Читать книгу Silk, Slaves, and Stupas - Susan Whitfield - Страница 13

THREE

A Hoard of Kushan Coins

IN THE SIXTH CENTURY a Christian church and monastery were built on the top of a steep-sided plateau in the highlands of what was then the kingdom of Axum (ca. 100–940), in present-day Ethiopia (plate 2). It is possible there was an existing shrine at the site, although no traces are recorded. Among the treasures found in the monastery grounds was a hoard of over a hundred gold coins, possibly contained within a wooden casket decorated with gold and green stone. The coins had been minted at the heart of the Silk Road when the Kushan Empire united Central Asia. They ranged in date from the early second century to the beginning of the third century.¹ The earliest showed signs of wear, but the newest—six gold staters of Vasudeva I (r. 190–230)—were in mint condition, suggesting they were probably taken to Axum during or shortly after his reign. Why, where, and how were the coins first minted, how did they travel to East Africa, and why did they end up in this Christian setting? And where are they now? To try to answer these questions, let us start their story in their homeland with the rise of the Kushan Empire in the first century AD and the Kushan use of coinage.

THE KUSHAN EMPIRE

The rise of the Kushan, like that of many Silk Road empires, is linked to the movement of peoples across the ecological rift that separates northern from

For places mentioned in this chapter see Map 4 in the color maps insert.

southern Eurasia (see chapter 1). Northern Eurasia, which David Christian has called “Inner Eurasia” but which is also called “Inner Asia,” has an arid continental climate that cannot support large areas of sustained agriculture. Much of Inner Eurasia has long been home to peoples surviving by varying mixtures of agriculture and pastoralism, moving in search of pasture one or more times a year, and creating small settlements.² Christian calls the border with Outer Eurasia—a land of settled people living primarily from agriculture and with large cities—“the dynamo of Inner Eurasian history.”³ One such group of agricultural pastoralists living in Inner Eurasia were called the Yuezhi by Chinese historians.⁴ They were a group of tribes living around the Tianshan (Heavenly Mountains), including in the Hexi corridor of what is now northwestern China. Attempts have been made to link the Yuezhi to the burials from an earlier period found in the Tarim basin to the west, but without further evidence such connections remain hypothetical. We can only speculate on the Yuezhi’s language and ethnicity, if indeed they were one people.⁵ When we do read about the Yuezhi in Chinese histories, it is after they have been pushed out of the Hexi corridor by invaders from the north. These were another confederation of steppe pastoralists, named the Xiongnu in Chinese (see chapter 1), who in the third century expanded their territory south across the Gobi Desert and into the Yuezhi homelands.⁶

For this early history we have to rely almost exclusively on the Chinese records.⁷ But these were informed by firsthand accounts, primarily that of the imperial envoy Zhang Qian (d. 113 BC). He was sent in 138 BC by the Chinese emperor to seek an alliance with the Yuezhi against the Xiongnu, who were also pushing at the northwest borders of China. It is his intelligence on his return to China that informs Shiji, the great work of the so-called first historian of China, Sima Qian, and Hanshu, the subsequent history of the Former Han dynasty (206 BC–AD 9). The latter was compiled by Ban Biao and his son and daughter Ban Gu and Ban Zhao. Ban Biao had another son, Ban Chao, who also contributed his knowledge. Ban Chao rose high in the military, fought several decades of battles against the Xiongnu alliance, and became “protector general of the Western Regions.” From the Han onwards the chapter on the “Western Regions” was to remain a mainstay of Chinese official histories.

We are told in the biography of Zhang Qian in these same histories that he was captured by tribes loyal to the Xiongnu alliance on his journey out and managed to escape only after a decade, having married a local woman and fathered a son. He reached the Yuezhi in 126 BC, and it is from his accounts given in the histories that we learn of their exodus:⁸ “The Yuezhi originally lived in the area between the Qilian or Heavenly Mountains and Dunhuang, but after they were defeated by the Xiongnu they moved far away to the west, beyond Dayuan [Ferghana Valley], where they attacked and conquered the people of Daxia [Bactria] and set up the court of their king on the northern bank of the Gui [Oxus] River.”⁹

Elsewhere in the histories the account is expanded. It is recorded that the Xiongnu killed a Yuezhi ruler and that the son of Modu, the Xiongnu leader, used his skull as a drinking cup. Some Yuezhi moved south, but most, it is recorded, first fled north across the Tianshan to settle in the Ili and Chu valleys—now on the borders of Kazakhstan and China.¹⁰ When Zhang Qian embarked on his mission, he might have believed that they still lived in this region, close to the western borders of Xiongnu territory. However, probably around the middle of the second century BC, the Yuezhi were pushed out of this new homeland by other invaders from the steppe, called the Wusun in the Chinese accounts, and moved west again. They eventually settled in the land between the Sogdians—who lived north of the Amu Darya (Oxus)—and the Bactrians, who lived in the region just to the south of the river, in present-day northern Afghanistan. This was a long way from their old enemies, and Zhang Qian’s mission thus became longer than expected. But his journey further west in search of them was to result in intelligence that is traditionally seen as an impetus for the Chinese expansion into Central Asia.¹¹

Zhang Qian spent a year with the Yuezhi and in Bactria to the south of the Amu Darya. He must have had many tales to tell on his return. Unfortunately, Chinese histories are terse, giving only brief descriptions of each foreign people.¹² That in the Shiji for the Yuezhi simply reads: “They are a nation of nomads, moving place to place with their herds and their customs are like those of the Xiongnu. They have some 100,000 or 200,000 archer warriors.”¹³

The Chinese histories also suggest that the Yuezhi had conquered the lands immediately to the south of the Amu Darya, including the Hellenistic city of Balkh (see chapter 5), but that they remained a confederation with many leaders:

