The revise the paper of the Eurasian nomad history

THE SCYTHIC AND HUNNIC ERAS: 1000 BCE-SOO CE BARBARIAN INVASIONS BEFORE 500 CE

acquired many aspects of the traditional steppeland cultures. According t Wolfram, by the fourth century, 'the Scythization of the eastern GOths i completed: the armored lancer, who covered incredible distances and fought on horseback; the practice of hunting with falcons; shamanism; the adoption:;f by the royal family of the Sassanian royal vestments; in short, the life-style of~ the Iranian-Turkish peoples of the steppe became part of the Gothic World. '4\

Late in the fourth century, there is evidence of at least one powerful dom in the Pontic steppes. Ammianus, writing in 395, described the and extensive realm' of a Gothic group called the Greuthungi, whose leader:, ~, was Ermanaric, 'a warlike king whose many heroic exploits had made a terror to his neighbours' .42 Jordanes' account, which relies heavily Ammianus, can be interpreted to imply that Ermanaric ruled an reaching as far as the Urals and the Baltic. If so, it would have been an ephemeral predecessor to the later state of the Rus'. However, as Heather has argued, it is unlikely that an empire this extensive would have left so little trace in the written sources. The archaeology of the Chernyakhiv Culture, suggests, instead, that there were as many as six separate fortified centres around the shores of the Black Sea, each of which may have been the capital of a powerful regional 'king'.43 As in the Scythic period, these kingdoms probably took tributes from local peasantries, and taxed the trades passing through the northern silk roads or from the northern forests. As Wolfram has suggested, Jordanes' account of Ermanaric's bitter conflict for control of the lower Don may reflect attempts to control the trades through the 'Cimmerian Bosporus,.44 It may be that Ermanaric claimed suzerainty over a number of semi-autonomous regional kingdoms in a loose political confeder­ ation similar to that of Athanaric. Such a confederation, like those of the Scythians and Sarmatians, would have combined considerable regional autonomy with the ability to mobilize rapidly in a crisis, and it might explain the claims Ammianus makes for Ermanaric's empire, despite its limited impact on Rome.

If this account is accurate, it suggests that the Goths, despite their agrarian background, created political systems typical of the Pontic steppes. These were based on the exploitation of local farming populations, and regional trade routes, but were never quite wealthy or powerful enough to extort the wealth of Rome as the Hsiung-nu had extorted the wealth of China. The great difference between the Goths and their pastoralist predecessors and successors was that the Goths were farmers, and if they failed to set up viable frontier states, they had the option of entering the agrarian world. After the disasters of 376, this was exactly what they did, with spectacular results.

HUNS: 370s-450s CE The arrival of the 'Huns' in the 370s, transformed the siruation along the western borderlands. For the first time in many centuries, the inhabitants of the Pontic region encountered a large pastoralist army in its most mobile and warlike phase. The terror that this first encounter generated amongst Goths and Romans still reverberates in the accounts of Ammianus or Jordanes.

Most modern historians accept that the 'Huns' were, in a loose sense, descendants of Hsiung-nu tribes who had migrated west after the collapse of the Hsiung-nu empire. However, the connection may be quite distant, and the Huns seem to have preserved no memory of their imperial antecedents in Mongolia. The name, 'Hun', certainly seems to be of Altaic origin (Turk. kiln, meaning 'people', Mong. kiimiin, meaning 'man'). The few fragments of European Hunnic that survive show different linguistic elements, but ],iongolian and Turkic seem to predominate. As Golden puts it: 'We do not possess an unbroken chain of evidence that directly links the Hsiung-nu to the Huns. But such a chain seems likely. The Hsiung-nu core (whether ethnic or political is unclear) remained, retaining its prestigious name while undergoing ... ethnic and very likely lin'guistic changes. ,45

After the collapse of the Hsiung-nu empires, many Hunnic groups mixed with Turkic, Ugric, Siberian and Iranian elements as they shifted westwards. This migration ended the ancient linguistic domination of the Kazakh steppes by speakers of Indo-European languages. There emerged in Kazakhstan a large, eclectic group of tribes mainly of Hunnic or Turkic origin, but with many other linguistic elements, and strongly influenced by Chinese, Iranian and Central Asian culture. 46 The earliest western report of those Hunnic groups who moved even further west places them between the Urals and the Aeal Sea in c.160 CE. Here Hunnic tribes may have settled for over 200 years, long enough to lose all memory of their imperial past. In the middle of the fourth century some were driven even further west, perhaps by the growing power of the Juan-juan confederation, which had driven the Chionites into Central Asia in the same period.

