Empires and Barbarians

by Peter Heather

  If we can’t get very far with numbers, the comparative migration literature does prompt several more general observations about the Hunnic expansion into central Europe. The first stage of activity recalls the way in which many better-documented migration flows build up on the back on the activities of ‘scouts’, which demonstrate to a broader population the benefits of relocation. And although not something observed in the modern world, even the en bloc migration of large Hunnic groups in phase two is in accord with the fundamental principle that migration units will be of a size and nature that are appropriate to the task of accessing wealth in the particular context in which the migrant flow is operating. For the same reasons we have met before, the kind of large-scale predatory migration eventually undertaken by the Huns also necessarily involved women and children. The numerous dependants of large military forces assembled from non-professional sources cannot be left behind in safety when the military activity encompasses any intent to migrate. As with so many of the other immigrants we have encountered, moreover, the Huns had established traditions of mobility which, all the comparative evidence again emphasizes, must have greatly facilitated their decision to respond to potential gains to be had from the Roman world by upping sticks and moving closer towards it. The biannual migrations common to the nomadic lifestyle meant that the Huns had a greater than usual capacity to organize large-group movement.

  As with the Goths, Vandals and Alans on Roman soil, another major reason why there was a substantial chronological gap between the two main phases of Hunnic intrusion into Europe must have been the need to build up geographical knowledge about the new possibilities that opened up for them after they had displaced Goths and Alans from regions north of the Black Sea. From this perspective, the massive Hunnic raid launched into the Roman and Persian Empires through the Caucasus in 395 can be seen as part of a learning curve. This caused huge disruption and attracted a great deal of coverage in Roman sources, not least because one group of raiders even got close to the Holy Land. But the raiders suffered heavy losses, and the experiment was never repeated.17 This does not suggest that the Huns themselves viewed the raid as a major success, and its relative failure may well have played a role in their eventual decision to move further west on to the Hungarian Plain rather than in any other direction. The knowledge of European geography necessary to make this move was no doubt also built up from feedback from the activities of smaller Hunnic groups we find west of the Carpathians before 405 – some of the mercenaries employed there in the 380s, for instance, or indeed the Huns of Uldin.

  As has been observed in so many other cases, moreover, the process of migration triggered major sociopolitical restructuring among the Huns. When Olympiodorus visited them in 411/12 he encountered, as we have just seen, a political structure based on a series of ranked kings, which was highly appropriate for a nomadic society. Economic logistics require nomad populations to be relatively dispersed. Bunched populations with herds would quickly lead to exhausted grazing and economic disaster. At the same time, subgroups need their own organization for matters such as settling disputes, and the larger group has to be able to act decisively as one on occasion, above all to protect the grazing rights upon which all depend. Well-organized devolution rather than centralized rule is a natural political form for nomadic societies, therefore, and a kingly hierarchy fits the bill nicely.18

  But when a second east Roman historian and diplomat – the famous Priscus – visited the Huns in the mid-440s in the time of Attila, the system of ranked kings had disappeared. Attila was surrounded by many great men, and although he had originally shared power with his brother, there were no other individuals of royal rank to be seen. No source records how the system of ranked kings was swept away, but one major bone of diplomatic contention between Attila and Constantinople was the protection it accorded to Hunnic fugitives of prominence.19 I take it, therefore, that Attila’s line, at the latest in the time of his uncle Rua who was active in the 430s, had ousted and/or demoted the other kings – a political process we have already observed among Alaric’s Goths, and will observe again among the Ostrogoths and Merovingian Franks.

  This all relates to migration, and for the following reasons. What went on in these cases, in broad terms, was that one leader came to monopolize the political support that used to be divided between several. This requires the successful leader to have access to unprecedented wealth so as to outbid his rivals in the patronage stakes and win over enough of their supporters, in the process forcing them either to leave the group or to accept more junior, non-royal positions. In the case of the Huns, the source of that new wealth was the profits that flowed from the new relationships they were able to develop with the Roman Empire. Putting yourself by hook or by crook in charge of distributing the combined profits flowing from a potent mixture of raiding, mercenary service and diplomatic subsidy was the shortest path to political triumph. Although this was surely not one of its envisaged aims, Hunnic migration to the Middle Danube naturally brought political revolution in its wake.