Daxia is located over 2,000 li southwest of Dayuan, south of the Gui river. Its people cultivate the land and have cities and houses. Their customs are like those of Dayuan. It has no great ruler but only a number of petty chiefs ruling the various cities. The people are poor in the use of arms and afraid of battle, but they are clever at commerce. After the Great Yuezhi moved west and attacked the lands, the entire country came under their sway. The population of the country is large, numbering some 1,000,000 or more persons. The capital is called the city of Lanshi [Balkh] and has a market where all sorts of goods are bought and sold.¹⁴

Here we see a group of agro-pastoralists of Inner Eurasia, the Yuezhi, forced away from their home and eventually settling across the geographical rift, in the Outer Eurasia of agriculture and cities.¹⁵ We might hope for archaeological sources and further written records from this time to supplement the Chinese histories, but both remain frustratingly scarce.¹⁶ Some of the main sources are coins.¹⁷

What has been pieced together is that a leader of one of the regional rulers of the Yuezhi took overall power in the mid-first century AD, marking the start of the Kushan Empire. This ruler, Kujula Kadphises, expanded his territory south into the Kapisa region (Bagram), Taxila, and Kashmir, and his successors moved into northern India. At the end of his reign he sent an army into the Tarim basin. This is reported in the Hou Hanshu (History of the Later Han; 25–220):

More than a hundred years later, the xihou (“Allied Prince”) of Guishuang (Badakhshān and the adjoining territories north of the Amu Darya), named Qiujiu Que (Kujula Kadphises), attacked and exterminated the four other xihou (“Allied Princes”). He set himself up as king of a kingdom called Guishuang. He invaded Anxi and took the Gaofu [Kabul] region. He also defeated the whole of the kingdoms of Puta and Jibin. Qiujiu Que (Kujula Kadphises) was more than eighty years old when he died.¹⁸

His son, Yan Gaozhen (Wima Taktu), became king in his place. He returned and defeated Tianzhu (Northwestern India) and installed a General to supervise and lead it. The Yuezhi then became extremely rich. All the kingdoms call [their king] the Guishuang (Kushan) king, but the Han call them by their original name, Da Yuezhi [Great Yuezhi].¹⁹

The stability brought by the Yuezhi enabled a growth of trade north from Sogdiana, south into India, and from there to Barbarikon and the Indian Ocean routes, as well as east to the Tarim basin and thence to China, and west into Parthia and thence to Rome. As Millward notes in relation to the Tarim basin, this is “a phenomenon displayed over and over . . . a nomadic royal house and its followers forging a confederation and establishing imperial control over sedentary populations.”²⁰ The rise of the Kushan confederation could be seen as the most important factor supporting the growth of sustained long-distance trade across Eurasia, the Silk Road: one striking example of how, again quoting Christian, “Inner Eurasia has played a pivotal role in Eurasian and world history.”²¹

The paucity of written and archaeological sources continues to impede our understanding of this Central Asian empire.²² The sources give little information, for example, about their religion. What there is points toward the Kushan kings practicing an eastern form of Zoroastrianism but patronizing other religions: in Raymond Lam’s words, demonstrating a “profound cultural-religious eclecticism.”²³ Buddhism was included in their patronage. Buddhist missionary activities had originally been given impetus in India by the Mauryan king Aśoka (r. ca. 268–232 BC). He made Buddhism his state religion and vowed to disseminate the faith throughout the world. According to tradition he sent his son and daughter to Sri Lanka and prominent monks to Central Asia, West Asia, East Asia, and Southeast Asia. Inscriptions above cave temples in Sri Lanka support the arrival of Buddhism at this time.²⁴ From the earliest period, therefore, Buddhism was a proselytizing faith with a penchant for long-distance travel.²⁵ The Maurya Empire (322–180 BC) included the city of Taxila, lying at the junction of trade routes leading between India and the Tarim, and this was to become one of the Kushan capitals. The region became a major center of Buddhism by the end of the Mauryan period, with famous stupas at Dharmarajika in Taxila and Butkara near Mingora in the Swat valley (see chapter 4). By the start of the second century BC the Maurya Empire had shrunk, and Menander, the Greek king of Kapisa/Gandhāra, had expanded his kingdom to the Kabul and Swat valleys.²⁶ The Pali text Milinda Pañha (The questions of Milinda) records that he became a Buddhist and recounts his debates with the Indian Buddhist sage Nāgasena.

By the start of the Kushan Empire under Kujula Kadphises in the first century AD there were Buddhist monasteries and stupas around the ancient city (near Jalalabad).²⁷ Buddhism spread northwest into the Kabul region during Wima Kadphises’s reign and then moved further north into Bactria. The Kushan kings used imagery borrowed from Greek and Indian religions to represent their gods. For example, when Kujula Kadphises minted coins he chose the imagery of the Greek god Heracles to represent the Kushan god Wesho. Wesho was also shown in the imagery of the India god Śiva. Buddha appears on coins from the time of Kaniška, the only direct representation of an Indian “deity.” As well as providing information about their beliefs, it is largely thanks to the Kushan coinage that it has been possible to construct a chronology of the kings of Kushan.²⁸

KUSHAN COINS

Money has been needed from the earliest human cultures, with many objects taking the role of an object of value used for exchange—from livestock to obsidian, grain, and silk. Early societies also used objects with little value—such as cowrie shells and beads—to exchange for higher-value goods. But coins, manufactured solely for the purpose of exchange, appear in cities around the Aegean Sea, in India and in China between 700 and 500 BC, seemingly representing separate developments. By the time of the rise of the Kushan Empire, coins were in use by neighbors in regions they conquered and there were also coins in circulation from more distant cultures, such as Rome.