After 360, Hunnic tribes crossed the Volga river and defeated the Alanic confederation of Sarmatian tribes, that dominated the eastern Pontic region. Once in control of the lands north of the Caucasus, the Huns, like the Scythians a millennium earlier, began to raid into northern Mesopotamia. There were huge Hunnic raids into the Sassanian empire in the 360s. In the 390s, there were massive Hunnic invasions of Syria and Mesopotamia, which may have been as destructive as the Scythian raids of the seventh century BeE. Many Huns settled permanently in the north Caucasus, where their descendants converted to Christianity in c.520Y

Other Hunnic tribes moved west. In c.375, an army of Huns, with Alan allies, crushed the empire of Ermanaric, and conquered the Pontic steppes. According to Ammianus, Ermanaric committed suicide. 48 His son died in battle, and his grandson, Vitheric, became leader of the demoralized Greuthungi. However, as he was a minor, it was two other leaders, Alatheus and Saphrax, who led a large contingent of Greuthungi west where they eventually settled in the Hungarian plain. Many eastern Goths clearly remained in the east under Hunnic suzerainty, for Gothic artefacts appear in the Pontic steppes throughout the next century. They undoubtedly paid tributes and served in Hunnic armies, but otherwise life may have con­ tinued much as before. Goths played an important part within the Hunnic confederation. Priscus reported that Attila's empire was led by Attila and his family, then by leading Hunnic nobles, and then by Gothic nobles. 49

226 227

THE SCYTHIC AND HCNNIC ERAS: 1000 BeE-SOO CE

In Crimea, Gothic kingdoms survived for a long time, as Scythian kingdoms had before them.

In 376, under Hunnic pressure, many groups of Tervingi and Greuthungi sought permission to cross the Danube and settle in the Balkans. The scale of the crisis gave the Roman emperor, Valens, little choice. However, with much of his army fighting in Mesopotamia, he and his officials botched what turned into a huge and dangerous refugee crisis. Ammianus describes the desperate crossing of the Danube:

Once the emperor's permission to cross the Danube and settle in parts of Thrace had been granted, the work of transportation went on night and day. The Goths embarked by troops on boats and rafts and canoes made from hollowed tree-trunks. The crowd was such that, though the river is the most dangerous in the world and was tl1en swollen by frequent rains, a large number tried to swim and were drowned in their struggle against the force of the stream. 50

The Roman armies failed to police the newly arrived Gothic contingents. Eventually, the corruption of Roman generals (Ammianus claims that local commanders sold dogs to the starving refugees in return for slaves), and their failure to provide food relief and to help the newly arrived Goths settle, led to a massive Gothic revolt within the Balkans. This culminated in the humili­ ating Roman defeat at the battle of Adrianople in 378 CE, when the emperor Valens was killed. 51

After 376, the Huns emerged as the major power along Rome's Balkan fron­ tiers. According to Ammianus, they had no overall leader when they first attacked eastern Europe. They probably lived in independent but allied tribes of several thousand people, each of which could probably field up to 1000 sol­ diers.sz Ammianus' account, though stereotypical in some of its detail, suggests that in this period the Huns practised a highly nomadic form of pastoralism, which is exactly what we would expect during a period of mass migrations.

They have no buildings to shelter them, but avoid anything of the kind as carefully as we avoid living in the neighbourhood of tombs; not so much as a hut thatched with reeds is to be found among them. They roam at large over mountains and forests, and are inured from the cradle to cold, hunger, and thirst. ... They wear garments of linen or of the skins of field-mice stitched together, and there is no difference between their clothing whether they are at home or abroad. Once they have put their necks into some dingy shirt they never take it off or change it till it rots and falls to pieces from incessant wear.... [They] remain glued to their horses, hardy but ugly beasts, on which they sometimes sit like women to perform their everyday business. Buying or selling, eating or drinking, are all done by day or night on horseback, and they even bow forward over their beasts' narrow necks to enjoy a deep and dreamy sleep.53

They were, he wrote, hostile to agriculture or sedentism.

None of them ploughs or ever touches a plough-handle. They have no fixed abode, no home or law or settled manner of life, but wander like refugees with

BARBARIAN INVASIONS BEFORE 500 CE

the wagons in which they live. In these their wives weave their filthy clothing, mate with their husbands, give birth to their children, and rear them to the age of puberty. No one if asked can tell where he comes from, having been conceived in one place, born somewhere else, and reared even furrher off. 54

After their initial attacks, Hunnic tribes raided and probed along Rome's frontiers, but also proved willing to act in alliance with the Empire. For almost a quarter of a century, most were content to remain in the Pontic steppes. Then, in the first 20 years of the fifth century, groups ofHuns began to move around the Carpathians and into modern Hungary.55 As they con­ solidated their hold first on the Pontic and then the Hungarian steppes, they presumably settled into a pattern of 'Semi-nomadic migrations. Clan and tribal groupings established regular migratory routes, new chiefly hierarchies emerged, and close relationships of trade or tribute-taking were established with subordinate agriculturalists and towns. 56 Regional Hunnic chiefs proba­ bly took over the trading and tributary systems of the Gothic kingdoms. They certainly showed great interest in cross-border trade, purchasing grain and weaponry from Rome. Hunnic elites were particularly interested in the prestige goods and coins they received in tribute, for these allowed them to purchase luxury goods that enhanced their status and influence amongst their own followers. Finds of Hunnic cauldrons along the Kama river, and Attila's gift of furs to visiting ambassadors, suggest that Hunnic tributary systems, like those of the Scythians, reached deep into the forest regions of western Inner Eurasia. 57

