In 510 or thereabouts, Theoderic the Amal, Ostrogothic King of Italy, wrote to the eastern Emperor Anastasius:
You [Anastasius] are the fairest ornament of all realms, you are the healthful defence of the whole world, to which all other rulers rightfully look up with reverence: We [Theoderic] above all, who by Divine help learned in your Republic the art of governing Romans with equity. Our royalty is an imitation of yours, modelled on your good purpose, a copy of the only Empire.
This looks like sucking-up in spades, but it wasn’t. In 507/8 Anastasius’ fleet had been raiding the eastern Italian coastline, and the Emperor had also provided diplomatic support for Frankish attacks on one of Theoderic’s main allies, the Visigothic King Alaric II. Against this backdrop, the really significant part of the letter comes in its next few lines:
And in so far as we follow you do we excel all other nations . . . We think that you will not suffer that any discord should remain between two Republics which are declared to have ever formed one body under their ancient princes, and which ought not to be joined together by a mere sentiment of love, but actively to aid one another with all their powers. Let there be always one will, one purpose in the Roman kingdom.
Theoderic’s initial flattery thus leads into a carefully crafted argument, which, in context, amounted to a diplomatic demand note. It can be paraphrased roughly as follows: ‘The eastern Empire is the model of complete divinely ordained rectitude, I follow it entirely; therefore I am the only other properly legitimate Roman ruler in the world, and superior – like you – to all the other successor state kings. You should be in alliance with me and not the Franks.’ And his pretensions were not greatly misplaced. He eventually came through the crisis of 507/8 with his position greatly enhanced. The Franks smashed the Visigothic kingdom at the battle of Vouillé in 507, but east Roman sea raids didn’t prevent Theoderic from picking up the pieces. From 511 he became sole ruler of both the Visigothic and the Ostrogothic kingdoms, comprising Italy, Spain and southern Gaul. He also ruled a chunk of the old Roman Balkans, exercised a degree of diplomatic hegemony over both the Burgundian and the Vandal kingdoms, and ran an alliance system that stretched up into Thuringia in western Germania. At the height of his powers, he dominated the western Mediterranean and ruled over a structure encompassing a good third to a half of the old western Empire. His ruling style – as the letter suggests – was also entirely Roman. He built Roman-style palaces, held Roman-style ceremonies in them, maintained the public amenities and even subsidized the teaching of Latin language and literature. The signals were easy to read. One of his Roman senatorial subjects hailed him in an inscription as ‘Augustus’, the most imperial title of them all.33
Faced with Theoderic in all his intimidating pomp, you could be forgiven for wondering if the fall of the Roman west to intrusive military powers from beyond its frontiers had made any real difference at all. In the second decade of the sixth century, forty years after the deposition of Romulus Augustulus, western Europe was still dominated by a Mediterranean-based imperial power. The fact that the edifice was under new management might seem neither here nor there. But appearances are deceptive. Theoderic’s revival of a Mediterranean-based western Empire proved entirely transitory. In the second half of the millennium the centre of imperial power in the west was to be located not in the Mediterranean at all, but much further north.
Empires of the Franks
First impressions might suggest that the Franks inherited Rome’s imperial sceptre through an entirely contingent sequence of events. Theoderic’s Gothic Mediterranean Empire failed to survive a succession crisis generated by his own death in 526. At that point, and contrary to the king’s wishes, the Ostrogothic and Visigothic parts of his Empire were redivided, each being ruled separately by different grandsons. And any lingering hopes that some future successor might rekindle the flames of the western Empire on the back of a Gothic power base were utterly extinguished in twenty years of warfare, starting in 536, when the east Roman Empire under the rule of Justinian (527–65) destroyed the Ostrogothic kingdom, as it had destroyed that of the Vandals in the early 530s. There was also a strong element of chance in Justinian’s decision to launch these western campaigns in the first place, since they were born of defeat in an earlier war against Persia that had left him desperately in need of a victory to shore up his waning prestige. Looked at more closely, however, the failure of Theoderic’s Empire reflects much more fundamental adjustments in Europe-wide balances of power – themselves the result of an interaction between half a millennium of development in barbarian Europe and the collapse of the Roman state.
By the late fourth century AD, before the great era of migrations began, the European landscape was marked, as we have seen, by massive inequalities of development. South of the River Danube and west of the Rhine lay the territories of the Empire, characterized by Europe’s highest population densities, the most developed exchange systems capable of supporting towns and considerable short- and long-distance trade, a relatively small and relatively rich landowning elite, and state structures of real power. These could mobilize the resources of the Empire’s territories to support such large-scale enterprises as professional armies, major building programmes and a governmental bureaucracy. Roman territory was bordered by its inner periphery of semi-subordinate client kingdoms, which generally received substantial trade privileges and diplomatic subsidies from the Empire but were subject to political interference and notionally owed certain – particularly military and economic – services to the Empire, which it was periodically able to extract.
These client states were part of the broader imperial system, but their relationships with the Empire were never entirely smooth. They often used force, or the threat of it, to attempt to maximize the financial benefits that could follow from a close relationship with the Roman state, and to minimize any accompanying exploitation. As we have seen, the Empire periodically launched campaigns to achieve exactly the opposite outcome. The new wealth that collected in the periphery via interactions with the Empire – trading, raiding and diplomacy – had also played an important role in generating political transformation. Dynasts who wished to control all the new wealth that came from proximity to the Empire had had to build up their military power accordingly, and were enabled to do so by the extra wealth that they came to control. But there was also an element of consent within this process, since the greater power of these leaders meant that they could offer their supporters the best hope of fending off imperial interference. All of these processes pushed political organization in the inner periphery towards the creation of larger and more powerful entities, and their aggregate effects show up in the new confederations that appeared along the entire length of Rome’s European frontiers in the third century.