Coins had a use beyond simple objects of value for exchange. They also acted as symbols of power and authority, and some argue they were a means of asserting legitimacy and hegemony.²⁹ For the Yuezhi, who were invaders into the region, Daniel Michon argues that the minting of coins that were familiar to those they were seeking to rule presented an idea of continuity of power rather than change. He sees the Yuezhi making a conscious choice among the variety of coins that were in circulation, replicating ones that showed a horse rider on the obverse, for example, as they were themselves horsemen.³⁰ However, Joe Cribb sees the primary purpose of continuity in coin design as enabling coins to circulate and argues that existing designs were reused for this purpose: “The horseman design was the standard design for coinage before the Kushan arrived, so they were just copying it, not necessarily with any intention of alluding to their own horsemanship.”³¹

We know the name of Kujula Kadphises only from the coins and one Bactrian inscription (he is assumed to be the Qiujiu Que mentioned in Chinese histories), yet as uniter of the Yuezhi clans under the Kushan he was of immense importance in Central Asian and indeed world history. During his reign both copper and silver coins were produced. They were diverse, both based on and reflecting the variety of existing coinage.³²

His successor, Wima Taktu, reduced the variety of coins. But he still produced coins only in silver and copper. It was with the third ruler, Wima Kadphises (r. early second century AD), that the first gold coins were minted.³³ These were not an innovation: gold staters had been produced by the Hellenistic kings of Bactria and of Kapisa/Gandhāra and by Rome, the last imported into India. On current understanding, it appears that techniques of the first two of these were used by the Kushan, alongside newly developed techniques, but not Roman ones. However, in appearance they probably borrowed from existing Roman coins. Harry Falk makes the case that Wima Kadphises made a deliberate choice to follow the standard of Roman coins under Augustus (r. 27 BC–AD 14), although they are not struck to this standard.³⁴ In terms of the weight, Robert Bracey argues that they followed a Bactrian standard derived from the Greek stater.³⁵

Under Wima Kadphises gold staters appear in a variety of forms showing production by the same mint but in five different chronological phases.³⁶ The double staters, as found in the Axum hoard, date from phase 4, presumably well into his reign and the most productive of the phases.³⁷ They weigh about sixteen grams—double that of the standard, hence the name. Bracey notes that it is unusual to find coins of Wima Kadphises among coin hoards of later kings and suggests that this might be a result of a low level of production or a short circulation of such coins. In this respect, the Axum hoard is unusual if not exceptional. They were probably minted in Balkh.³⁸ New reverse and obverse dies were made for each phase, and Bracey suggests that each phase saw a period of days or weeks of intensive production with multiple workmen but that there might have been long periods when no coins were minted.³⁹ The Kharoṣṭhī inscription and images were prepared freehand onto the dies by different craftsmen, probably using a prototype or drawing. The Greek legend was created by drilling holes into the die to mark the end point of the letters and then joining these up with lines. The obverse die was then fixed into an anvil, and the reverse die joined others in a die box. To make each coin, the gold blank was placed between the observe die and the reverse die. The reverse die was struck twice to create the impression. The coin was then removed and the impression checked. If not clear enough, it was replaced and struck again. To create the impression would require strong blows, probably holding a hammer with two hands. Therefore either the coin and reserve die were fixed or a second workman held them in place. Other coin-making workshops used hinged dyes or sleeves to hold them in place, but the evidence suggests neither of these were used by the Kushan. The dies, however, were probably up to 50 percent larger than the finished coins.

The source of the gold for the coins has been subject to discussion. Gold is found in the Zerafshan and Ferghanan valleys, north and east of the Kushan in the lands of Sogdiana and Dayuan, and also in the steppes to the north. Trade is commonly found across the ecological border of Inner and Outer Eurasia, the steppe providing goods such as horses and gold in trade for grain, silk, and other commodities (see chapters 1, 6, and 8). There would be little remaining evidence of such trade—the grain consumed and the gold reworked—and this source cannot be dismissed.⁴⁰ Southern Arabia has also been suggested as a source, and for a long time it was argued that a large part of the gold came from Rome in the form of coins. In an often quoted text, the first-century Roman historian Pliny the Elder (AD 23–79) estimated that the Roman Empire spent some 150 million sesterces a year for luxury products from India, including silks, gems, and spices.⁴¹ Yet while coin hoards have been found in Southern India, very few Roman coins have been found in Kushan territory. Moreover, recent analysis has shown the gold used in Roman coins to have a different source than that used in Kushan.⁴² But perhaps we should consider a more local source. As Falk points out, there is plentiful evidence of gold in the Kushan area, in the mountains of what is now eastern Afghanistan, and also brought down by the rivers.⁴³ And a large gold hoard has been found in graves to the west of Balkh, at Tillya Tepe.⁴⁴ The archaeologist in charge of the 1978 excavations of it, Victor Sarianidi, claimed that these were graves of Yuezhi rulers, pre-Kushan.⁴⁵ However, the Yuezhi/Kushan attribution has been questioned, as the grave goods have Parthian, Saka-Samaritan, Xianbei, and Chinese affinities. Whether the people buried there had links with these cultures or whether they came from these cultures, both show the rich interconnections of Central Asia in the first century.⁴⁶ Sarianidi also claimed that the gold was local river gold, a view supported by more recent analysis of some pieces.⁴⁷

The obverse of the five double gold staters found in the Axum hoard shows Wima Kadphises in profile facing right (figure 7).⁴⁸ He is bearded and wears a diadem and tall peaked cap. He is seated cross-legged on clouds with flames emanating from his shoulder. His dynastic symbol or tamgha is to the left (figure 8). Tamgha seem to have been developed by pastoralist people as a tribal symbol and used, for example, to brand their cattle, so this is another legacy of the Kushan’s past.⁴⁹ The inscription on the coin—reading in translation “King Wima Kadphises”—is in Greek language and script. The reverse shows the Kushan god Wesho (sometimes associated with the Indian deity Śiva).⁵⁰ He is standing with a battle-ax trident in his right hand and an animal skin in his left in front of a bull—often identified as Nandi, the mount of Śiva. The symbol to the left is a nandipada symbol, sometimes associated with the Buddhism Triratna, but Bracey argues strongly against this, attributing its appearance to imitation: “It appears because it appeared on other coins. Though some value was attached to it that is not why it was used.”⁵¹ The inscription is in Prākrit, an Indian language, written using the Central Asian Kharoṣṭhī script.⁵² It reads: “Of the great King of Kings, lord of the world, great lord, Wima Kadphises, the savior.”