The Huns soon adapted to the agrarian world of eastern Europe, and their own settlements became more sedentary. When he visited Attila's court in Hungary in 449, Priscus, our second maior source on the Huns, described a world of pastoralist elites settled in pastureJands surrounded by farming regions. Priscus passed through many villages on his way to visit Attila. Eventually, he reached Attila's camp, which was a relatively permanent set­ tlement in the middle of a treeless steppe in the Hungarian 'Alfbld', probably east of the river Tisza. Priscus described it as 'a village, ... like a great city, in which we found wooden walls made of smooth-shining boards, whose joints so counterfeited solidity that the union of the boards could scarcely be distin­ guished by close scrutiny. There you might see dining halls of large extent and porticoes planned with great beauty, while the courtyard was bounded by so vast a circuit that its very size showed it was the royal palace.'58 Inside the palisade of Anila's palace were many circular wooden dwellings, mod­ elled, presumably, on nomads' tents. Inside the hut of Attila's wife, Kreka, Priscus found a scene typical of many a wealthy pastoralist household, with the lady seated on felt mats, while her servants embroidered fine linen cloths. The wood for these royal palaces and their largely decorative palisades was all imported, as the nearby steppes were treeless. Nearby, was the house of Anila's lieutenant, Onegesius, which was distinguished by the presence of a stone bath-house, built from specially imported stone by an architect cap­ tured in Sirmium. 59 As Pletneva points out, the structure of Anila's steppe settlement, with its concentric squares, rectangles or circles, and the royal

228 229

THE SCYTHIC AND HUNNIC ERAS: 1000 BCE-SOO CE

court at the very centre, was typical of many pastoral 'towns' that grew up around large winter camps.60

In the early fifth century, there emerged a more organized Hunnic polity. In 422, Hunnic tribes resumed large-scale attacks on the Balkans. In the 430s, these attacks were renewed under a leader called Rua, and his brOth_ ers, Mundiuch and Octar. In c.433, Rua demanded the return of Hunnic 'fugitives' from the Black Sea region, who were serving Constantinople. This suggests that Rua now claimed suzerainty over at least the western parts of the Pontic steppes. Indeed, Rua may have been the real founder of Hunnic

61 power. After invading Pannonia and Thrace, Rua concluded a treaty under which the Huns were to receive 350 pounds of gold a year.

Rua died in 432 or 433. To prevent a planned Hun attack on Roman terri­ tory, Emperor Theodosius 11 (408-50) raised the annual tribute to Rua's successors to 700 pounds of gold, and agreed to open border markets.62 Enriched by the flow of Roman tributes, Rua's nephews, Bleda and Attila, sons of Mundiuch, maintained the authority he had established over both Hunnic and non-Hunnic tribe's.63 They spent the next years consolidating and extending their authority. Priscus refers to warfare in Scythia, and even to the possibility that the Huns extended their authority to the Baltic shores. In 433, Pannonia was surrendered once more to the Huns. In 441, when Constantinople's armies were engaged in North Africa and Syria, Hunnic armies invaded the Balkans, citing Constantinople's refusal to return 'fugi­ tives' and failure to pay tributes in full. They devastated several major cities, including Sirmium and Singidunum (Belgrade), and left with huge amounts of booty and captives. In 443, Hunnic armies sacked the base of the Danube fleet at Ratiara, and the populous cities of Naissus and Serdica (Sofia) and crushed imperial armies in the Balkans. Travelling through Naissus six years later, Priscus found the river bank still covered with the bones of the dead. 64

Theodosius 11 had to negotiate humiliating terms to prevent the fall of Constantinople itself.

In c.445, Attila murdered his brother, Bleda. He became for eight years the sole leader of a Hunnic confederation that controlled much of modem Hungary and Romania, most of the steppelands north of the Black Sea, and even had considerable power in Germany and parts of Gaul. Attila's ambi­ tions probably extended even further, for, according to Priscus, he intended, after subordinating the Romans, to turn on the Persians, following the route of earlier Hunnic armies through the Caucasus. As he built up his power, he showed extreme sensitivity to issues of honour and prestige, for these were important matters to a ruler whose diplomatic status did not yet match his real power. Priscus reports many incidents showing Attila's extreme interest in the return of fugitives who had betrayed him, and in the exact status and rank of diplomats sent to him from Constantinople. 65

The brief glimpses of Attila contained in Priscus' account suggest that he wielded autocratic authority over his immediate followers. His power depended on the huge flows of revenues, through tributes, ransom payments and sheer booty, that Hunnic leaders had concentrated in their hands since the 420s. These he used to bind regional chiefs into loyal retinues. Under

