The Silk Road: A New History Read online

  “Silk” is even more misleading than “road,” inasmuch as silk was only one among many Silk Road trade goods. Chemicals, spices, metals, saddles and leather products, glass, and paper were also common. Some cargo manifests list ammonium chloride, used as a flux for metals and to treat leather, as the top trade good on certain routes.

  Another common trade item was paper, invented during the second century BCE, and surely a far greater contributor to human history than silk, which was used primarily for garments.1 Paper moved out of China via these overland routes first into the Islamic world in the eighth century, and then to Europe via its Islamic portals in Sicily and Spain. People north of the Alps made their own paper only in the late fourteenth century.2

  The term “Silk Road” is a recent invention. The peoples living along different trade routes did not use it. They referred to the route as the road to Samarkand (or whatever the next major city was), or sometimes just the “northern” or “southern” routes around the Taklamakan Desert.3 Only in 1877 did Baron Ferdinand von Richthofen coin the phrase “Silk Road.” He was a prominent geographer who worked in China from 1868 to 1872 surveying coal deposits and ports, and then wrote a five-volume atlas that used the term for the first time.

  His map, reproduced in color plate 2–3, depicted the route between China and Europe in Roman times as a trunk route. Von Richthofen read Chinese sources in translation and was the first European geographer to incorporate data from the dynastic histories into a map of the region. The orange line shows information from the classical geographers Ptolemy and Marinus; the blue line, from the Chinese histories.4 In many ways his Silk Route resembles a straight railway line cutting through Eurasia. In fact, von Richthofen was charged with designing a potential railroad line from the German sphere of influence in Shandong through the coalfields near Xi’an all the way to Germany.5

  Gradually the term gained acceptance. Sven Hedin’s 1936 book about his Central Asian explorations carried the title The Silk Road in its 1938 English translation. In 1948 the Times of London included the following question in its “Fireside Questions for the Family: A Test of General Knowledge”: “From where to where do, or did,” the Silk Road run? The answer: “China borders by various routes to Europe.”6 The term has shown considerable staying power as a designation for overland trade and cultural exchanges across Eurasia.

  From its inception, the Silk Road was shown as relatively straight and well traveled, but it never was. Over a hundred years of archeological investigation have revealed no clearly marked, paved route across Eurasia—nothing remotely like the Appian Way of Rome—but instead a patchwork of drifting trails and unmarked footpaths. Because there was rarely a discernible route, travelers almost always hired guides to take them along a particular section, and they frequently shifted to another path if they encountered obstacles.

  These meandering trails converge at oasis towns—the towns this book explores. When flying over this region today one merely has to identify the highest peaks to locate the principal sources of the streams nurturing the main Silk Road cities of ancient times. Because the documents are largely from these towns, this book is organized around seven ancient Silk Road sites—six in northwest China and one to the east of Samarkand—that form the chapters of this book.

  These towns were semi-independent city states ringing the Taklamakan Desert. The rulers, whether on their own or on behalf of Chinese dynasties, strictly supervised trade and played a major role as the purchasers of goods and services. This produced a paradox: once the trade passed through totally wild regions and entered one of these oasis communities, it was suddenly highly regulated.

  This was especially true when the Chinese stationed troops in Central Asia—primarily during the Han dynasty (206 BCE-220 CE) and the Tang dynasty (618–907 CE). The central government made massive expenditures to supply these armies with grain and uniforms and to pay thousands of soldiers. Bolts of silk took on another important function during the Tang dynasty, which was unable to mint enough bronze coins to cover the expenses of the central government. The authorities recognized three commodities as currency: bronze coins, grain, and bolts of silk. Since they often suffered coin shortages, and since grain rotted, most of these payments were in bolts of plain-woven silk, shown in plate 5A. Many of the military subsidies to the northwest were paid in silk, and bolts of silk circulated widely in the Western Regions as a result. When the soldiers made many purchases at local markets, trade boomed. But when rebellion threatened the emperor and he summoned all troops back to central China, trade fell off markedly.