FIGURE 7. Coin of Wima Kadphises, found in the Debra Damo hoard. After Göbl (1970).

FIGURE 8. Tamgha of the Kushan kings used on the coins.

Here we see in one coin a microcosm of Central Asian complexity, of the meeting and mixture of peoples, languages, gods, and symbols. This is not an accidental or random mixture of cultures but the deliberate exploitation and adaptation by a ruler of certain symbols, languages, and gods to send a message both to the peoples under his reign and to neighboring and more distant cultures.

Gold coins of the next king, Kaniška, also appear in a variety of forms, and it is not clear whether the five gold staters in the hoard were of one or more types.⁵³ Under Kaniška, for the first time some coins show images of Buddha on the reverse: up until then the gods depicted were local, Iranian, Indian, or Hellenistic. Robert Göbl indicates that the hoard did not include the issue with Buddha but one showing a bearded Kaniška with his head turned to the left (figure 9). Like all the kings depicted on these Buddha coins, he is shown standing. He wears a tall conical hat and holds a spear in his left hand and a goad in his right. With his right, he is tending an altar. The legend is no longer in the Greek language but in the local Bactrian language, although still in modified Greek script. It reads, “Shah of Shahs, Kaniška the Kushan.”⁵⁴ The Greek term found on his predecessor’s coins, basileus (βασιλεύς), often translated as “king,” has been replaced with the Iranian term for ruler, shah, and this continues to be used thereafter. On the obverse is the solar deity Mithra standing with head facing left, nimbate, holding the hilt of a sword in his left hand and with right hand outstretched in benediction. The Bactrian legend at right gives his name in Greek script as Miiro, while Kaniška’s tamgha is to the left.

FIGURE 9. Coin of Kaniška I, found in the Debra Damo hoard. After Göbl (1970).

Coins from Huviška’s reign form by far the greater part of the hoard—88 out of the 105 coins—and represent at least seven types (figure 10).⁵⁵ All show a bust of Huviška on the obverse. On these coins he is nimbate, faces left, and holds a filleted spear over his shoulder in his left hand and a mace scepter in his right.⁵⁶ The inscription—Bactrian in modified Greek script—reads, “Shah of Shahs, Huviška the Kushan.” The reverse shows a variety of male and female gods. The ones in the hoard are all from the Iranian pantheon: Pharro, Nana, Ardoxsho, Miiro (Mithra), Teiro, and Ashaeixsho.⁵⁷

FIGURE 10. Coins of Huviška, found in the Debra Damo hoard. After Göbl (1970).

The last coins found in the hoard are six gold staters from the reign of Vasudeva I showing a standing king facing left with a spear in his left hand and his right tending the altar (figure 11). The obverse shows Wesho with a bull. Although the history of this period remains obscure, this is generally accepted to be a time of crisis and retrenchment for the Kushan Empire. Much of their territory in northern India was lost, and to the west their old neighbors, the Parthian kings (like the Yuezhi, originally settlers from the steppe), had been replaced by an Iranian dynasty, the Sasanians, commencing under Ardashir I (r. 224–42). He was threatening Kushan territory from the north and west. Some accounts tell that Vasudeva sought alliances at this time to counter these threats, joining with the Armenian king Khosroes I (r. 217–52) against Ardashir. The Chinese histories also report a diplomatic visit, probably sent by Vasudeva, to the Wei court of northern China in 230.⁵⁸ We do not know of any alliance resulting from this. However, given the composition of the hoard, with only six coins from Vasudeva’s reign, it may be assumed that it left Kushan early in Vasudeva’s reign before the start of the Sasanian campaign, perhaps on a more general diplomatic mission.

FIGURE 11. Coin of Vasudeva I, found in the Debra Damo hoard. After Göbl (1970).

If we speculate that the hoard of coins of Vasudeva and his predecessors was a gift sent in a diplomatic embassy across the Indian Ocean, then they would have traveled from one of the Kushan capitals—possibly Peshawar or the summer capital Bagram. The mission would have undoubtedly had many other treasures to take as gifts for foreign rulers. From the city they would have followed well-traveled routes to the valley of the Indus River or one of its major tributaries.⁵⁹ From here it was possible to travel by river downstream. We have an early account of such a journey through Alexander the Great (r. 336–323 BC). He returned home this way in late 326 BC, marching first to the Jhelum River, where he had established the cites of Nicaea and Bucephala on opposite banks earlier in the year. He assembled eight hundred river ships, many of them newly built, and set sail in November after the end of the monsoon when the river was navigable. The boats were rowed by Phoenicians, Cypriotes, Carians (from southwestern Turkey), and Egyptians, all seafaring peoples. Soldiers accompanied them on both banks. They traveled for ten days before reaching the confluence with the Ashkini, where several boats lost control and many people were drowned. Further downstream his army was attacked, and, in the ensuing battle Alexander was seriously wounded. He survived and his army eventually won. He had to fight other battles as they progressed, and the army did not reach the ocean until the monsoon in 325.⁶⁰

At its greatest extent Kushan rule extended along the Indus River to the ocean, so there was less threat of such delays.⁶¹ Vasudeva might not have had his own fleet and might instead have relied, as did many diplomatic missions of the time, on getting passage on a merchant ship, probably for the river voyage and, at Barbarikon, changing to a seafaring vessel for the crossing of the Indian Ocean.