Beyond this inner periphery – still viewing matters from a Roman perspective – lay an outer periphery. Here the populations generally stood in only an indirect relationship with the Empire, but their territories were a source of raw materials and other resources that were shipped into it via the inner periphery. Hence they had received some of the economic benefits that flowed into barbarian Europe from five hundred years of interaction with Rome. The outer periphery also shared in some aspects of the inner periphery’s political transformation, not least because elements of its population periodically organized themselves to take control of the greater wealth that was available closer to the frontier. This began to happen through raids in the first century AD, but shows up in more substantial transfers of militarily organized population groups from the outer to the inner periphery in the so-called Marcomannic War of the second century, and above all in the expansions of Goths and others from the outer periphery towards the Roman frontier in the third.
The boundary between the inner and outer peripheries cannot be fixed with precision, and in practice the two probably elided. As well as the periodic population intrusions from outer to inner, Roman diplomatic contacts spread to some extent beyond the innermost ring. In the fourth century we know, for instance, that from time to time the Empire had diplomatic relations with the Burgundians of the Main valley, ‘behind’ the inner Alamanni on its southern Rhine frontier. The boundary of the outer periphery, however, is reasonably easy to fix. There is no sign of any interaction – direct or indirect – with the Roman world in archaeological remains of the period very far east of the River Vistula or north of the forest steppe zone of southern Russia (Map 2). Down to the late Roman period, the populations of most of this huge territory continued to preserve the very simple Iron Age farming lifestyles that had marked out these landscapes from long before the birth of Christ.
In the broadest of terms, therefore, development in the Roman period had created four bands across the European landscape compared with the three-speed Europe that existed around the start of the millennium. Then, Roman/Celtic Europe west of the Rhine and south of the Danube was generally far more developed than Germanic Europe up to the Vistula, which, in turn, was more developed than the lands further to the east. By the fourth century, the more intense development generated by contact with the Empire had subdivided the old middle band of Germanic Europe into the inner and outer peripheries that we have just examined.34 The migratory activity associated with the fall of the Roman west not only reflected the four bands of unequal development running across the European landscape, but also transformed them out of all recognition. The profound nature of these changes shows up clearly if we take a close look at the kingdom of the Merovingian Franks north of the Alps and Pyrenees, and consider why it was left as the only possible centre of supra-regional power in the west after the destruction of Theoderic’s Empire.
As we saw in the last chapter, the Franks’ progression towards superpower status had accelerated in the reign of Clovis. Not only did he unite some previously separate Frankish warbands, but he used this power base to conquer large swathes of territory in what is now France and western Germany. And once Theoderic’s Gothic Empire fell apart, the way was open for a dramatic escalation in Frankish expansion. From the early 530s, Clovis’ sons and grandsons extended a mixture of hegemony and conquest over a much wider swathe of territory. Like Theoderic earlier, they were well aware of their achievements. When Clovis’ grandson Theudebert wrote to Justinian in about 540, he declared himself the ruler of many peoples, including Visigoths, Thuringians, Saxons and Jutes, as well as the lord of Francia, Pannonia and the northern seaboard of Italy. Theudebert went very close indeed to claiming the imperial title outright. A prerogative of Roman emperors was that they alone could issue gold coins. This had been inherited from the period when many Roman cities had had their own base-metal coinages, and continued generally to be observed after 476. Theudebert, however, started to issue his own gold coins, and many of his Merovingian successors followed suit.35
To explain the rise of Frankish imperial power just on the basis of contingent events, like Justinian’s destruction of the Ostrogothic kingdom, would be to miss much of its real significance. For all his glory, even Theoderic had had to strain every sinew to hold the rising Frankish tide in check. His alliance system of the 510s was designed to contain further Frankish expansion, and the whole edifice rested on a rickety unification of the Visigothic and Ostrogothic kingdoms, which, for all Theoderic’s ambitions, must always have been odds on to collapse after his death. And once this extraordinary but improvised Gothic power block fell apart in 526, Frankish expansion began again, well before Justinian’s assault on Ostrogothic Italy. In the early 530s, the Burgundian and Thuringian kingdoms, deprived of Gothic support, quickly fell under Merovingian sway, and all this before the first east Roman troops set foot in Italy.36 Justinian’s campaigns removed any chance that a Gothic counterbalance to the Franks might re-emerge in the western Mediterranean, but that had been looking unlikely anyway. The main point to emerge from all this is that in the post-Roman world the most likely source of supraregional power in western Europe lay not in the Mediterranean, but north of the Alps and Pyrenees.
This was an unprecedented development, which must not be taken for granted, as it easily can be if you view events just from a modern perspective. It is not that odd, given everything that’s happened since, to find France, Benelux and western Germany providing the basis of a militarily and politically powerful entity. But when this first happened in the sixth century AD, it represented a huge break with the past. The ancient world order in western Eurasia was one where the resources of the Mediterranean reached such a precocious level of development that states based upon them had always been more powerful than anything further north.37 In the sixth century, for the first time in recorded history, an imperial power emerged that was based on the exploitation of more northern European resources. This very signifcant phenomenon was a direct result of the processes of development that had been operating on the fringes of the Empire during the Roman period. The social, economic and political transformations under way on its periphery had all tended to generate ever stronger economies and political societies, and the rise of the Franks exploited this new strength to the utmost.