  Even with the Chinese military presence, there was no documented traffic between China and Rome during the years of the Roman Empire. Contrary to popular belief, Romans did not exchange their gold coins directly for Chinese silk. The earliest Roman gold coins found in China are Byzantine solidus coins, including many imitations, as shown in plate 4a. They come from tombs dated to the sixth century, long after Emperor Constantine (reigned 312–37 CE) moved the empire’s capital to Constantinople.

  Geographically, the Silk Road goes through an astonishingly diverse landscape, much of it treacherous. Beginning in Xi’an and traveling westward, travelers first traversed the Gansu Corridor. This is a 600-mile (1,000 km) route running mainly east-west between the Qinghai Mountains on the south and the Gobi Desert of Mongolia on the north. After reaching the oasis city of Dunhuang, in Gansu Province, they had to decide whether to take the northern route or the southern route around the Taklamakan Desert, which converged in Kashgar. If both routes were impassable, those making the trip could take a central route right through one of the most inhospitable deserts on earth.

  After passing through Dunhuang, travelers entered the region called Xinjiang, literally the “New Frontierlands,” a term used by the Qing dynasty when it conquered this area in the eighteenth century. Before that, the Chinese called this region Xiyu, meaning “Western Regions,” an area spanning parts of Uzbekistan and Tajikistan to the west and the Chinese provinces of Gansu and Shaanxi to the east.7 Modern Xinjiang encompasses most of the Silk Road routes in western China.

  Here modern tourists will see the breathtaking vistas of modern Xinjiang, and comprehend why there was not one Silk Road but multiple routes. The first daring peoples to traverse this region learned how to cross deserts in the winter when the sun was not too hot, and where to cross mountain passes in the summer when the snow was light. Above all, they learned to skirt the edge of the desert, pausing to drink, rest, and learn about the route ahead. At each oasis community they might stop for days, weeks, or much longer, in order to plan the next step.

  Travel was painfully slow. In 1993 a British officer and explorer named Charles Blackmore led an expedition on foot through the Taklamakan. His men and camels managed to cover 780 miles (1,400 km) across the Taklamakan between Loulan and Merket, southwest of Kashgar, in fifty-nine days, averaging just over 13 miles (21 km) a day. Walking over the dunes in the sandy part of the desert was strenuous, and they did not always make ten miles (16 km) in a day, but walking on the flat pebbled surface, they reached as much as 15 miles (24 km) per day.8 These rates give a good approximation of what travelers in previous centuries endured.

  Once across the desert, travelers faced towering peaks separating the Taklamakan from all points west and south. It is here that the earth’s largest mountain ranges crash together in a Mardi Gras of snow and ice—the Pamir Knot—where the Himalayas, Tianshan, Karakoram, Kunlun, and Hindu Kush meet. Once through, travelers descend west to Samarkand or south toward India.

  Few individuals traversed all of Central Asia, covering the distance of some 2,000 miles (3,600 kilometers) between Samarkand and Chang’an. The most famous (though not the most reliable) Silk Road traveler, Marco Polo (1254–1324), claimed to have traveled all the way from Europe to China by land and to have returned home by sea. Most travelers moved on smaller circuits, traveling a few hundred miles (around 500 km) between their hometown and the next oasis and no furth
er. Because goods were traded locally and passed through many hands, much of the Silk Road trade was a trickle trade. Long-distance caravans with hundreds of animals are rarely mentioned anywhere in the historical record—and usually only when states exchanged emissaries.

  Today the region between Dunhuang and Samarkand attracts many tourists who come to see the famed ruins, including those now buried deep in the desert like the Rawak monastery outside Khotan, the walled cities of Turfan, and the caves of Dunhuang and Kucha. Local museums display artifacts found in tombs, such as silver and gold vessels and textiles combining Eastern and Western motifs in lively and exquisite designs. In a handful of places, the desert’s dry climate has preserved the mundane as well as the visually striking: Chinese dumplings lie buried alongside rounds of North Indian naan flatbread baked over a thousand years ago by ancient Silk Road residents.

