk Road and why. One of the most illuminating documents about cultural contact on the Silk Road surfaced because Stein’s men continued to dig on their own even when he did not pay them. After excavating Dandan Uiliq for seventeen days (from December 18, 1900, to January 4, 1901) Stein dismissed some of the crew and took the remaining men to a nearby site, some 7 miles (11 km) away.

  When Stein returned to his camp on the first evening, he was startled to see some of the dismissed workers waiting for him. He was even more surprised when they presented him with their finds: near the corner of Dandan Uiliq Structure 13, they had uncovered a crumpled document with Hebrew letters on it. Stein explains why he believed his men had found it where they said they had; the paper was genuinely old (eighth century), and making and planting a forgery would have required a great deal of preparation.46 Stein was particularly leery of fakes because he had just exposed the fraud of Islam Akhun, the man whose skillfully forged documents tricked Hoernle into thinking that yet another new language had been discovered.47

  While at Dandan Uiliq, Stein had cleared the sand from Structure 13, where Turdi reported finding multiple silver ingots worth two hundred rupees, or thirteen pounds British sterling (with a rough value of perhaps one thousand pounds today) in his youth.48 Even though the structure was big, measuring 60 feet (18 m) on one side, with one room 22 by 18 feet (6.7 by 5.5 m), Stein decided not to excavate when his crew found nothing but a fireplace and a wooden frame.49 After Stein departed, the dismissed workers dug through a pile of waste, left near the ruin by earlier looters, and recovered the document written in Hebrew letters.

  The letter was in New Persian, the language that replaced Middle Persian in Iran during the ninth century. A handful of Jewish-Persian documents have been found at various locations around the world, near Herat, Afghanistan, the Malabar Coast of south India, and Baghdad. The Dandan Uiliq document is not the oldest such example—the oldest dates to the 750s—but it is among oldest Jewish-Persian documents surviving today.50

  The fragmentary letter is difficult to make out because only the center of the page remains; the words at the beginning and end of each line are gone. The author writes to a business associate, apparently his superior, about different transactions involving sheep, clothing, spikenard (a plant used in medicine and scents), a saddle, stirrups, and straps. Most likely a merchant, he mentions wanting to know his “profit and loss.” We do not know why he left Iran, but we can speculate that he (or his ancestors) moved east to escape the Islamic conquest, and he ended up in the Khotan region during a particularly turbulent time.

  The letter provides direct testimony that at least one Persian-speaking Jew was in Dandan Uiliq in the late eighth century, but the fragmentary state of the letter makes it difficult to say more. More than a hundred years after the initial find, the utterly unexpected occurred: a second Jewish-Persian letter, almost undamaged, appeared for sale, and the National Library of China purchased it. Zhang Zhan, a graduate student who earned his master’s degree at Peking University and who is working on his doctorate at Harvard, published a full transcription and translation into Chinese in 2008; he plans to bring out the English translation in the near future.51

  The similarities with the old letter are so great that Zhang Zhan is confident that the two letters were written at the same time, in the same place, and by the same person: in the opening years of the ninth century in Khotan. His dating hinges on a sentence in the letter reporting the latest news from Kashgar: “The Tibetans have all been killed.”52 If, as comparison to several Khotanese letters also found at Dandan Uiliq suggests, the letter refers to the Uighur defeat of the Tibetans, then the letter dates to 802, when the Uighurs took Kashgar and expelled the Tibetans.

  The new letter opens with eight lines of greetings by the writer “from far away” to the family of his “lord master Nisi Chilag,” the recipient of the letter, who was most likely a Jew living in Dandan Uiliq. The new letter then details a dispute about sheep that the writer had with a “landlord”: although he gave presents including musk and candy (the meanings of the other commodities he used as gifts are so far unknown), he has not yet received the sheep due to him. Interestingly, he relates a conversation in which the landlord mistook him for a “Sogdian,” an understandable error given how many of the merchants along the Silk Road were Sogdian and how few were Jewish.

  Jewish merchants have left only a few traces on the Silk Road. Recall that one of the latest inscriptions on the Karakorum Highway was in Hebrew. An Arabic account of a late ninth-century massacre mentioned Jewish merchants living in the southern city of Guangzhou (Canton) alongside Muslims, Christians, and Zoroastrians. And cave 17 contained a folded sheet of paper with an eighteen-line prayer as well as a selection from the Psalms in Hebrew (shown in color plate 12).