As first Roman control and then Gothic influence evaporated from north of the Alps, Childeric, Clovis and their descendants swallowed up both Roman territory west of the Rhine and large parts of the Empire’s old inner and outer peripheries on the other side of the river. East of the Rhine other, so-called Ripuarian, Franks were quickly incorporated into the new enterprise, as were the Alamanni, again in Clovis’ own lifetime, after the huge defeat they suffered in 505/6. In subsequent generations, a mixture of conquest and domination established different degrees of Merovingian control and hegemony over more easterly neighbours – Frisians, Saxons, Thuringians and Bavarians. Some kind of Frankish superiority was even acknowledged, it seems, in parts of Anglo-Saxon England. This new supraregional power base ran from the Atlantic in the west more or less up to the River Elbe in the east (Map 13). It combined, therefore, both a substantial portion of former Roman territory west of the Rhine, and large portions of the old Empire’s inner and outer peripheries to the east.38
The ancient world had seen Mediterranean-based supraregional powers of several shapes and sizes, and in this sense the successor states created there by longer-distance migrants such as Theoderic’s Ostrogoths were just a new variation on a long-established theme. But a brand-new phenomenon now emerged in the European landscape, and the existence of the Frankish superpower firmly reflected the five centuries of transformation effected in the north by the diplomatic, economic and other workings of the Roman state. Without the increased socioeconomic and political development that its existence stimulated in the inner and outer peripheries east of the Rhine and north of the Danube, these lands could never have provided the basis for a truly imperial power. By sponsoring the emergence of the new Frankish-dominated northerly power block, therefore, these earlier transformations set the developing pattern of European history off on a new trajectory in the second half of the first millennium. And from the sixth century onwards, the new superpower was itself to become a major factor in the further development of the European landscape.
That said, the path of Empire north of the Alps in the second half of the millennium did not run so smoothly as it had for its Mediterranean counterpart in the first. Roman emperors had come and gone, and a few peripheral territories had been lost in the third century. The Empire’s modes of government, likewise, were transformed over time. But this was largely a process of organic, internal evolution, at least up to the third century. Essentially, the Roman Empire remained the same state ruling over broadly the same territories for the best part of five hundred years.39 The same was not remotely true of imperial western Europe in the second half of the millennium. Merovingian greatness peaked in the sixth century, and by the second half of the seventh, much of the real power had fallen into the hands of regionally dominant blocks of local landowning elites, in Neustria, Austrasia and Burgundy (Map 13). This in turn allowed peripheral areas to reassert their independence. The Thuringians seem to have been independent from the revolt of Radulf in 639, the Saxons and Bavarians soon after 650. Even the long-subdued Alamanni reasserted their independence by the early eighth century.40
Later Merovingian political fragmentation was followed by a dramatic resurgence of Empire in the same century at the hands of a second Frankish dynasty, the Carolingians. The new dynasty originally came to prominence as Merovingian loyalists with lands between Cologne and Metz, more or less modern Belgium, where the Merovingians too had first burst on to the historical stage. Those who study the first millennium AD have a distinct advantage when it comes to the age-old game of Name Five Famous Belgians. There is no need to tell the Carolingian story in any detail here, but in the late seventh century the first really prominent member of the dynasty, Pippin, made himself dominant in northern Francia, over both Austrasia and Neustria, after the battle of Tertry in 687. Ruling at first through a Merovingian frontman, in the next generation Pippin’s son Charles Martel successfully reunited to this northern Frankish heartland all the old territories of Gaul controlled by the Merovingians at their height. By 733 he was moving his supporters into Burgundy to establish his control in the south-east. After a lengthy struggle, Aquitaine in the south-west was conquered in 735. Charles Martel also campaigned east of the Rhine, forcing the Frisians, and notably the Saxons in 738, to start paying him tribute again.41
Imperial momentum had been established, and his sons and grandsons did more than maintain it. First, they ditched the Merovingians. After securing his own position, Charles’s son, another Pippin, made the final leap to royal lightspeed, deposing the last Merovingian, Childeric III, and having himself crowned king in 752. Now royal, the Carolingians quickly expanded the area under their control, the second half of the century being consumed in an orgy of conquest, initially under Pippin, but more particularly under his son and heir Charles, better known as Charlemagne: ‘Charles the Great’ (Karolus Magnus, 768–814). East of the Rhine, direct Frankish rule was asserted for the first time over peoples who had been periodically subordinate to the Merovingians, but often autonomous and sometimes entirely independent. By 780, self-assertive ducal lines in Alamannia, Thuringia and Bavaria had all been extinguished, and further north, Frisia had been subdued. Saxony, too, was eventually conquered outright, for the first time, but only in the early ninth century after two decades of punishing campaigning that had included forced baptisms, population transfers and wholesale massacres. On the back of these successes, Charlemagne directed his attentions further afield. The independent kingdom of the Lombards was crushed in the mid-770s, and in campaigns that culminated in 796 he destroyed the central European Empire of the Avars. The plunder taken on this expedition became proverbial.42
Charlemagne’s conquests thus united Gaul, the territory between the Rhine and the Elbe, northern Italy and much of the Middle Danube region, together with parts of northern Spain, into one vast imperial state (Map 13). The Merovingians had at times exercised influence over large parts of this territory, but not everywhere in the form of direct rule, and their domination had never extended into Italy or the Middle Danube. Consonant with this, Charlemagne was much less shy in asserting that he had created an empire. From about 790 onwards, a consistent thread appeared in the writings of his team of resident royal intellectuals, extolling his success and his piety and declaring that both showed him to be a (or even ‘the’) true Christian emperor. There is not the slightest doubt, therefore, that Charlemagne’s imperial coronation on Christmas Day 800 in St Peter’s basilica in Rome happened not by accident but by design. Three hundred and twenty-four years after the deposition of Romulus Augustulus, full-blown Empire in the west was reborn.43