INDIAN OCEAN TRADE

Chapter 2 looked briefly at the maritime trade between India, the Red Sea, and the Gulf up to the first century BC. In the first centuries AD this trade continued and grew as new players came into the equation. In northern India and Central Asia the Kushan Empire provided a climate for trade by both land and sea. This was a major factor in the growth and success of the Silk Road. The stability provided by the Kushan enabled safe travel from the steppe to India through north-south routes, and from the Iranian Plateau to the Tarim, by east-west routes. Striking evidence for the trade is seen in the rock-cut graffiti at a way station called Shatial, high in the Indus River valley on the route from the north through Gilgit.⁶² Most are in Sogdian, left by merchants from Samarkand, Penjikent, and other cities north of the Kushan. But there are also inscriptions in Bactrian, Middle Persian, and Parthian. The routes south led from the Kushan capitals—such as Bagram and Taxila—to the Indus valley and thence to the seaport of Barbarikon—near Karachi in present-day Pakistan. There is plentiful evidence, archaeological and textual, to show the existence of Kushan-Roman trade at this time, most notably the Roman bronzes and glass and the ivories uncovered during excavations at Bagram.⁶³

One of the main textual sources is the Periplus of the Erythraean Sea, an anonymous handbook on Indian Ocean trade, written in Greek probably by a merchant based in Egypt in the mid-first century. Although at this time the Kushan Empire did not extend to the seaport of Babarikon, it is likely that items found in Bagram and elsewhere came in through this route and were traded at a site a little way inland. The Periplus lists glass as among the items traded there, along with gemstones, frankincense, textiles, silverware, wine, and coins.⁶⁴ It is supplemented by other texts, and although the imports from Asia have mainly long disappeared, scholars have made a strong argument for regular trade “undertaken for commercial profit, facilitated by the use of coinage and underwritten by accumulated capital.”⁶⁵ By this time ships were making use of the monsoon winds. These are caused by the Asian land mass heating up during the summer months; the rising hot air creates a vacuum that is filled by air from the ocean, causing strong winds that blow from the southwest. The opposite happens in the winter months, with the winds changing direction to come from the northeast. This system is supplemented by other regular patterns, such as equatorial winds and southeast trade winds. It was possible for ships to take advantage to sail from the mouth of the Red Sea and the coasts of East Africa and southern Arabia in summer, helped by the southwest monsoon, and for ships to set sail in winter from the ports on the east coast of South Asia, from Barbarikon to Taprobane (Sri Lanka), to be helped across to the Gulf and Red Sea. There was also the alternative of coast hopping from port to port, used from earliest times.

From the western edges of the Indian Ocean, ships came from the ports of the Red Sea, southern Arabia, and the Gulf. Those from the Red Sea included Roman boats that sailed from the ports in the north but also ships made by the Axumites, a kingdom in East Africa in what is now Eritrea and Ethiopia.⁶⁶

THE AXUM EMPIRE

In the third century the Axumites were a major political and trading power. Axum was listed as one of the four most important kingdoms in the world by the Persian religious teacher Mani (ca. 216–74)—alongside Persia, Rome, and China.⁶⁷ Two centuries earlier, the Periplus had noted Axum’s importance as a market for ivory: “Here is brought all the ivory from the land beyond the Nile, across the region called Cyenum and thence to Adulis.” This is sometimes given as the reason for the siting of the Axumites’ capital, Axum, relatively far west from the coast, as it acted as a hub for ivory hunters working inland to bring their goods over the Sudanese steppes and through the Nile valley.⁶⁸ But the city also lay on fertile ground with ready access to water. Adulis, described by the Periplus as a “large village,” had become the major port by the third century. From Axum, it was an eight- to ten-day journey over the highlands and down into the coastal plain and the port. Although the Axumites had a written language, Ge’ez, and also used Greek, there are no extant documents from this period. The name of the Axumite king contemporary with Vasudeva’s reign is known from inscriptions found in southern Arabia. The script used for this gives consonants and no vowels, and in this unvocalized form he is named “GDRT.”⁶⁹ This is tentatively vocalized as Gadarat. He was succeeded in about 230, the same time as the end of Vasudeva’s reign, by “DBH” (possibly Azaba or Adhebah).⁷⁰ Although the Axumites started minting regular coinage only under King Endybis (r. ca. 270–ca. 300), they were familiar with the concept before this. The Periplus mentions that they were importing brass “for cutting as money” and also that they imported money for “use by the foreign community.” When coins were minted by Endybis they were probably based on Roman coinage and produced in gold, silver, and bronze. They showed the head of the king and stalks of the indigenous wheat (Eragrostis tef) on the obverse, with a disc and crescent at the top. The inscription on the gold coins was in Greek, but the silver and bronze coins used the local language and script, Ge’ez, suggesting that the latter were for local use and the former for international trade.

Archaeology in Axum has revealed complex palace structures but also, most notably for this period, large stone built tombs (assumed to be royal burials) marked with carved stelae. The largest of these, which weighed 517 tonnes and would have been 32.6 meters tall if successfully erected, would have been the largest monolith to be raised by humans at that time.⁷¹ It was made from granite quarried on a hill close by and possibly transported by elephants—although they are no longer found here. The stelae are carved with representations of the local architecture of multistoried monumental buildings. These were constructed of dressed granite with recessed facades, rebated walls, wooden tie beams, and monumental staircases. Archaeology of sites in Axum dating from the second century onward reinforces this, with remains of several substantial complexes tentatively identified as palaces but also large elite residences, indicating a society with considerable wealth.⁷² Finds in the tombs and elsewhere show evidence for trade both north and south from Adulis, including Roman glass and Indo-Pacific beads.⁷³

Even though the written and archaeological evidence is scarce, it suggests that the Axumites were probably used to foreign visitors—traders, religious figures, and diplomats. But we enter the area of speculation by proposing that the gift of coins was brought by a diplomatic mission sent by the Kushans directly to the Axumites. Possibly it was intended for some other recipient and was waylaid here, or it might not have been a single hoard at all.⁷⁴ We therefore have to make a leap of imagination to picture the Kushan mission disembarking at Adulis after their long sea journey. From here they would have taken the journey inland, perhaps accompanied by local officials and soldiers or simply joining a trading caravan. The gift was handed over and perhaps placed in the royal treasury. Here it possibly remained through a succession of kings.