BARBARIAN INVASIONS BEFORE 500 CE

Attila, powerful chieftains acted as members of their master's personal retinue, and were liable for any military or diplomatic service on which he might send them. The loyalty of Attila's retinues depended on his ability to reward them abundantly with the goods seized in imperial raids, with the prestige of noble rank, clothing and foodstuffs, and the privilege of going on embassies to Constantinople, during which they could expect to receive valuable gifts. As Thompson puts it: 'Arrila's endless demands that Hun ambassadors should be received with "gifts" at Constantinople is not merely to be ascribed to a politique de prestige on his part ... ; it was a vital support of the social order then prevailing among the nomads.'66 However, within their own tribal lands, there can be no doubt that such chieftains continued to wield great power, even if they were not Hunnic, for there is no sign of the existence of a central bureaucracy of any kind. At most, Attila enjoyed the services of Latin secretaries sent to him by his Roman ally, Aetius.G7

In 447, Hunnic armies, with Gothic and Gepid allies, invaded the Balkans once more, and sacked large areas. Theodosius 11 had to raise the annual tribute to 2100 pounds of gold, to pay arrears of 6000 pounds immediately, and to evacuate a no man's land between the two powers. Though consider­ able sums, these tributes represented just over 2 per cent of the 270,000 pounds of gold received by the Empire each year in the fifth century. They were certainly not enough to bankrupt the empire, as Mommsen claimed. 6B

They proved, however, that for a brief time, the Huns could play the 'Outer Frontier Strategy' in the west as their ancestors had played it in the far east. At his funeral, Arrila's followers praised him, among other things, because he 'terrorized both empires of the Roman world after conquering their cities, and, placated by their entreaties that the rest might not be laid open to plun­ der, he accepted an annual tribute'. 69

In 451, Attila led an army westwards to attack the Visigothic rulers of Gaul. Though Jordanes ,claimed the army numbered 500,000, in reality it was probably no larger than 100,000, the majority of whom were probably Goths. At the last minute, the western Roman emperor, Valentinian, decided to support his old enemies, the Visigoths, and Attila found himself facing a combined Visigothic and Roman army. His armies were defeated in June 450, during a momentous battle near Troyes, but allowed to escape east. In 452, Attila invaded Italy and destroyed many of the great cities of the north. This time, too, Attila turned back, probably because of an outbreak of plague, shortages of supplies in a land recently devastated by famine, and the fact that a Byzantine army had entered the Hunnic lands of Pannonia. 7o

A year later, in 453, Attila was dead of a hemorrhage, and his empire began to unravel.

His followers gave him the funeral, or strava, of a great steppe ruler. According to Jordanes (who r~lied on Priscus here): 'His body was placed in the midst of the plain and laid in state in a silken tent as a sight for men's admiration. The best horsemen of the entire tribe of the Huns rode around in circles, after the manner of circus games, in the place to which he had been brought and told of his deeds in a funeral dirge.' His followers cut their faces as they rode around the grave, buried their chief with many precious

230 231

--- -----THE SCYTHIC AND HUNNIC ERAS: 1000 BCE-SOO CE objects, and then slaughtered those who had buried him to keep the bUrial site a secret. 71

Attila's empire now fell apart, as his sons quarrelled over his inheritance. Unlike Motun, Attila had established no durable structures to hold the empire together. This may be partly because the Hungarian steppes can sup_ port smaller numbers of pastoralists, so that the Huns were greatly Outnu!ll_ bered by subject tribes. Or perhaps it was because he had less understanding than Motun of the importance of bureaucratic structures and rituals. With Attila's death, the flow of tributes and booty began to dry up, undermining the willingness of subordinate chieftains to follow new and untried leaders. 72

The Germanic tribes revolted; a Hunnic army led by some of Attila's Sons was defeated in 454; and the Huns soon had to leave Hungary and flee to the Pontic steppes. Attila's empire broke into its many tribal segments. In 468, a mere 13 years after Attila's death, one of his sons, Dengizikh, demanded the opening of frontier markets along the Danube for Hunnic merchants, a sign, perhaps, of a serious shortage of prestige goods, weapons, and perhaps even foodstuffs. When the Byzantine government refused to reopen the markets, Dengizikh invaded but was defeated and killed, and his head was handed over to the imperial commander, Anagast, in Constantinople. Small, mixed groups described as Huns survived along the Danube frontiers for another decade or two, acting as mercenaries, petty raiders, or even settlers. But most 'Huns' returned to the Pontic steppes, where they merged with other pastoralist groupS.73

Attila's career suggests that. the 'Outer Frontier Strategy', worked so successfully by Motun, was harder to manage in the west. This was because of the remoteness of Rome, Constantinople and the empire's wealthier cities, the narrowness of the Balkan frontier, and the lack of good pasturelands west of the Carpathians. 74 Here, pastoralist rulers had to rely mainly on the wealth of local populations of farmers, forest-dwellers and pastoralists, while staying in the steppelands of Hungary or the Pontic steppes.