  In addition to the single Hebrew prayer sheet and tens of thousands of Chinese and Tibetan-language documents, the library cave at Dunhuang contained about two thousand Khotanese-language documents, many of them fragments.53 Like many peoples living in small countries surrounded by powerful neighbors, the Khotanese were adept at learning languages. One Khotanese scribe was so skilled that he copied Tibetan texts; we know that he was Khotanese only because he numbered the pages using Khotanese numerals.54 How did the Khotanese pick up languages so quickly even though they had no dictionaries?

  A NEWLY DISCOVERED JUDEO-PERSIAN LETTER

  In the early ninth century, the author of this letter, most likely a Persian-speaking Jew, used the Hebrew alphabet to
write in New Persian to another Jew living in Dandan Uiliq. The author describes a dispute he is having with his landlord, who has not paid the sheep he owes him. Courtesy of National Library of China.

  Several sheets from phrasebooks in Khotanese and Chinese were preserved in the Dunhuang library cave.55 Dispensing with Chinese characters, these learning aids simply record the sound of the Chinese sentence using Brahmi script and then give the meaning in Khotanese. The extremely skilled philologists who have worked on these texts have reverse-engineered the Khotanese version of tenth-century Chinese pronunciation to reconstruct the original Chinese sentences. Like any good language textbook, the Chinese-Khotanese bilingual list repeats key sentence patterns, all very short, so that the student can practice:

  Bring me vegetables!

  Bring me cucumbers!

  Bring me a gourd!

  The phrasebook also includes sentences one can use at a market when buying and selling. Given the many contacts between Khotan and Dunhuang in the tenth century, it seems likely that Khotanese from different walks of life—envoys, monks, merchants—could have benefited from basic instruction in Chinese.

  In contrast, the Sanskrit-Khotanese bilingual texts aim at a narrower audience.56 Sanskrit was easy for Khotanese speakers to learn; because it was written in the same Brahmi script, language-learners could simply copy Sanskrit sentences and commit them to memory. The Khotanese-Sanskrit phrasebook, 194 lines total, begins with a simple conversation:

  How are you?

  Good, thank you!

  And how are you?

  Where do you come from?

  I come from Khotan.

  The dialogues mention other places, too: India, China, Tibet, and Ganzhou (the seat of the Uighur Kaghanate in what is now Zhangye, Gansu). The phrase-book teaches how to buy a horse and obtain fodder, how to ask for needle and thread, and how to request that clothes be washed. Some of the dialogues hint at conflict:

  Don’t be angry with me.

  I shall not pull out your hair.

  When you speak unpleasantness

  then I shall become angry.

  And some even mention sex:

  He loves many women.

  He makes love.

  Some of the dialogues make it possible to identify the intended users:

  Do you have any books?

  I do.

  [Which books?]

  Sutras, Abhidharma, Vinaya, Vajrayana.

  Which of them do you have?

  Which of them do you like?

  I like to read Vajrayana.

  Only a monk or an advanced student of Buddhism could use such sentences. “Sutras” was a general term for all Buddhist texts, “Abhidharma” referred to the doctrinal texts, “Vinaya” to Buddhist regulations, and “Vajrayana” to Tantric texts. Sanskrit was spoken in monasteries all the way from China to India, as well as Khotan. One conversation reveals more about the target audience:

  I am going to China.

  What business do you have in China?

  I am going to see the bodhisattva Manjushri.

  The intended users were monks traveling on the pilgrimage route that became popular in the eighth century. Starting in Tibet or Khotan, they traveled east, stopping at Dunhuang, with their final destination the devotional center for Manjushri at Mount Wutai in Shanxi (about four hours northwest of Beijing by car).

  A gap in surviving documents means that we know very little of Khotan’s history between 802, when the Dandan Uiliq documents end, and 901, when documents from the library cave record that an official in Dunhuang provided Khotanese envoys with one bundle and eight additional sheets of fine paper.57 In the tenth century, the Khotanese kings were part of the same international order as the Cao-family rulers of Dunhuang: they sent emissaries to and received emissaries from each other, as well as the Ganzhou and Turfan Uighur Kaghanates and the different dynasties of central China. To travel to central China, Khotanese envoys went first to Dunhuang and then Ganzhou before proceeding to Lingzhou (modern Wuling County, Ningxia), an important stopping point for delegations on their way to the capital. Travel to central China was so uncertain that the Khotanese and the two Uighur Kaghanates frequently sent envoys bearing tribute only as far as Dunhuang, which they sometimes referred to as “China.”58