A century later King Ezena (r. ca. 333–ca. 356) converted to Christianity, and a cross replaced the disc and crescent on the Axum coinage. Churches were built and monasteries endowed over the following centuries. In the late fifth century it is recorded that nine Christians came to Axum from various parts of the Roman Empire (27 BC–AD 1453) to avoid persecution after the Council of Chalcedon in 451. The council, convened by the Eastern Church, declared that Christ had two natures in one person, God and man. This was not a doctrine accepted by many parts of the Christian Church and resulted in a split in the Eastern Church. The churches not accepting this doctrine are often referred to as the Oriental Orthodox or Old Oriental Churches. The Axumite Church was among them—along with Coptic, Syrian, Armenian, and Malankara (Indian) Orthodox. The lives of the nine Christians are recorded in much later biographies, albeit with many contradictions. Among them was ʾAragawi Zä-Mikaʾel—the Elder Zä-Mikaʾel.

According to his later biography, Zä-Mikaʾel was the son of the Roman prince Yeshaq and of Edna (an Ethiopian name). The biography states that at age fourteen he received his name from and became a monk with Pachomius (ca. 292–348) in Egypt.⁷⁵ Other Christians joined them in what Pachomius established as a new form of monastic community, where monks and nuns formed one community with common property and were presided over by an abbot or abbess. This broke with the earlier tradition of ascetics living largely as hermits. Zä-Mikaʾel’s mother Edna later became a nun here. Later Zä-Mikaʾel traveled to Rome and thence to Axum, which was by this time Christian. He invited his eight fellow believers to join him, and together with his parents they were welcomed by the king.⁷⁶ They lived at the court for twelve years before separating to evangelize in the countryside. Zä-Mikaʾel went with his mother and a disciple called Mattéwos to Eggala in the Tigray district. Here he decided to found a monastery—associated with Debra—on top of a steep-sided plateau. But he was unable to climb the cliffs until a serpent, living on top, let down his tail and pulled him up. The monastery was erected on the order of King Gabra Masqal, a large ramp being constructed to transport the building materials. On completion, the ramp was removed, making the perilous cliff climb the only route up. Zä-Mikaʾel’s biography also reports that his mother Edna became part of the community, which suggests there might also have been nuns, as in Pachomius’s original monastery.⁷⁷

Tradition says that the king endowed the monastery with treasures. This is possibly another clue in the journey of our coins. Could the king have presented the coins to the monastery at this time? We have no evidence of this, but it is a possible hypothesis. It is, of course, also possible that the monastery was built on an existing shrine and that the coins were already there, or, indeed, that they did not make its journey to Axum until later. It is almost certain that we will never have the answers to such questions.

Treasuries are found in churches throughout the Christian world (see chapter 8) and present-day Ethiopia is no exception, laying claim to holding one of the greatest treasures of Christianity, the Ark of the Covenant. According to tradition, this gilded wooden chest contains the tablets giving the Ten Commandments received by Moses on Mount Sinai and was carried with the Israelites until the temple was built in Jerusalem to house them. Ethiopian tradition further records that the Ark left Jerusalem at the time of King Solomon, carried by his son with the Queen of Sheba, Menelik. It was kept safe until finally being enshrined in the Church of St. Mary of Zion in Axum, built by Emperor Haile Selassie in 1960 for this purpose. Today it is said to be in the treasury building adjacent to the church. Only the high priest is allowed to see it.

While most believe this story to be apocryphal, the treasury probably does contain some ancient treasure, just like many other churches in Ethiopia. Other treasures have been found at Debra Damo, including an Axumite coin of Armah (r. 614–31), gold and silver Arabic coins dating from about the eighth to the tenth centuries, and some textiles from the sixth to twelfth century, probably originating in Egypt.⁷⁸ Our coins probably remained safely there for centuries. However, oral and written histories tell of two major threats to Debra Damo and the Christian heritage of the Axumites in the intervening centuries.

The first, which appears mainly in oral tradition, was during the reign of Queen Gudit around the mid-tenth century. Some corroboration is provided by a contemporary Arab traveler and geographer Ibn Ḥawqal in his work Kitāb Ṣūrat al-ʿArḍ (Picture of the earth), written in 977. “The country of the habasha [Abyssinians] has been ruled by a woman for many years now: she has killed the king of the habasha who was called Haḍani [from Ge’ez haṣ́ani, modern aṣ́e or atse]. Until today she rules with complete independence in her own country and the frontier areas of the country of the Haḍani, in the southern part of [the country of] the habasha.”⁷⁹

Gudit (Judith) is said to have come from Axum’s Jewish community, which had coexisted there for centuries alongside the Christians. Tradition tells that she killed the king to take the throne. It was usual at the time that royal princes were exiled on hilltop settlements such as Debra Damo—presumably to keep them from seeking power.⁸⁰ Queen Gudit duly went there, built a ramp to gain access, as Gabra Masqal had done, and killed the exiled princes to rid her of her rivals.⁸¹