For all its grandeur and the lasting importance of some of its cultural legacies, however, the Carolingian Empire proved no more stable than its Merovingian predecessor. By the later ninth century there were still Carolingian kings, but their effective power was confined to a limited block of territory around Paris. Elsewhere in west Francia, authority had devolved once again to a constellation of local princes who each exercised within his own domain the kinds of powers (over justice, the minting of coins, ecclesiastical appointments and so forth) that Charlemagne had previously wielded over the entire Empire. Sometimes these rights had been formally granted, sometimes merely usurped. As one contemporary chronicler, Regino of Prüm, put it in a charming phrase, after the death of Charlemagne’s grandson Charles the Bald in 871, each area had made a prince ‘out of its own bowels’. In the west, the Carolingian Empire had come and gone within a century of Charlemagne’s coronation.44
In east Francia, beyond the Rhine, greater unity prevailed, providing – eventually – the third Frankish imperial moment of the second half of the millennium. Here, another of Charlemagne’s grandsons, Louis the German, enjoyed an unusually long reign, providing a greater heritage of political cohesion for the constituent duchies of east Francia – originally Saxony, Thuringia, Franconia, Suabia and Bavaria, to which Louis added Lotharingia and Alsace (Map 14). East Francia’s cohesive tendency survived the extinction of Louis’s line at the turn of the tenth century, and the region was brought to still tighter unity first by Conrad (King of east Francia during 911–18), originally Duke of Franconia; and then by Henry, son of Otto son of Liudolf, the Duke of Saxony from 912, then King of the east Franks from 919 until his death in 936. Within three years of his accession, Henry had forced the Suabians and Bavarians to recognize him as King of the east Franks. He then provided effective war leadership against the pagan Magyar nomads, who had moved into the Middle Danube region from the western steppe around the year 900. They rapidly made themselves public enemy number one, raiding far and wide across northern Italy and southern France and defeating no fewer than three major east Frankish armies between 907 and 910. On the basis of a carefully crafted programme of military reforms in the late 920s, Henry was able finally to defeat them at the battle of Riade (in northern Thuringia) in 933. This victory secured Henry’s position as king, but it was his son Otto I who took the dynasty’s authority to a new level.
Not that this was easy. It took Otto until 950 finally to subdue different combinations of rebellious dukes and familial rivals. He also effectively continued another of his father’s key policies: expansion in the east. Otto then launched a powerful expedition into Italy in 951, taking control of most of its northern and central regions. This demonstrated that he was now the most powerful ruler of Latin Christendom, a status he confirmed by inflicting a crushing defeat on the still pagan Magyars at the battle of the Lech in 955. This gave him an irresistible combination of overt power and ideological legitimacy, since it was clearly God who had allowed him to defeat the pagans where so many other Christian rulers had failed. So armed, Otto was able to browbeat the papacy into another imperial coronation. A second Italian expedition was mounted in 961, after which he was crowned Emperor in 962. The third Empire of the second half of the first millennium had come into being. Although based ultimately on Otto’s inherited position in Saxony, the Ottonian Empire was a distinctly sub-Carolingian entity, a reasonably direct descendant of the east Frankish kingdom of the ninth century.45
A succession of Frankish empires based firmly in northern Europe thus dominated large parts of western and west-central Europe between about 500 and 1000. They were never as stable as their Roman predecessor, the exercise of imperial power being punctuated by two periods of considerable political chaos, c.650–720 and c.850–920. This was because all three were based on a weaker type of state structure. In the Roman Empire, much day-to-day control was in the hands of local communities, but the the central authorities had always retained some key levers of power. They had the right systematically to tax the largest sector of the economy – agriculture – in order to generate substantial annual revenues. These were used to support a large professional army, a governmental machine and state legal structures (both laws themselves and the courts), which were the font of legitimacy in the Roman world. For all its limitations, and there were many, the Roman Empire thus operated a relatively big state structure in pre-modern terms. The three Frankish Empires of the second half of the millennium differed considerably in detail from one another, but none taxed agricultural production systematically to maintain large professional armies. They all drew the bulk of their armed might from the militarized landowners in the localities under their control. Sometimes this support could just be extorted, but usually it had to be attracted via reward. And since these later rulers were not renewing their revenues annually through large-scale taxation, this meant that wealth tended to flow outwards from the imperial centre to the more local landowning elites.
As has been well observed, the three Frankish imperial moments of the later first millennium all occurred when circumstances favoured expansionary, predatory warfare. Profits from this activity allowed the imperial dynasts, whether Merovingians, Carolingians or Ottonians, to reward their militarized landowning supporters without having to impoverish themselves. But when expansion stopped, political fragmentation quickly returned, as rewards flowed outwards again from a now fixed body of resources.46 As we shall see, this particular aspect of later first-millennium imperialism would play an important role in the further transformation of barbarian Europe, as well as providing much of the explanation for the rather stop–start character of Frankish imperialism. Even so, for most of the second half of the period, the view from outside would have identified a predominant western European power whose influence encompassed large parts of the continent. And it is, of course, precisely the view from outside – that of the barbarians in the rest of Europe – that will concern us in the chapters that follow. Before we can properly examine how the rest of European society responded to the stimulus provided by this entirely new north European imperial power, and the patterns of expansion inherent to it, we must take account of two further major reconfigurations of the ancient world order.