THE NORTH CHINESE BORDERLANDS AFTER THE

HSIUNG-NU: 220--550 CE

Whereas the history of central and western Inner Eurasia in the early centuries of the modern era was driven by the migratory movements started by the Hsiung-nu, in the eastern steppes, pastoralists were sucked into a vacuum left by the collapse of empire on both sides of the great wall. There followed an extended period of complex politics conducted between weak and unstable political confederations until, in the sixth century, new imperial dynasties appeared in China and in the steppes.

After the fall of the Han, the eastern steppes were dominated by a series of weak tribal confederations, most of which adopted primitive strategies of raiding, so that conflict was endemic. However, as Thomas Barfield has shown, beneath the complex politics of the period it is possible to detect a pattern of events which tended, eventually, to restore political power both in

BARBARIAN INVASIONS BEFORE 500 CE

China and in the steppes. First, there appeared in the borderlands regional states ruled by pastoralists with some understanding of the agrarian world,

75 as well as dynasties ruled by agrarian 'barbarians' similar to the GothS. Manchuria played a crucial role at this stage, for its remoteness and eco­ logical variety made it possible to experiment with symbiotic systems com­ bining pastoralism and agriculture, in an environment of relative security. Eventually, as these polities began to depend more and more on the wealth of their Chinese populations, and the skills of their Chinese officials, they became Sinicized or were replaced by indigenous Chinese dynasties. As powerful new Chinese dynasties emerged, it became possible, once again, for powerful steppe leaders to contemplate some form of 'Outer Frontier'

strategy.

THE NORTH CHINESE FRONTIER LANDS IN AN ERA OF

ANARCHY

Chinese sources describe most of the tribes along the north Chinese borders after the collapse of the Hsiung-nu as Hsien-pi. The Hsien-pi tribes came from western Manchuria, and had once been part of the Tung-hu confed­ eration over which Motun had his first great victory. However, the name certainly masks great ethnic and linguistic diversity and many Hsien-pi were probably Hsiung-nu under a new name. The language of the dominant Hsien-pi tribes was probably close to later Mongolian and Kitan.

76

Though the Hsien-pi were more likely to practise agriculture than the Hsiung-nu, particularly near Manchuria, it is unlikely that their lifeways were very different from those of the Hsiung-nu era. Here is a Chinese description of the life of the Hsien-pi and the closely related Wu-huan:

They are experts in riding and shooting and following the waters and grass to pasture their animals, and they have no definite place to live. They use yurts as their homes, and all the yurts point towards the east. They hunt birds and animals every day. They eat meat, drink milk, and wear wool and skins .... They depend on the Middle Kingdom for their supply of seed and grain. The ta-jen leaders are able to make bows, arrows, saddles and bridles, and to forge metals to make arms. They are able to dress hides, embroider, and weave woollen clothes. 77

Though the Turkic title of 'kaghan' is first recorded amongst the Hsien-pi in the third century CE, they never formed a durable pastoral state.

78 This

meant that Chinese officials had to deal with a multiplicity of local chiefs. As early as 49 CE, Han officials paid Hsien-pi chieftains bounties for the heads of northern Hsiung-nu. The Hsien-pi brought Hsiung-nu heads to markets in Liao-tung where they received gifts and cash. Many Hsien-pi tribes also received cash subsidies. After the defeat of the northern Hsiung-nu in 89 CE, many Hsiung-nu tribal leaders became Hsien-pi, and the Hsien-pi became the dominant peoples of Mongolia. However, the collapse of the northern Hsiung-nu deprived the Hsien-pi of the bounties on Hsiung-nu heads. As a result, Hsien-pi tribes began once more to raid northern China.

79

232 233

---THE SCYTHIC AND HUNNIC ERA.S: 1000 BeE-SOO CE difficulties Soviet historians faced in fitting Central Asia of this period into the Marxist 'slave mode'.

20 Mukhamedjanov, 'Economy and social system', in f/CC4, 2:268-70 and 265. and see Tolstov, Po drevllim de!'lalll, pp. 94-5. '

21 Ocherki iSlOn'i SSSR: PervobyT1lo-obshchiIlIlYI slmi, p, 540; and see Mukhamedjanov 'Economy and social system', in HCC4., 2:268-74; on the historiographieal debate, see Gafurov, Tadc-,hiki, 1:224-30,

22 Liu, AIlCie12l IlIdia alld AIlcle12l China, pp. 7-9; Gafurov, Taddllki, 1:205-7. 23 Tolslov, Po drevllim del'cam, p. 225; see also Narain, 'Indo-Europeans in Inner

Asia', in CHEIA, p. 163; Golden, ImroducTioll, p. 56; Mukhamedjanov, 'Economy and social system', in HCCA, 2:284-7; Ocherkl iSTOnl SSSR: PmJObylno-obshchinnyi slroi, pp. 539, 546.