  Before the end of the nineteenth century, no one realized that the sands of Xinjiang preserved so many documents and artifacts from the distant past. In 1890, the British officer Lieutenant Hamilton Bower traveled to Kucha, an oasis on the northern route around the Taklamakan, to investigate a murder. While there, he bought an ancient manuscript consisting of fifty-one leaves of birch bark with writing on them and announced the discovery to the Royal Asiatic Society of Bengal. Within a few years scholars identified it as a medical text from the fifth century CE, making it the oldest known Sanskrit manuscript in the world by almost one thousand years.9 Alerted to the importance of such discoveries, European diplomats in Asia began to purchase various manuscripts and sent them to Europe, where scholars trained in philology could decipher them.

  DESICCATED DUMPLINGS FROM TURFAN

  The dry conditions at Turfan preserved many perishable items, including food. Here we see four wontons and a single dumpling, dating to the 600 or 700s. By examining the dumplings that have cracked open, archeologists have identified Chinese chives and some type of meat, most likely pork, since Xinjiang was not yet Islamified at this time. Xinjiang Museum.

  In 1895 the Swedish explorer Sven Hedin launched the first scientific mission to Xinjiang, the source of many of these ancient manuscripts. Departing that April from the town of Merket on the Yarkand River, Hedin entered the Taklamakan Desert in search of the source of the Khotan River. After fifteen days, he discovered that he was not carrying enough water for himself and the four men with him. Still, he did not turn back because he did not want to admit that his expedition had failed. When their supply ran out, he began a desperate search for water. As his men and camels collapsed one by one, the exhausted Hedin forced himself to crawl along a dry riverbed. On the sixth day without water, he located a stream, drank his fill, and carried back enough water in his boots to save the life of one man.

  As he made his way out of the desert, Hedin encountered a caravan of four merchants and various pack animals, from which he purchased three horses, “three pack-saddles, one riding-saddle, bits, a bag of maize, a bag of wheat-flour, tea, jugs, bowls, and a pair of boots.”10 This is a revealing list. Even at the beginning of the twentieth century, just as in earlier times, almost all the goods traded in the Taklamakan were locally made necessities, not foreign imports. After leaving the desert, Hedin learned that shepherds had aided another of his men, but two others had perished.

  A chastened Hedin returned to the Taklamakan in December of the same year. This time he brought along enough water for his men. Entering the desert from Khotan, one of the main oases on the southern edge of the Taklamakan, they discovered the ruins of the Dandan Uiliq site. Among the wooden posts and the remnants of walls in the sand lay several Buddhist statues. Hedin did not excavate; as he later explained, “I was not equipped to make a thorough excavation; and, besides, I was not an archeologist.”11 European newspapers carried extensive coverage of Hedin’s Taklamakan explorations, which were as exotic and dangerous as space explorations are today.

  One of these news reports was sent in late 1897 by a manager of a Polish coal mine to his brother, Aurel Stein, who was working as an education official in the British colonial city of Lahore in India (now Pakistan).12 A native of Hungary, Stein had completed his doctorate in Sanskrit at Tübingen in 1883 and continued to study the language with the learned Indian scholar Pandit Govind Kaul in Lahore. Sanskrit was an enormously popular field throughout the nineteenth century; many people wanted to study the Indo-European language that was more ancient than, and closely related to, Latin and Greek. During his studies in Germany, Stein had learned the importance of obtaining the earliest and most complete manuscripts.

  Immediately recognizing the implications of Hedin’s discovery for the study of ancient manuscripts, Stein applied to the British archeological authorities for funding to go to Khotan. Systematic investigation of the site, he argued, would provide far more information than the pillaging that had so far occurred. He also hinted at the international competition to acquire antiquities already under way. Hedin, he reported, was bound to return to the region, and the Russians were contemplating launching an expedition too. The Government of India funded his application.