  In the qualitative sense used in migration studies, therefore, there is not the slightest doubt that the Huns’ intrusion into Europe in the later fourth and early fifth centuries must be considered mass migration. It was a flow of gradually increasing momentum, not a sudden, single migratory pulse, but the political shocks the Huns inflicted north of the Black Sea, and then in central Europe, could not be more obvious. Just as powerful, indeed, was the shock that eventually swept away their own political structures. Further analysis is limited by our inability to identify the precise trigger that set the Huns in motion. Roman sources highlight random chance, telling a charming story of wandering hunters who blundered though marshes, then to emerge into a land of plenty, but this is only a story and one based – again, like most of Ammianus’ digressions – on classical antecedent.20 In the absence of any other information, though, it may well be correct in suggesting that it was the wealth of the Roman Empire’s periphery that first sucked in the Hunnic raiders, and that migration momentum built up slowly from that point. New information on climate change or on political developments may transform this view in due course, making us redistribute the emphases we presently place on the various factors involved, but for the moment, the attractions of the wealthy imperial periphery seem the best option.

  Hunnic-era migration affected not just the Huns themselves, however, but also the many and varied peoples who made up Attila’s Empire. Everything suggests that the migratory motivations and processes that brought so many others to the Middle Danube in the period of Hunnic domination were very different from those of the Huns themselves.

  TRIBAL GATHERING

  Among Attila’s non-Hunnic subjects, Gepids, Sueves and Sarmatians had already occupied lands in the Middle Danube in the fourth century, long before the Huns arrived in Europe. Sueves and Sarmatians appear in Ammianus’ account of the Emperor Constantius II’s intervention in the region in 358, and Gepids are mentioned in other sources as inhabiting lands to the north-west of the old Roman province of Dacia (modern Romania) on the region’s eastern fringes. Their presence in the Middle Danube in the time of Attila is entirely unremarkable. The same is not true, however, of most of the other non-Hunnic components of Attila’s Empire. The fourth-century territories of Goths and Alans certainly lay east, not west, of the Carpathians, as probably did those of the Heruli, Rugi and Sciri. The best geographical fix we have for any of these latter three concerns the Heruli, who, though unmentioned in fourth-century sources, certainly occupied land north of the Black Sea in the third. The evidence for the Sciri is not so explicit, but in c.380, when the group name first appears in Roman sources, Sciri mixed with Huns attacked across Rome’s Lower – not Middle – Danubian frontier, which implicitly places them to the east of the Carpathians. We have no explicit information for the Rugi from the third or fourth centuries, but they had, like the Goths, formed part of the Wielbark system in the first and second. When we find them, again alongs
ide Goths, in the Middle Danube region in the fifth century, this raises the distinct possibility that they ended up there via a Gothic-style trajectory heading south-east to the Black Sea in the third.21 So like the Goths and Alans, Sciri, Rugi and Heruli had probably all moved west of the Carpathians into the Middle Danube basin only at some point in the late fourth or early fifth century.

  We have rather more information, in fact, about the Gothic contingent within Attila’s Empire. It came in a number of separate groups. One was dominated by the Amal family and their rivals, and rose to prominent independence in contemporary sources from the early 460s. A second was led in the mid-460s by a man called Bigelis, while a third remained under the tight control of Attila’s son Dengizich until the later 460s. Some of these Gothic groups may, like that of Radagaisus in 405, have moved into the Middle Danube region before the Huns established their dominance there in c.410. Others may have been moved there by the Huns at the height of their power. Still others perhaps followed a different trajectory. According to Jordanes, the Amal-led Goths moved west of the Carpathians only after the death of Attila in the mid- to late 450s, although they had acknowledged his overlordship in the 440s.22 Aside from all the migration implicit in the story, the Gothic evidence thus has a further dimension of importance. It warns us that each of the Hunnic subjects named in the sources may in fact have operated in a number of independent contingents, and that the history of their transfer west into the Middle Danube region may have been correspondingly complex.