24 Cited by B. A. Litvinsky, 'Cities and urban life in the Kushan kingdom', in HCCA (pp. 291-312), 2:303, and see 2:302,309; see also Litvinskii in Gafurov, Tadzhiki, 1989, 2:351-2.

25 On this system of fortifications, probably built by Khosrow I, see Frye, Golden Age of Persia, pp. 13-15, and Frye, 'The Sasanian system of walls for defense', in Islamic Iran alld Celllral Asia, HI: 12-13; see also Artamonov, ISloriya khazar, pp. 117-26, which includes a plan on p. 121; according to Novoscl'tsev, Khazarskoe gosudarslVo, p. 101, recent excavations have shown fortifications here dating from the eighth to seventh centuries BCE; on Darband's history, see Bartol'd, Sochmnziya, 3:419-30 and see Barthold, A HislOry of lhe Turkman People, in Four STudies (pp. 75-170), p. 87.

26 Banhold, TurkesTan down TO The Alongol Invasion, p. 181, and p. xxix; on the decline of Samarkand before the late fifth century, see Shishkina, 'Ancient Samarkand', pp. 90-3; and B. 1. Marshak, 'Sughd and adjacent regions', in HCCA (pp. 233-58), 3:233-58; Azarpay, Sogdian Paiming and Belenitsky, Cemral Asia.

27 Chegini and Nikitin, 'Sasanian Iran', in HCCA, 3:38; on the complex debates over the origins of the Chionites, see Gafurov, Tadzhiki, 1:255-8.

28 Citations from Sung Yiin from Beal, Si-yu-ki: BuddhiST Records, I:xci; see also Menander, The HisTOry, p. 115; Frye, Golden Age of Persia, p. 38; Golden, IlIlroduceioll, p. 80-2; according to al-Biruni, Kunduz was their capital, Vogelsang, Rise and Organisalion, p. 62; see also E. V. Zemal, 'The Kidarite kingdom', in HCCA, 3: 119-34, Litvinsky, 'The Hephthalite empire', in HCCA, 3: 135-62, and Litvinsky and Zamir Safi, 'The later Hephthalites', in HCCA, 3: 176-83; see also Chavannes, Documenls sur les Tou-Kiue, pp. 224-6.

29 The best account of Gothic history is now Heather, The GOlhs; on Jordanes, see pp. 9-10.

30 See Jordanes, The GOlhic HislOry, pp. 57-8; also, Heather, The GOlhs, pp. 21-49; Wolfram, HislOry of lhe GOlhs, pp. 12, 23; and Kazanski, Les GOlhs, pp. 19, 25.

31 Heather, The GOlhs, pp. 46-8. 32 Wolfram, HislOry of lhe Goths, pp. 48, 51-5; Brown, The Makillg of Late

Amiquily, pp. 22, 25. 33 Heather and Matthews, The GOThs, p. I, on Gothic mercenaries; and see Bums,

Barbarians, pp. 19-20, 22, and map, p. 20; Burns, HislOry of lhe OSlrogolhs, pp. 32-4; Wolfram, HiSTory of The GOThs, p. 56.

34 Wolfram, HislOry of The Gorhs, p. 10 I; and Kazanski, Les GOlhs, p. 25; see also Heather and Matthews, The GoThs, p. 98, and pp. 57-8 and 63.

35 Burns, HislOry oflhe OsTrogoThs, pp. 139-42, and pictures on pp. 140-1. 36 Heather and Matthews, The GOlhs, p. 90 and pp. 82-4; on combs, see Heather,

The GOlhs, pp. 81-3.

BARBAIUAN INVASIONS BEFORE 500 CE

37 Heather and Matthews, The GOlhs, p. 93, and sec pp. 71, 91. 38 Heather, The G01hs, pp. 60-2, and see p. 58. 39 Heather. GOThs "lid Rmnans, p. 107; 'W'olfram, Hislory oflhe GoThs, p. 62,

40 Zosimus, New HislOrv, tr. Ridley, bk 4:11, p. 75. 41 Wolfram, Hisror" oflhe GOlhs, p. 115; on use of the term, 'Scythian', see p. 11. 42 Ammianus lvlarcellinus, Laler Roman Empire, p. 415. 43 See, e.g., Burns, HIslO~\' of lhe OSlrogolhs, p. 38; Jordanes. The GOlhic HisTorl',

pp. 841'1'; Heather, The GOlhs, pp. 55-7 and see p. 53, for Jordanes' debt to

Ammianus. 44 Wolfram, History of lhc GOlhs, pp. 87-8. 45 Golden, IntroduceiOlI, pp. 87, and see p. 89; see also Golden, Khazar Studies,

1:28; Maenchen-Helfen, World of lhe HUllS, insisted that there was no connection between the two, but most modern specialists are less sceptical.