The second event is better attested. It took place long after the end of the Axum Empire, starting in 1529 with the invasion of Ethiopia by Aḥmad ibn Ibrāhīm al-Ghāzī (r. ca. 1506–43), the ruler of the neighboring Muslim sultanate of Adal. He was named Gragn—“the Left-Handed” in Amharic. During this invasion many churches and monasteries were destroyed—including the Church of St. Mary of Zion, and, it is reported, many church treasuries were looted.⁸² The Ethiopians asked for assistance from the Portuguese, who landed a force in 1541, and the invaders were eventually driven out. As Paul Henze noted, although these events took place centuries ago, they remain alive in Ethiopian culture: “In Ethiopia the damage which Ahmad Gragn did has never been forgotten. Every Christian highlander still hears tales of Gragn in his childhood. Haile Selassie referred to him in his memoirs, ‘I have often had villagers in northern Ethiopia point out sites of towns, forts, churches and monasteries destroyed by Gragn as if these catastrophes had occurred only yesterday.’”⁸³

During the invasion, the emperor, Dawit II (Lebna Dengel) (r. 1508–40), was forced to flee his capital and take refuge at the monastery at Debra Damo. He was wounded in a battle nearby in 1540 and died and was buried at Debra Damo. One of the Portuguese, Miguel de Castanhoso, records that Lebna Dengel’s widow remained in Debra Damo, during which time Gragn laid siege to the mound for a year. But he was unsuccessful in gaining access. This makes sense when one reads Castanhoso’s description of the place:

The summit is a quarter of a long league in circumference, and on the area on the top there are two large cisterns, in which much water is collected in the winter: so much that it suffices and is more than enough for all those who live above, that is, about five hundred persons. On the summit itself they sow supplies of wheat, barley, millet and other vegetables. They take up goats and fowls, and there are many hives, for there is much space for them; thus the hill cannot be taken by hunger and thirst. Below the summit the hill is of this kind. It is squared and scarped for a height double that of the highest tower in Portugal, and it gets more precipitous near the top, until at the end it makes an umbrella all round, which looks artificial, and spreads out so far that it overhangs all the foot of the mountain, so that no one at the foot can hide himself from these above; for all around there is no fold or corner, and there is no way up save the one narrow path, like a badly-made winding stairs, by which with difficulty one person can ascend as far as a point whence he can get no further, for there the path ends. Above this is a gate where the guards are, and this gate is ten or twelve fathoms above the point where the path stops, and no-one can ascend or descend the hill save by the basket.⁸⁴

When the Portuguese arrived, they met with Lebna Dengel’s widow, and she accompanied them to the court of her son, Emperor Galawdewos. They joined forces and finally managed to defeat the invaders at a battle in 1543 with the help of Portuguese firearms that had been hidden at Debra Damo.⁸⁵

It is possible that these events might have impinged on the life of our casket of coins. Perhaps it was not given to the monastery until this time, taken by the emperor when he took refuge. Or perhaps it was already there, stored in the treasury but removed to a niche on the cliff face during this period in case the site was ransacked. We do not even know for certain the date when it was found, said to be around 1940. The Italian archaeologist Antonio Mordini, who was carrying out excavations at the site, gives the following account:

Toward the beginning of that year [1940], a young monk, entrusted with the repair works of the small wall sustaining the terrace where the church rises in the cemetery area of the convent, found in a small natural cave below the terrace, the relics of a wooden box along with eleven small gold plates, tweleve [sic] small bands of same metal, and of various lengths, and a considerable number of gold coins. The coins along with the gold plates and bands (which presumably formed part of the ornamentation of the box), were brought to Asmara, and the prior of the convent, mamher Takla ab Tasfai sold the entire lot to an Italian jeweller; who, in turn, sold it to a person of culture, who felt interested in the hoard.⁸⁶

This raises the question of the casket, for whose story we have even less evidence than for that of the coins, since at least we can be fairly certain of their origin. For a start, it is assumed that the pieces of wood, gold, and green stone found in the same place as the coins originally formed a casket. Mordini, one of the few people to see these, is confident on this point and describes them: “They consisted of ten small plates, partly rectangular and partly square; one large plate of hexagonal shape, slightly pyramidal and twelve bands of various sizes all decorated in light bas-relief with ornamental vines and stylish flowers of peculiar character.”

In 1943 Mordini located the nook and sieved the earth in it to find more remains of the casket, including pieces of wood, about thirty gold nails, “which, it is clear, must have served to fix the plates and the bands to the casket,” and some thin plates of green stone, which he could not immediately identify. He also found some potsherds that he judged to be from the Axumite period. In a later article, the Russian scholar S. I. Berzina suggests that the casket was Indian made and similar to a casket excavated by Neville Chittick.⁸⁷ However, since the casket was found in pieces and, as far as I have been able to ascertain, the pieces were not photographed or drawn, this conclusion must remain extremely tentative. David Phillipson gives information suggesting that the pieces of the casket were later discarded, so it is unlikely that we will ever know more.⁸⁸ The assumption that the coins were held in the casket is reasonable, given that we could expect such a rich hoard to be housed in some sort of receptacle. But whether this was the original receptacle or one used later—perhaps when the coins were moved—and where the box originated remain too uncertain to call.

The sale of monastery treasures was not unusual—and continues to this day (see chapter 9). It is fortunate that Mordini had the opportunity to study the coins after they were sold on.⁸⁹ This was sometime prior to 1959, when Mordini published his original account. I have not been able to find the identity of the Italian jeweler or the “person of culture,” although both might be noted in archives. The 1939 census of Asmara counted fifty-three thousand Italians among a total population of ninety-eight thousand in the city and a total of seventy-five thousand Italians in all of Eritrea. Among them there were presumably several jewelers. The nationality of the “person of culture” is unclear, but again details might be found in archives. As far as I know, this is the last record of the coins. While they might remain intact in a private collection, there is also the possibility that they were melted down to realize their value as gold.⁹⁰

Whatever the story of these coins and how they traveled, they reflect many aspects of the Silk Road. Created in the “dynamo” of Inner Eurasian history by an exiled agro-pastoralist people who had perhaps largely adapted to a settled city life, they reflect the importance of the Kushan for trade by both land and sea. While the early pieces probably circulated as money, the newest issues were probably never used for this purpose, instead being kept unmarked and unworn as treasures. At some point they traveled from Central Asia to East Africa by land, river, and sea on routes used by Silk Road merchants, possibly in a wooden casket. They later became the property of a Christian church, surviving over centuries. We can only hope that they remain as a treasured collection. If so, although inaccessible to most, they continue their entanglement with their collector and to tell their story.