  The Cao family of Dunhuang and the royal family of Khotan had close ties. The Khotanese king Visa Sambhava, who ruled Khotan from 912 to 966, also used the Chinese name Li Shengtian.59 Sometime before 936 he married the daughter of Cao Yijin. The Khotanese royal family maintained a residence in Dunhuang where Visa Sambhava’s wife often stayed and where the heir to the Khotanese throne lived.60 The crown prince’s residence functioned as a representative office for the Khotanese, and it is very possible that the Khotanese-language documents found in cave 17 were an archive donated by the crown prince’s residence to the Three Realms Monastery.61

  In 938 Visa Sambhava sent envoys from Khotan to Kaifeng, Henan, the capital of the Later Jin. This is one of five instances during his reign when the Khotanese sent delegations to China, which was unified in 960 by the Song dynasty.62 The Chinese-language records of these exchanges are typically brief: for example, “On the fourth day of the twelfth month [of the second year of the Jianlong reign, or 961] the king of Khotan Li Shengtian sent an emissary to present one jade tablet and one box” to the founder of the Song dynasty.63 The Chinese would normally record the date, the name of the country sending gifts, the item presented, and sometimes the name of the lead emissary, but little else.

  In contrast, a group of some fifteen Khotanese-language documents preserved in cave 17 offer a wealth of detail about one mission, which consisted of seven princes and their entourage who left Khotan, probably in the mid-tenth century near the end of Visa Sambhava’s reign.64 These documents reveal much about the nature of trade on the Silk Road, particularly in the difficult conditions of the tenth century.

  The princes and their entourage set off with some 800 pounds (360 kg) of jade.65 In addition, they carried some leather goods, most likely saddles, harnesses, or other horse tack. Horses and jade were the most common tribute items from Khotan, and other recorded gifts included camels, falcons, yak tails, textiles, furs, medicines, minerals, herbs, some types of fragrances, amber, and coral.66 As was fitting in the subsistence economy of the time, rulers also presented slaves to one another.67

  THE KHOTANESE KING AND QUEEN AS DONORS IN A DUNHUANG CAVE

  This painting from cave 98 at Dunhuang shows the Khotanese king Visa Sambhava (reigned 912–66) and his wife, who was the daughter of the Dunhuang ruler Cao Yijin. Ties between the two families were close, and the Khotanese royal family frequently contributed funds for the construction of caves at Dunhuang. Drawing by Amelia Sargent.

  Rulers liked these gifts and said so. At one point when the Khotanese and the Uighur Kaghanate of Ganzhou had failed to exchange gifts for ten years, the kaghan wrote to the Khotanese king. (The letter survives only in its Khotanese translation; he probably wrote in either Chinese or Tibetan, the two diplomatic languages of the northwest in the tenth century.) The Uighur kaghan longed for “the many various wonderful things” that the Khotanese delegations had previously brought to him.68 Most of all, he probably missed the intelligence, particularly about the military strength of rival powers, that only envoys could provide.69

  Travel from one kingdom to another seemed slow, even to people at the time. As one envoy traveling with the princes from Khotan complained, “I shall go as far as Dunhuang making a difficult journey in forty-five days on foot, which with power to fly in the air I had done in one day.”70 Those who proceeded on horseback took eighteen days to cover 950 miles (1,523 km) by road.71 No wonder they envied birds the power of flight.

  The princes never made it to the Chinese capital; the ruler of Dunhuang felt that travel was too dangerous to go to Ganzhou, where three armies were battling each other in a succession dispute after the Uighur kaghan had died. The Dunhuang ruler insisted that they stay in Du
nhuang. The trip ended in total failure, the princes bitterly complained in letters home. Forced to spend the different gifts they carried, the princes ended up utterly destitute, as they wrote:

  All the animals our men had are lost. Our clothes are lost.… There is no one with whom we can get out and go to (?) Ganzhou. How can we then come to Shuofang [where the envoys visiting the Chinese capital were first received], since we have neither gift nor letter for the Chinese king? … Many men have died. We have no food. How were then an order to come? How can we have to enter a fire from which we can not bring ourselves back?72

  A letter from their escorts explains how each of the animals was lost.73

  The ruler who did not allow the princes to continue their journey saw the princes’ purpose as quite different from that of the monks traveling with them. Monks sometimes traveled as pilgrims and sometimes as members of official delegations. Rulers received monks because they believed that hosting a powerful monk could bring immediate benefits, whether in the form of miracles or the enhanced prestige accruing to Buddhist patrons. When the princes’ delegation fell apart, the monks left the group, took some of the gifts intended for the Chinese, and settled down with wives. This is hardly what we might expect of Buddhists who had taken vows of celibacy, but entirely in keeping with what excavated materials reveal about other Buddhists, whether in Niya or Dunhuang.