46 Altheim, Ani/a, ch. 2; Golden, Introduclion, p. 86; Bona, Hwmell7'eich, p. 29. 47 Golden, Introduceion, pp. 107-8; Maenchen-Helfen, World of lhe HUllS, pp. 43-4;

and sec Sinor, 'The Hun period', in CHEIA (pp. 177-205), p. 182; and

Artamonov, ISlOriya khazar, p. 53. 48 Ammianus, Laler Roman Empire, 3.2, p. xxxi. 49 Burns, History of the OSlrogolhs, pp. 46-7 and see pp. 53-5; and Bona,

Humlellreich, pp. 12-13. 50 Ammianus, Laler ROmall EmpIre, bk 31, section 4, p. 417. 51 See Burns, HislOry of lhe OSlrogolhs; on this critical period, Ammianus, Laler

Roman Empire, bk 31, sections 4 and 5, pp. 416-19; and Heather, GOlhs and

Romans, pp. 128-35. 52 Thompson, HislOry of Anila alld The HUllS, pp. 48-56; Ammianus, Larer Roman

Empire, p. 412; however, Bona, Hwmenreich, pp. 25-9, argues that Ammianus, who had no direct knowledge of the Huns, underestimates the power and

sophistication of their political structures. 53 Ammianus, Laler Roman Empire, pp. 411-12.

54 Ibid., p. 412. 55 Heather, The GOlhs, pp. 102-8. 56 Pletneva, Kochevniki, p. 44-5. 57 Bona, Hunnenreich, p. 40; and see Thompson, The Huns, pp. 177-82. 58 Cited by Jordanes, The GOlhic HislOry, p. 101; on the site of Attila's capital, see

the discussion in Thompson, The Hum, pp. 276-7. 59 Gordon, ed., Age of Allila, p. 84, and see p. 90; almost 800 years later, the

Franciscan missionary, William of Rubruck, was as surprised as Priscus to find

European craftsmen in a barbarian capital. 60 Pletneva, Koche~J!liki, p. 45. 61 See, e.g., Thompson, The Huns, p. 227, but see also, pp. 84-5 on the limits of

Hunnic control in the Pontic steppes; and see pp. 79-80; see also Sinor, 'The

Hun period', in CHEIA, p. 187. 62 Sinor, 'The Hun period', in CHEIA, p. 187-8. 63 Attila's name may come from the Turkic word for the Volga, 'Itil' (Golden,

Imroduceioll, p. 90); if so, this may mean he began his career in the Pontic steppes. 64 Gordon, Age of Allila, p. 74; and see Thompson, The Hum, pp. 71, 84, 89. 65 See, e.g., Gordon, Age of Anila, p. 94; see also p. 91. 66 Thompson, The Huns, p. 192, and see pp. 178-9, 181-2, 187-9.

67 Ibid., pp. 122, 139. 68 Maenchen-Helfen, World of lhe HUllS, p. 180, and see p. 182; and see Sinor, 'The

Hun period', in CHEIA, p. 190; Thompson, The HUllS, pp. 99-101.

240 241

THE SCYTHIC AND HUNJ\:IC ERAS: 1000 BCE-500 CE

69 Maenchen-Helfcn, In,rld of thc Huns, p. 275; but (p. 276-7) he doubts the authenticity of the dirge which is from Jordanes.

70 Sinor, 'The Hun period', in CHEIA. p. 196; on the SilL' of tIle Hunnic armies Maenchen-Helfen, World ot the HUllS, p. 213. '

71 Jordanes, The Gothic History, p. 124. 72 See the superb analysis in Thompson, The Hurts, pp. 199-2U2, which is based, in

turn, on the ideas of Lattimore; on the ecological aspects of the problem, Sinor 'Horse and Pasture', pp. 182-3. '

73 Golden, Imroduction, p. 92. 74 Lindner, 'Nomadism, horses and huns', pp. 14-15 and Sinor, 'The Hun period'

in CREIA, p. 204. ' 75 Barfield, Perilous frontier, ch. 3. 76 Golden, IntroduClion, p. 69. 77 from the 'Book of We!', cited in Jagchid and Symons, Peace, Wa1; and Trade,

p. 170. 78 Golden, ImroduClion, p. 71; there have been no completely successful attempts to

trace the etymology of the term. 79 Barfield, Perilous Fromier, p. 86-8; it was traditional in the Chinese army to

measure the success of a campaign by the number of heads brought back; Jagchid and Symons, Peace, !.f7a y, and Trade, p. 178, point out that the need to fulfil quotas encouraged the Chinese to execute non-combatants.

80 Barfield, Perilous Fromier, p. 97, and see p. 89; there is a good account of his career in de Crespigny, Northenz Frontier, pp. 330-42; and see Ishjamts, 'Nomads in eastern Central Asia', in HCCA, 2; 156.