The Strange Death of Germanic Europe
The first unfolded more or less simultaneously with the rise of the first of the Frankish imperial dynasties: the Merovingians. Their Empire, as we have just seen, stretched from the Atlantic to the Elbe, and comparing this area with prevailing patterns of development across the European landscape as they stood in the sixth century, it quickly becomes apparent that its extent east of the River Rhine coincided closely with those parts of the old Roman periphery – inner and outer – that had maintained a considerable continuity in their long-standing Germanic-type material cultures and associated levels of socio-political organization during the period of Roman collapse. This is a point of critical importance that is easily lost because it concerns areas of Europe whose history finds little or no coverage in the surviving historical sources. Its importance emerges immediately, even from a quick overview of the archaeological evidence.
In the late Roman era, the largely Germanic-dominated inner and outer peripheries of the Roman Empire comprised huge swathes of territory running broadly north-west to south-east across the map of Europe. Its breadth in the north was approximately 1,000 kilometres, from the east bank of the Rhine to just beyond the Vistula. In the south, it was broader – more like 1,300 kilometres from the Iron Gates of the Danube to the west bank of the River Don (Map 15). Societies within this block of territory had relatively dense and increasing populations, relatively developed agricultures, relations of some kind with the Roman Empire, and material cultures that characteristically included substantial amounts of carefully crafted metalwork and pottery. By the sixth century, culture collapse had engulfed most of the area. In Ukraine and southern Poland, this occurred when the Cernjachov and southern Przeworsk systems disintegrated, not long after 400 AD. In middle Poland, collapse can be dated to c.500, and in Pomerania by the Baltic to c.500–25. In the Elbe–Saale region, complete collapse came right at the end of the sixth century; between the Elbe and the Oder, there is no sign of any Germanic continuity into the seventh. To the south of this zone, in Bohemia and Moravia, a thinning-out of Germanic-type remains is once more observable in the fifth and sixth centuries, followed by the total disappearance of such material between the mid-sixth and early seventh. By c.700, characteristic styles of traditional Germanic material culture were thus confined entirely to areas west of the Elbe (Map 15).47
The fact that Merovingian Frankish expansion did not extend into any of the areas affected by Germanic culture collapse was not an accident. Like its Roman counterpart, Frankish expansion was accomplished by military annexation, whose potential benefits had always to be weighed against its many costs. Battles had to be fought, and these were many and fierce even if the historical evidence is not good enough for us to reconstruct them in detail. Sometimes, though, you get lucky. The nature of the Frankish takeover of the Alamannic kingdom, for instance, shows up beautifully in the evidence of widespread and dramatic destruction from the old hillfort sites, which, as we saw in Chapter 2, had emerged in the late Roman period as the centres from which the authority of kings was exercised. About the year 500, when historical sources tell us Clovis won his great victory, they – or all that have been investigated – were taken by storm, and, more generally, huge material-cultural discontinuities show up right across Alamannia. Not only were the hillforts abandoned, but new burial rites appear in the cemeteries, and in some places entirely new cemeteries came into use. The degree of investment of human and other resources required for such an aggressive takeover would only be made when its rewards were going to be commensurately large.48
The collapse of long-standing patterns of central European, largely Germanic material culture in the fifth and sixth centuries meant that, east of the Elbe, there were no similarly coherent political structures to confront, and no relatively developed economies with accumulations of movable wealth to ransack. In the centuries either side of the birth of Christ, the Roman Empire had expanded to the limit of that era’s profitable warfare, and in the sixth century the Merovingians did the same. The one area of old Germanic Europe that maintained the old cultural patterns and didn’t fall under Frankish domination was southern Scandinavia: the Jutland peninsula, the south-western Baltic islands and the southern coast of what are now Norway and Sweden. But Merovingian power was exercised in neighbouring Saxony only in the form of hegemony rather than outright conquest, and this probably insulated Scandinivia from any wider Frankish ambition. This one partial exception doesn’t negate the general point, though. Only those areas of Rome’s inner and outer periphery where continuity of development had been maintained were worth the effort of Frankish conquest. In this sense, the new trajectories of development from the late Roman period played an important role in defining the limits of the new supraregional power of the post-Roman world.49 So far so good. But what exactly had happened in those other areas of Germania that saw such a dramatic disruption to well-established material cultural patterns?
Thinking about this phenomenon, it is important to be absolutely clear about its nature. As the Polish archaeologist Kazimierz Godlowski above all demonstrated, culture collapse involved the disappearance, in the fifth and sixth centuries, of long-standing patterns of material-cultural development over vast tracts of central Europe. These patterns often ran back at least to the start of the millennium and sometimes beyond. But when discontinuity hit, it manifested itself in virtually every area of life reflected in the material-cultural remains: everything from the enduring economic links with the Mediterranean world that generated regular flows of Roman imports, to established craft traditions in pottery and metalwork. Technologically, pottery production simplified dramatically, the use of the wheel was abandoned. This was matched by a marked diminution in the range of pot forms and even in the overall quantities being produced. Metallurgical production similarly declined in scale – the range of ornaments being produced (or at least deposited) shrank almost to zero. Settlements also became much smaller.50 Essentially, the archaeological record shows striking simplification in every category by which the activities of the populations of the region are customarily analysed, compared and dated in the Roman period, and it all adds up to a massive change in lifestyles.
What human history underlay these striking archaeological discontinuities?