  If the rise of Hunnic power, on the face of it, involved a great deal of human displacement into the Middle Danube, its unravelling generated an even longer procession of comings and goings. Prominent among the early departures was that of the Huns themselves. As more and more of Attila’s former subjects established their independence after the victory of the Gepids at the battle of the Nedao, where the chapter began, the military potential of the Hunnic Empire declined drastically and suddenly, to the extent that by the later 460s the two surviving sons of Attila – Dengizich and Hernac – concluded that life north of the Danube had become too precarious. They therefore sought new lands inside the eastern Roman Empire. Hernac was well received, obtaining land for himself and his followers in Scythia Minor, but Dengizich was defeated and killed by an east Roman army in 469, and his head put on display in Constantinople. By 470, within seventeen years of Attila’s death, the Huns had ceased to exist as an independent force in the trans-Danubian world: an astonishingly dramatic passage of history. And, in fact, the decisions of Dengizich and Hernac to move south of the Danube had been prefigured by some other refugees just a little earlier. Different sources record the intrusions on to east Roman soil, in or around 466, of another Hunnic group led by a certain Hormidac, and a Gothic one led by Bigelis.23 The circumstances are not known in any detail, but these moves clearly belong to that period when different population fragments originating from Attila’s Empire were looking to escape the fighting in the Middle Danube region.

  Not that the demographic history of the Middle Danube after Attila’s death was all about emigration. According to our one connected account of the action, provided by Jordanes writing in Constantinople in c.550 AD, it was only at this point – in the mid-450s – that the Amal-led Gothic group moved west of the Carpathians, under the leadership of a certain Valamer. The story was probably first written down at the court of Valamer’s nephew, Theoderic, King of Ostro-gothic Italy in the 520s, however, and Theoderic, born in the mid-450s was over seventy at this point, which does increase the story’s credibility. But Jordanes shows some uncertainty about Valamer’s early career (which we will come to later), so a degree of doubt must remain.24 But if the Amal-led Goths were new arrivals, this might explain why the struggles of the 460s seem to have taken the form of the other occupants of the Middle Danubian region – particularly Sciri, Sueves, Rugi and Gepids – all allying against them. Be that as it may, the net outcome of the struggle was more emigration. Not only was their failure to control this competitive conflict the fundamental cause of the Huns’ decision to seek asylum in the Roman Empire, but, very much in the same spirit, substantial numbers of Sciri, most famously Odovacar son of their defeated king Edeco, made their way into the west Roman Empire in the aftermath of a second major defeat by the Amal-led Goths in 469/70. Next to leave the region were the Goths themselves. Having – again according to Jordanes – defeated a coalition of their rivals in a bloody battle beside yet another unknown river in Pannonia, the Bolia, the Amal-led Goths moved into the east Roman Balkans in 473/4. This inaugurated a fifteen-year Balkan interlude during which Theoderic came to power. He succeeded his father Thiudimer, who had inherited control of the group when his brother Valamer was killed in an earlier battle against the Sciri. The Goths’ Balkan sojourn eventually came to an end in the autumn of 488, when Theoderic’s followers marched into Italy to create the Ostrogothic kingdom.25

  Not even the eclipse of the Huns, the destruction of the Sciri and the departure of the Goths were enough to end the struggle for mastery in the Middle Danube. By 473/4, three main powers were left in the region: the Rugi, Heruli and Gepids. The next major casualty was the kingdom of the Rugi. Settled north of the Danube opposite the former Roman province of Noricum in what is now lower Austria, the Rugi incurred the wrath of Odovacar, ruler of Italy from 476. In 486, he sent a major expedition north of the Danube which heavily defeated them and killed their king, Feletheus. This was the end of an independent Rugi kingdom, though a body of survivors under Feletheus’ son Frederic fled south into the Balkans to attach themselves to the following of Theoderic the Amal in 487/8. They subsequently joined in the Ostrogothic trek to Italy.26