  The first to locate and map many of the sites discussed here, Aurel Stein found a number of stunningly important objects and documents. Leader of four different expeditions to Xinjiang between 1900 and 1931, Stein wrote extensive formal reports as well as more casual narratives. His excavations were imperfect by today’s standards; he hired workmen to dig for him and rewarded them extra pay for any finds, a widespread practice that sometimes resulted in overly rushed excavation. But few of the excavators who found documents in Xinjiang—Paul Pelliot of France, Albert von Le Coq of Germany, Otani Kozui of Japan—matched the level of detail found in Stein’s archeological reports. None went to as many places as Stein or published nearly as much material.

  Stein’s descriptions are essential to reconstructing the original condition of each site. His explanation of the circumstances leading to the burial of the documents is also important; every subsequent scholar has relied on Stein as a point of departure even when they have updated his explanations. Stein’s and other accounts from the late nineteenth and early twentieth centuries are informative because their authors, with few exceptions, traveled along the same routes, using the same means of transportation as did travelers in earlier times. Their accounts fill in many details left unmentioned by past travelers, making it possible to relive the experience of travel along the ancient trade routes.

  These explorers, and many who followed, revealed what was hidden beneath the sand. First, they discovered archeological evidence showing that long-distance overland trade began long ago. Different peoples living in Xinjiang sent goods to central China as early as 1200 BCE. At the time, the kings of the Shang dynasty (1766–1045 BCE) ruled the lower Yellow River valley and wrote using the earliest extant Chinese characters. The lavish tomb of one king’s consort, a woman named Fu Hao, contained over one thousand jade implements, some carved out of the distinctive milky green jade of Khotan. Large quantities of seashells found in Central Asia, particularly at the Wubao site near Hami in Gansu Province from the same period, testify to trade with coastal regions either to the east in China, to the south in India, or to the west along the Mediterranean.13

  Secondly, they revealed that diverse ethnic groups once inhabited the area. For example, at sites in Xinjiang and Gansu dating from 1800 BCE to the early centuries BCE, the dry desert climate has preserved about five hundred desiccated corpses.14 Many of the males measure over six feet (1.8 m) tall, much taller than their Chinese contemporaries, and the deceased often have non-Chinese—sometimes called Caucasoid—features like fair hair and pale skin. Their appearance has led scholars to propose that many of the people traveling along and settling in the different oases around the Taklamakan Desert were descended from the speakers of Indo-European languages. These peoples, linguists believe, migrated to ancient India and Iran, probably from their original homeland, possibly the Pontic steppe north of the Black Sea, sometime between 2000 and 1
000 BCE.15 Some of the corpses wear woolen textiles with plaid patterns resembling those in second-millennium BCE Ireland, further evidence for Indo-European roots.16 Some scholars have proposed that they spoke Tocharian, an Indo-European language discussed in chapter 2. Yet because none of these tombs have produced any written evidence, we cannot know what language(s) these peoples spoke.17

  There are also discoveries of trade with people to the north, at the site of Pazyryk in Siberia, which dates to the fifth century BCE. The residents of this site buried Chinese bronze mirrors and silk in their tombs.18 One silk fragment bears an embroidered phoenix, most likely a Chinese motif (or a motif copied from something originally Chinese), since this had positive associations in Chinese culture. A similar textile, also from the fifth century BCE and found at Turfan, shows a beautifully embroidered phoenix on a faded yellow silk background.19 These finds indicate that overland trade was certainly taking place in the centuries before the Common Era, but no documents reveal who was carrying these goods or why.

  The first written description of the Silk Road trade concerns Zhang Qian (d. 113 BCE), a Chinese envoy from Chang’an to Central Asia in the second century BCE, during the reign of Emperor Wu of the Han Dynasty (reigned 140–87 BCE). The emperor hoped that Zhang Qian would persuade the Yuezhi people, living in the Ferghana region of modern-day Uzbekistan, to ally with the Chinese against their common enemy to the north, the Xiongnu tribal confederation, based in modern-day Mongolia. The earliest surviving account about Zhang Qian was written at least 150 years after his trip and does not provide many basic facts about his journey, such as his exact itinerary.