According to the interpretation championed by Godlowski, traditional cultural patterns vanished because the population producing them had itself largely disappeared. Where we have relevant literary sources, material-cultural collapse is geographically and chronologically coincident with the known movements of Germanic-speakers on to Roman soil. The Cernjachov and Przeworsk systems collapsed at the same time as Goths, Vandals and their other constituent populations were being displaced by the rise of Hunnic power in central Europe (Chapter 5), while the thinning-out of Germanic material culture along the Elbe in the fifth century has long been associated with the transfer of Angles, Saxons and others to Britain, and the southern movement of Lombard groups into the Middle Danubian region. These flows both continued into the sixth century, as we have seen, not least in response to the extension of Frankish imperial power eastwards, which led a large number of Saxons to join the Lombards in their trek into Italy.51
The chronological links are much too tight to be accidental, but the total departure of Germanic populations is not the only possible, nor even the most likely, explanation of this extraordinary phenomenon. Since archaeological cultures must be understood as systems, the disappearance of established cultural forms can a priori have a number of causes. In this case, as other commentators since Godlowski have stressed, what we are dealing with is the disappearance of ornamental metalwork, weaponry and specialized wheel-made pottery, and these were largely produced for a Germanic social elite. The absence of such items from the observable spectrum of archaeological remains could reflect, therefore, the disappearance from these lands only of the political and militarized class for whom they were manufactured. A numerous – possibly very numerous – but archaeologically invisible peasantry, users of a much simpler material culture, might have been left behind.52 In theory, therefore, it is possible to explain culture collapse by positing anything from a total evacuation of the landscape at one extreme of the spectrum, to what you might term elite departure at the other. Where within this range does the evidence suggest the human history behind Germanic culture collapse fell?
We will need to return to some of the evidence in more detail in the next chapter, when we look at the Slavic populations who eventually took control of these de-Germanized areas of central and eastern Europe. For the moment, a few more general observations can be made. First, Germanic culture collapse surely does not reflect a total evacuation of the affected landscapes. As we have seen in the case of the Goths north of the Black Sea, there is good reason to suppose that many groups among the indigenous population, who had become subordinated to Gothic intruders in the third century, did not form part of the later Gothic migration units that moved on into the Roman Empire from 376. Nor, again in general terms, do the numbers of Germanic migrants moving into the Empire in the late Roman period seem anything like large enough to have created large empty landscapes.
It is obviously not possible to say exactly how many people were caught up in the migratory activity generated by the rise and fall of the Hunnic Empire and the new opportunities for expansion that then became open to Rome’s nearer neighbours as it lost the capacity to maintain frontier security. One negative thought experiment, however, is worth running with. This involves considering how many migrants are known to have emerged from the areas that suffered culture collapse. There are reasonable indications, for instance, that both Visigoths and Ostrogoths could field around, or a few more than, twenty thousand fighting men. The armies of the Vandals, Alans and Sueves were between them probably just as large, certainly in 406 before they suffered such heavy losses in Spain, while Burgundian manpower, if probably smaller, was not minimal. We have little conception of how many Middle Danubian refugees were recruited into the army of Italy or east Rome’s Balkans military establishment; but to judge by the numbers given for the Heruli, the many different groups we hear of will between them have amounted to at least another 10,000-plus warriors, and quite possibly double that. Migrant Anglo-Saxon numbers are perhaps the most controversial of all, with guesses ranging from 20,000 to 200,000.53
If for the moment we take a maximum view of this evidence – for reasons that will become apparent – it would suggest that the largest possible figure you could reasonably calculate for the number of Germanic warriors who departed from the zones suffering culture collapse was something over 100,000 men, but certainly not 200,000. There is much guesswork here, but it is not a vastly inflated figure, and this order of military magnitude really is required to explain how the immigrants were able, between them, to bring down a west Roman state that determinedly resisted their intrusions. I suspect, anyway, that 100,000 is not making sufficient allowance for how many immigrant warriors died in the course of the action. Nonetheless, something over one hundred thousand does give us a ballpark figure to work with. How many people in total were on the move depends on how consistently women and children accompanied these warriors, and on the very murky subject of how many slaves came along for the ride. Here again, let’s take a maximum view – and in any case, despite some recent attempts to deny it, there is both a decent amount of evidence that most of the larger groups were mixed in age and gender, and also, further reason to accept that this was so. As we have seen, traditional accounts multiplied the numbers of fighting men by five to get total population figures for mixed groups, but something nearer to four may be more correct. On the other hand, none of this makes any allowance for slaves. Putting all this together, a reasonable maximum estimate might put the total exodus from the areas which suffered culture collapse at something around or perhaps a bit over half a million souls.54
The reason for bothering with such calculations is that we do know the size of the territory affected. Germanic culture collapse affected an area defined broadly by the Rivers Elbe and Vistula in the north and the Iron Gates and the Lower Don in the south. At a rough calculation, this weighs in at close to a million square kilometres. For the migrations of the late Roman period to have emptied this area, population densities across it would have to have been in the region of 0.5 per square kilometre. This is an impossibly low figure. Even allowing for the fact that agricultural regimes were not intensive, it is simply impossible for the departure of half a million people to have emptied such a huge area. The figures are only guesstimates, but one recent study has suggested (and reasonably so) that just what it calls the Pontic-Danubian region (Map 15) must have contained between three and four million people in antiquity, and the population of just the Great Hungarian Plain has been put at something like three hundred thousand in the early medieval period. For all that every number cited in the last two paragraphs is an approximation, we can nonetheless safely discount the possibility that culture collapse in central and south-eastern Europe was caused by the complete evacuation of its population.55
In general terms, then, Germanic culture collapse was caused by the disappearance only of particular elite groups from the affected areas. But this conclusion needs to be tempered with two further observations. First, for all the transformations of the preceding centuries, Germanic society of the fourth century was not dominated by a very small elite. New distributions of social power did emerge between the first and fourth centuries, but the elite of the Germanic world still represented a larger percentage of the population than the tiny landowning class, say, that had dominated in the Roman world. As we saw in Chapter 2, and as the events of the so-called Völkerwanderung confirm, we must think in terms of social and political power (and group identities) shared between fairly broad oligarchies of freemen, numbering between a fifth and a third of the warrior population. Nor was participation in the migrations, at least among the larger groups like the Goths and Lombards, limited just to this dominant oligarchy. At least two social strata of warriors – possibly to be equated with the free and freed classes documented in early medieval law codes – are observable in these intrusive groups, not just a single body of elite soldiery, and sometimes they brought slaves with them as well, not to mention families.56 Elite departure was thus not a very small-scale phenomenon.