  The demise of the kingdom of the Rugi left the Heruli and Gepids as the major powers in the Middle Danube, but now Lombard groups from Bohemia and the Middle Elbe began to move south into the region, initially into the lands previously dominated by the Rugi. In the first century AD, the heartlands of the Lombards were located in the Lower Elbe region, just south of the Jutland peninsula. The surviving written accounts of the intervening trek, which brought them to Lower Austria by the later fifth (Map 10), were written down only in the ninth century (hundreds of years afterwards) and are full of the kind of fanciful details that make it clear that no authoritative intermediate historical record underlies them. The first secure date we have for the Lombards’ progress south is provided by Roman sources, and comes with their entry into Lower Austria in 488/9 to take advantage of the power vacuum created by Odovacar’s destruction of the Rugi. Once there, Lombard power increased in two perceptible stages. In 508, first of all, their forces crushed the Heruli. The same late literary sources also report that the remnants of the Sueves were defeated and forced out of the Middle Danubian region at much the same time. The second extension of Lombard power came with the occupation of the old Roman province of Pannonia, south of the Danube. We are so badly informed that this may have occurred as early as the 520s or as late as the 540s, but the overall pattern is clear enough. Trickling in from regions along the Elbe, and most immediately from Bohemia, the Lombards made themselves the dominant power in the western half of the Middle Danube region by the second quarter of the sixth century.27 The Gepids, established further east, were now their great rivals.

  As for the Heruli, defeat in 508 caused an immediate split in their ranks. One subgroup moved away from the Danube altogether, ending up far to the north, in Scandinavia. A second sought refuge first among the Gepids. But the demands placed upon them by their hosts proved too heavy, and they quickly found an alternative sanctuary inside the eastern Empire, where the Emperor Anastasius granted them land on the Danube, sometime early in the 510s. There they stayed until the 540s, when the last surviving member of their royal clan died. Somehow they knew that the other group of Heruli had ended up in Scandinavia, and sent a mission there to find a suitable prince. It took so long to return, however, not least because the first-choice candidate died en route back to the Danube, that
the Emperor Justinian had picked out, at their own request, a new ruler for the remainder in the meantime. Civil war broke out when the Scandinavian mission finally returned, and the Danubian Heruli split again. One force remained inside the eastern Empire, the other returned to the Gepids. And in a subsequent war between the Lombards and Gepids, the Byzantines sent some of their remaining Heruli as military support for the Lombards. At that point, they found themselves up against their former comrades, who were fighting for the Gepids.28

  The fate of the Heruli, however, is no more than an appendix. With the rise of Lombard power, the revolution in the Middle Danube set in train by the rise of Hunnic power had finally worked itself out. An extremely complicated process in detail, it stretched out over pretty much a century, from the first arrival of the Huns west of the Carpathians in perhaps 410 AD to the defeat of the Heruli in 508. There is, however, an apparent logic to these events as they are reported in our various sources, which saw, first, a massing of militarized manpower on the Middle Danubian Plain orchestrated by the Huns, followed by an extended struggle for pre-eminence amongst their subject peoples after Attila’s death. Several of the combatant powers left the region as these struggles unfolded, so that the pattern of Middle Danubian affairs in the Hunnic era – many powers in close proximity to one another – eventually gave way in the sixth century to a division between Lombard and Gepid spheres of influence. Our primary interest, however, is in the migratory activity associated with these processes: largely, immigration into the region as Hunnic power increased, followed by emigration after Attila’s death, although the Amal-led Goths (possibly) and the Lombards (certainly) provide significant exceptions.29 That there was some movement of population would be denied by no one. Its nature and scale, however, are hotly contested.