81 Honey, fo,1cdie·val Rsiung-Nu, p. 7; Barfield, Perilous FrO/aier, p. ID!. 82 On Liu Yuan, see Honey, Medieval Rsiung-Nu. 83 Barfield, Pen"lous Frontier, p. 103. 84 Ibid. pp. 104-5. 85 Barfield, Pen"lous Frontier, p. 107, and see pp. 98, 106, Ill. 86 Ibid., p. Ill. 87 This is Golden's conclusion; Introduction, p. 74. 88 Golden, Imroduction, p. 75. 89 Cited in Barficld, Perilous Fromier, p. 122. 90 Golden, Imroduetion, p. 75; on 'barbarization', see Yu, Trade and ExpanSIOn,

pp. 213-15. 91 According to legend, a white horse brought the first Buddhist texts to Lo-yang,

Liu, Anciem India and Anciem China, p. 85; see also Zurcher, The Buddhist Co"quest ot China, 1:28-38; Zurcher, 'Yueh-chih and Kaniska', p. 356; and see Puri, Buddhism in Central Asia, pp. 12 and 7; and Lirvinsky, 'Outline History of Buddhism in Central Asia', in Gafurov, et aI., Kushan Studies in USSR, p. 68; on Buddhism under the T'o-pa, see Ch'en, Buddhism in China, ch. 6.

92 Liu, A"cie11l China and Ancient India, pp. 145, 155. 93 Golden, Imroductio", p. 76-7; Sinor, 'Establishment. .. of the Turk empire', in

CREIA (pp. 285-316), 291. 94 Barfield, Perilous Fromier, pp. 120, 122. 95 Chavannes, Documents sur les Tou-Kiue, p. 221, and Sinor, 'Establishment ... of

the Turk empire', in CHEJA, p. 293. 96 Barfield, Pen"lous Fromier, pp. 124-6; Jagchid and Symons, Peace, War, and Trade,

pp. 170-1 describes one spectacular consignment of 'gifts' to the Juan-juan leader; on links with the Hephthalites and Avars, Golden, Introduction, pp. 76-8,

97 Golden, Introduction, p. 79.

BARBARIAN INVASIONS BEFORE 500 CE

fURTHER READING

f{CCA, vols 2 and 3, and CliEL1, are valuable for many parts of this chapter. 'I1lere is wom for a book on the Yueh-chih anu the Kushan Empire, a crucial, but long-neglected problem m World history. The best recent surveys are in HCCA, vo!. 2; CHE/A; Golden, lll,rodllCtlOIl to lhe HisrolJ' (~f the Trukt'c Peoples; and Frye:, Heritage ol Central Asia. Chinese sources on the Yueh-chih are summarized in Ziin.:herl 'Yiieh-chih and Kaniska" and accessible in Hulsewe, Chi1lo in Central ASlU, while other Chinese sources on this period are available in Chavannes, Documents sur Ics ](w-Kwe, and Beal, Si- Yu-kl. On Buddhism in Central Asia, see Puri, Buddhism, and Litvinsky~ 'Outline History of Buddhism'. On both trade routes and religious e:l<changes between India and China, see Liu, Am;icnt India aHd AncicHl China, On Central Asia before Islan1, see Barthold, Turkcsran; Staviskij, La Bacrn'anc; and Frye, GoldcH Age of Persia, and Herirage of Central Asia. On the Gaths, the Innst recent studies include: Heother and Matthcws, The Gorhs; Heather, Goths and Rm-nons , and the s.ame author's, The Gorhs; Kazanski, Les Goths; Wolfram, Hiswry of the Goths; and Burns, History of the ()strogoths. The main primary sources are AmIl1ianus Marcdlinus and Jordanes, though horh must he tre3ted with caution. On the Huns, see Maenchen-Helfen, World of the Hum; Thompson, A History of Attila and the Hum (the n;vised edition includes an essay by Heather bringing it up to date); and Altheim.> Art{{a und die HumlCn . Two brief surveys are Sinor, 'The Hun period', in CHE/A, pp. 177-20'5, and Sinor, 'The historical Attila'. On the archaeology of the Huns, see Bona, Humzenreich. The main primary sources are Ammianus Marccllinus and Priscus. There are translaIions of Priscus' account of his embassy to Anila in Gordon, Age of Attzla, and Blockley, 'J7lC Fragmel1tary Ciassicising Hislon"ans. See also, Lindner, 'Nomadism, horses and Huns', and Sinor, 'Horse and pasture\ nn the ecological difficulties the Huns faced in Hungary. Tbe most recent history uf Byzantium is Treadgold, A Hi,torv of rhe Byzanrine Srare and Society. On the eastern steppes in this period, the best overview is in Barfield, Perilous Fromicr. See also Jagchid and Symons, Peace, l¥lar, and Trade; Honey, The Ri~e nJ the Medie·val Hsiung-l\lu; and Golden, Introduction to the History of the Turkic Peoples.

Z42 243