Second, as we shall see in the next chapter, the evidence indicates that population levels did nosedive dramatically in some particular localities. Once again, this suggests that Germanic migration may not have been entirely negligible in demographic terms, and the two points may well be linked. Because the Germanic elites were not so tiny in the first place, and had some dependent social groups (slaves and freedmen) attached to them, then when a concentrated group of migrants left a particular area, this may well have created empty districts.57
Not only did prevailing patterns of development dictate the working-out of the migratory processes of the late fourth and the fifth century, therefore, but the reverse was also true – the migrations affected patterns of development. One major consequence of this interaction, as we have seen, was the emergence of an unprecedented type of imperial power for western Eurasia, based on north European resources. Because, however, the Roman Empire came to an end in a process that saw substantial armed and organized groups from the periphery relocate themselves in the heart of its former territories, the process of imperial collapse was matched by parallel transformations in large parts of this periphery. Culture collapse caused by the departure of the still fairly broadly based elite of Germanic Europe changed totally the socioeconomic and hence political organization in the old periphery of the Empire, and marks a second major break with the ancient world order – a break quite as important as the rise of the Franks’ northern European Empire. It was to have enormous consequences for the emergence of Slavic Europe, as we shall see in Chapter 8, but this process was also profoundly shaped by the third major reconfiguration of the old world order that unfolded in these middle centuries of the first millennium.
Out of Arabia
Up to about 600 AD the eastern half of the Roman Empire, with its capital at Constantinople, maintained its imperial credentials as the dominant power of the Mediterranean. Strong though his position was in the 510s, Theoderic the Ostrogoth had held back from making his claim to imperial power absolutely explicit, for fear of alienating the rulers of Constantinople. And in the next generation, the astuteness of the king’s judgement showed through, when Justinian’s forces, in twenty years of brutal warfare from 536, played more than just a walk-on role in the emergence of imperial power north of the Alps by destroying the Ostrogothic Italian kingdom. This military adventure followed an astonishingly successful earlier conquest of the Vandal North African kingdom in 532–4. Then in the early 550s, in Justinian’s later years the east Romans established a toehold in southern Spain. Constantinople’s domination of the Mediterranean had moved from latent to manifest within the space of about twenty years.
East Rome’s collapse in the seventh century from these heights of imperial grandeur was every bit as dramatic as that of its western counterpart in the fifth. In the early 610s, it looked as though it was about to be conquered by its traditional bête noire, Sasanian Persia, which took control of many of its key revenue-producing districts: Syria, Palestine and Egypt. By 626, a Persian army was even camped on the south side of the Bosporus, while its nomadic Avar allies besieged Constantinople, just over the water. Astonishingly, the Empire clawed its way back from the jaws of this defeat. Constantinople survived the siege, and the Emperor Heraclius mounted a series of campaigns through Armenia into Mesopotamia which, by autumn 628, had brought Persia to the brink of collapse. The Sasanian King Khusro II, who had launched the war of conquest, was deposed, and most of the conquered territories were restored to Heraclius’ rule.
No sooner was the ink beginning to dry on the history of Heraclius’ great victory, however (provisional working title: The Original Comeback Kid), than it had to be deposited in the nearest waste-papyrus bin. Out of a long-neglected corner of the Near East burst a new enemy – Arab tribes united only within the last decade by Islam and Muhammad – sweeping all before them. Heraclius’ triumph turned to dust in his mouth as, before the end of his reign, Syria, Palestine and Egypt were all lost once more, and Asia Minor turned into a battle-ravaged wasteland. By 652, other Arab armies had conquered the entire Persian Empire, and within a further two generations the new Empire of Islam stretched from India to the Atlantic.58
The details of this astonishing revolution in world history are not central to this study. Suffice it to say – and this will come as no surprise – that nearly as many reasons have been offered for east Roman imperial collapse as for that of its western counterpart. Traditional lines of explanation have often centred on Justinian’s conquests in the western Mediterranean, arguing that they were overambitious and left his successors a poisoned chalice of bankruptcy and imperial overstretch. But if ‘a week is a long time in politics’, as one British prime minister famously commented, this link looks hard to sustain. Justinian died in the mid-560s, the Arab conquest came seventy years – or pretty much three whole generations – later. The events could still be interrelated in some way, of course, but they don’t look like simple cause and effect. More recently, those concentrating on internal reasons for Constantinople’s collapse have switched their attention to alternatives: the periodic sequence of plagues that afflicted the Mediterranean world from 540 onwards, and – perhaps related – signs of possible later sixth-century economic decline in the Roman Near East.