1. I owe a great debt to the work of numismatists for this chapter and would like to thank Joe Cribb and Robert Bracey in particular for their generous advice and suggestions. All mistakes, misunderstandings, and omissions are my own.

2. Agriculture is found here, but on a relatively small scale compared to Outer Eurasia. Cities are also found here.

3. Christian (1998: xxi).

4. See Chang et al. (2003) for a discussion of the Yuezhi as agropastoralists—farmers and herders.

5. Some have linked them to an Indo-European group known as the Tocharians, but this remains speculative.

6. The Yuezhi were renowned for their army of skilled archers, but the Xiongnu had mobile mounted warriors.

7. For a discussion of these, see Thierry (2005). For a detailed account of the Yuezhi migration, see Benjamin (2007). The historical record is always subject to support or otherwise from the archaeological record. They do not always tally, especially in the indiscriminate use of Xiongnu in the Chinese histories, where the archaeological record suggests different cultures.

8. Mainly in juan 123 of the Shiji and juan 96A of the Hanshu. Various translations are available for both, cited below.

9. Shiji 123, trans. Watson (Sima Qian 1993: 234).

10. Some probably stayed as well. A notable descendant of these was the Buddhist monk Dharmarakṣa (fl. mid-third century, Zhu Fahu). It is recorded that he came from a Yuezhi family that had lived for generations in Dunhuang.

11. And, by extension, as a major factor in the start of the Silk Road. See Whitfield (2018b) for further discussion of this.

12. As Holcombe has noted, “Three subjects that mainstream traditional Chinese historians seldom addressed were trade, Buddhism and foreigners” (1999: 285).

13. Shiji, trans. Watson (Sima Qian 1993: 234). The archaeological record suggests the Yuezhi were agro-pastoralists, but the nuances of the lives of different groups of the steppe were not necessarily acknowledged by the historians in China.

14. Ibid.

15. For a discussion of this interaction in relation to the Yuezhi, see Liu (2001).

16. As noted above, Chang et al. (2003) have studied contemporary settlements in southeastern Kazakhstan, supposedly the first stopping point of the Yuezhi, and have found evidence of an agro-pastoralist society.

17. See Michon (2015: 110–51) for a discussion of these.

18. There is considerable discussion about the equivalents of the place-names used by the Chinese. For example, Anxi is linked by some with Parthia and by others with the Parthian kingdom of Gondophares in Gandhāra.

19. Hou Hanshu 88, “Xiyu juan” [Chapter on the Western Regions], trans. Hill (2003).

20. Millward (2007: 15).

21. Christian (1994: 182).

22. “Writing Kusana history is like constructing a giant mosaic. Scholars have pieced together parts of the outer frame and a few internal configurations, but whole areas are still empty” (Rosenfield 2011: 10).

23. Lam (2013: 440).

24. Trainor (1997: 86).

25. By sea as well as land.

26. Holt (1988).

27. Dates for the Kushan and its early rulers are the subject of much debate, as discussed below, so no firm dates are given here.

28. For a recent discussion of the importance of coins for the historian, see Holt (2012), who calls them “the very backbone on which the frame of Central Asian history has been built” (31).

29. This is often asserted, but there is little evidence for these early-periods coins having a legitimizing purpose. Cribb, for example, is skeptical. For a discussion, starting from the coins of the early Yuezhi, see Michon (2015: 110–16). For a more general discussion, see Cribb (2009, esp. 500–503).

30. Michon (2015: 114).

31. Joe Cribb, pers. comm., May 2, 2016. Cribb also states that “the momentum of money is represented through the continuity of design. For money to continue to circulate it has to have some resemblance to pre-existing forms of money” (2009: 498) and that the debate on legitimacy has “obscured for many the primary function of the designs to enable the coin to be issued and used as money” (503).

32. “The copper coinage derived from silver and copper coinages of North India, Pakistan and Afghanistan current in the first century AD” (Bracey 2012: 188). Michon argues that the coins were part of a “process of ‘imaginative remaking’ of who he [Kujula Kadphises] was and who the Kuei-shang [Yuezhi] were” (2015: 130–31). Cribb argues that their diversity was to enable the coins to circulate rather than to assert legitimacy (see note 31).

33. Bracey (2009).

34. Falk (2015: 105–9). He further argues that the Kushan use of devaputra (son of god) to describe themselves was, at least in part, influenced by the use of divi filius on the coins of Augustus (107).

35. For a summary of discussions on this and alternative views, see Bracey (2009) and Falk (2015: 105–9).

36. Bracey (2009).

37. Göbl (1970).

38. Bracey (2012).

39. Bracey (2009).

40. As does Rosenfield, for example, when he writes, “It is difficult to imagine how the Kushans paid for the raw metal” (2009: 21).

41. Pliny the Elder, Natural History 6.26, trans. Bostock (1855).

42. Rosenfield (1967: 21).

43. Falk (2015). Marshall (1951: vol. 2:620) suggested sources in Dardistan and Tibet.

44. For the treasure, see Sarianidi (1985). For the archaeologist’s assertion that this was an early dynastic necropolis of the Kushan, see Sarianidi (1990–92: 103).

45. See S. Peterson (2017: 47–58) for a review of the evidence, including recent studies of the coins.