These explanations all have something to say, but outside factors also need to be taken into account: not least, the all-in knock-down twenty-five-year war between Constantinople and the Persians that immediately preceded the Arab conquests. Persia and eastern Rome fought one another on and off throughout the sixth century, but for the most part only in limited fashion: through surrogates in Caucasia, or by sieges designed to capture the odd strategic fortress. This restricted pattern of warfare fizzled out in the early seventh century, when the two powers fought each other head on, and ultimately to a standstill. There was a triumphant fightback by Heraclius when all seemed lost, but the terms of the 628 peace treaty show that the end result was actually a draw, through exhaustion. Despite Heraclius’ victories, Constantinople failed to get back every piece of territory lost since 602. This, of course, immediately provides part of the explanation for the Arab victories over both empires that quickly followed.59
But attention also needs to be paid to the Arab world itself. Here, the galvanizing effect of Muhammad’s new religion, creating unity within a previously fragmented population, ranks centre-stage. But, as with the appearance of new confederations capable of forming successor states out of the western Empire’s periphery in the late fourth and the fifth century, there is a backstory here of huge importance. Looked at in the round, the evidence demonstrates a steady growth in the size and power of Arab client states on the fringes of the Roman and Persian Empires between the fourth and sixth centuries, just as there had been in those of the western Empire’s European peripheries between the first and the fourth.60 What concern us here, however, are the broader effects of this seventh-century revolution on European-wide patterns of power. Two stand out.
First, the rise of Islam destroyed east Rome as a truly imperial, supraregional power. If you read texts produced in Constantinople after the deluge, this is not immediately obvious, and the city itself was not to fall to a Muslim power until Mehmet the Conqueror’s cannon finally blasted a hole through the city’s great Theodosian landwalls in 1453, near the modern Topkapi bus station. For most of the preceding seven hundred years, the rulers of the city had called themselves ‘Romans’ (even while writing in Greek), and maintained all the old Roman ideologies of supremacy: claiming to be god-appointed emperors, whose job it was to bring proper order to the entire human cosmos.
As in so many contexts, though, it is important to look beneath the surface. Then, what really strikes you about Constantinople after the mid-seventh century is how much state power had haemorrhaged away. Islamic conquest deprived Constantinople of many of its richest provinces: Syria, Palestine and Egypt in the first generation, quickly followed by North Africa about forty years later, and eventually Sicily as well. Asia Minor was retained, but became a major battlefield in further conflicts with the new Islamic state, and the archaeological evidence shows how badly its economy was affected. All the great cities of antiquity, where they survived at all (and some didn’t), ceased to be major centres of population, manufacture and exchange, being transformed into military fortresses and command posts. Coinage, likewise, became exceedingly scarce, and everything points to a massive simplification of the economy. Before these disasters, the east Roman Empire was quite similar in ‘shape’ to the Ottoman Empire of the sixteenth century, from which interesting tax records survive. These can be used to gloss the likely extent of Constantinople’s losses in the earlier period in terms of state revenue (although the overwhelming nature of the disaster is anyway clear). And if you do the calculations and make some appropriate adjustments, it becomes apparent that the rise of Islam deprived Constantinople of between two-thirds and three-quarters of its revenues; that is, of between two-thirds and three-quarters of its capacity to act.61
The consequences of this diminution show up with great clarity in the big picture of European history after 600 AD. From the early seventh century, Constantinople was no longer a pan-Mediterranean power and major player on the broader European stage. Though still important in the eastern Mediterranean, it became in many ways an unwilling satellite state of the Islamic world, no longer substantially in charge of its own fate. Its subsequent periods of prosperity and decline correlate closely and inversely with the history of the new Islamic power block. When Islam was politically united, Constantinople was condemned to decline; when – as sometimes happened – Islam itself fragmented, there was room for modest expansion. In short, the self-proclaimed imperial Romanness of the rulers of post-seventh-century Constantinople is a chimera. The losses suffered at the hands of Islam meant that these emperors were now ruling what was as much a successor state to the Roman Empire as any of the new powers of the Roman west a century earlier. My own preference, in fact, is to use ‘Byzantine’ rather than ‘east Roman’ from the mid-seventh century, as a reflection of how great a sea change the rising tide of Islam had created in Mediterranean history.62
Second, the reverse of the same coin, Islamic explosion created a new superpower on the south-eastern fringes of Europe. It engulfed not only much of the east Roman Empire, and certainly its richest territories, but its old Sasanian sparring partner too. The result, when some of the dust had settled by the early eighth century, was a gigantic Empire running all the way from Spain to northern India. Ruling such an enormous entity using pre-modern communications was always a logistic nightmare, in addition to which there were major ideological divisions over how the Islamic Empire should be run, and by whom. Not surprisingly, therefore, its internal history was rarely stable. Even if their political control was always a bit arthritic, though, and certainly declined with distance from their respective capitals, both the Umayyad Caliphate centred on Damascus between the 660s and the mid-eighth century, and the Abbasid Caliphate centred on Baghdad from the later eighth to the early tenth, represented huge concentrations of imperialwealth and power, on a scale that surpassed even that of the Roman Empire at its height.63 This superpower based in the Near Eastern fringes of the European landmass was too far away to intervene directly in the unfolding history of migration and development in barbarian Europe, but its indirect effects on these processes were enormous. Not only did it remove the east Roman Empire from the map of major players in European history, but, as we shall see in the chapters that follow, its diplomatic and economic tentacles stretched up through the Caucasus on to the western steppe, and from there beyond, into eastern and even northern Europe.