Empires and Barbarians

by Peter Heather

  Sparse and difficult though it is, then, the evidence strongly suggests that continental migration to Britain in the fifth and sixth centuries took the form of a flow, like that of Germanic-speakers south to the Black Sea in the third century or of Lombards towards the Middle Danube in the fourth and fifth (Chapters 3 and 5), or indeed of the Vikings to the west in the ninth, rather than a single concentrated pulse as was the case, for instance, in 376. Its minimum duration would appear to have been c.410 to 575, although even this might be a substantial underestimate. The movement was probably also not continuous, ebbing and flowing with the lows and highs of the struggle it generated with elements, at least, of the indigenous Romano-British population. Unless Gildas is substantially misrepresenting the career of Aurelius Ambrosius – and there is no reason to think he was, since to do so would have undermined the case he was trying to make to a British political audience who knew these events for themselves – immigration into Britain must have become considerably less attractive after the native victory at Badon Hill. Interestingly, both Gregory of Tours and Procopius note the presence of Germanic-speakers from north of the Channel among the continental Franks in the first half of the sixth century, suggesting that this period, which coincides with the aftermath of Badon Hill in most chronologies, even saw some reverse migration.27 Furthermore, the flow clearly recruited from a wide geographical area, to judge both by Bede’s historical account of the migrants’ origins and the geographically diverse origins of the material culture that spread among them.

  None of these sources gives you any sense of the overall numbers of immigrants involved in this highly extended flow. Many of the individual migrant groups, especially at the beginning, may well have been small. According to Gildas, the initial force of mercenaries came in just three boats, and could therefore have numbered little more than a hundred men. Three ships is possibly something of a folklore motif, however, and not all the groups need have been that small.28 On the continent, groups of Saxons up to twenty thousand in number – explicitly including women and children – took to the road in the fifth and sixth centuries, and some larger groups of this type may possibly have come to Britain. The large cremation cemeteries of eastern England, for instance, look like the remains of more unified migrant groups than the smaller inhumation cemeteries of southern England, though they are certainly not the burial sites of groups twenty thousand strong. It is likely enough, too, that the migration flow would have had to respond to the changing nature of British resistance. It is the central drift of Gildas’ narrative summary, and the basis of his unflattering comparison with the current state of affairs, that Aurelius Ambrosius had pulled substantial numbers of the native British together into a reasonably united response to the Saxons, but that this strength was now being dissipated by competition amongst his lesser successors. The immigrants, of course, would necessarily have had to respond in kind to Ambrosius’ success, fielding more substantial forces of their own to resist the greater level of British resistance that he organized. Even if the immigrants started out in small parties, therefore, concerted British counterattack would have forced them to reform into larger units.

  There is no narrative evidence to support this vision of ebb and flow, but its reality might well be reflected in, and is certainly compatible with, the seemingly late arrival into England of the ruling dynasties of the Anglo-Saxon kingdoms of c.600 and beyond. These were perhaps the leaders who provided the greater unity necessary to turn the tide of war once more in an Anglo-Saxon direction. If accepted in its outlines, such a picture of the migration flow would have many parallels. The evolution of migration flows into new forms to overcome obstacles, or to allow a range of greater ambitions to be fulfilled is, as we have seen, a common thread from the third-century Goths to the ninth-century Vikings, to the nineteenth-century Boers. It is also, of course, an underlying constant in work on group identities, that they form and strengthen in the face of conflict (Chapter 1). Nonetheless, it is important not to overstate the military problem that even a more united Romano-British world presented. The Saxons never faced an opponent with anything like the military power of the western imperial state encountered by continental migrants on to Roman soil. Hence it is not surprising that the end result of the Anglo-Saxon takeover was a multiplicity of smaller kingdoms, certainly at least ten and perhaps many more, in c.600. The new kingdoms of Anglo-Saxon England encompassed military forces which added up to far more than a few boatloads, but the lower degree of danger faced on British soil was insufficient to make them go through the kind of process of political unification that proved necessary to the Visigoths, Vandals, Franks or Ostrogoths, all of whom were operating in contexts requiring military forces in the tens of thousands.29

  It is also certain that the migration flow included women and children. The first mercenary groups presumably were all-male, but female dress items of continental Germanic origin (again, especially brooches) provide much of the archaeological evidence unearthed from the cemeteries. Some of this material could have arrived without women attached, destined for Saxon invaders’ Romano-British brides; but to suggest that there were no women at all looks rather forced, not least since migrant groups of Saxons on the continent certainly took women and children with them. There are two possible reasons why Saxon groups migrating to Britain may have been skewed to the male gender. The first is again the smaller scale of some of the action. Before the era of Aurelius Ambrosius, military retinues numbering in the few hundreds might have been sufficient to carve out niches within Britain for intrusive Saxon leaders. If so, these leaders would have had less or no need to recruit followers more broadly within their home societies, and hence there was less likelihood of men with families becoming involved. The second was logistics. The large-scale land-based migrations of this period – all those documented, at least – moved baggage and the physically less able in huge wagon trains, thousands of vehicles strong. This must have been cumbersome enough, but shifting families, animals and baggage across the Channel or North Sea to Britain represented an entirely different order of logistic difficulty. Shipping non-combatants would not only have meant finding more ships, but would also have incurred many other costs.

  Nonetheless, comparative evidence suggests that we should not overstate the likely effect of either factor. There is unimpeachable DNA evidence that the ninth- and tenth-century Norse took Scandinavian women with them to Iceland in substantial numbers, as well as other females they had picked up in the meantime from Scotland, Ireland and the northern and western isles. About one-third of modern Icelandic female DNA is derived from Norwegian ancestry. Even if the immediate forebears of these ancestors had moved over the North Sea first, so that the Norwegian babes journeying to Iceland came only from the islands or mainland of northern Britain, this would stress only that the initial Norse settlements in Britain had involved Norse women in large numbers. And all of these ninth- and tenth-century sea crossings – from Scandinavia to the northern British Isles and from there on to Iceland – were longer, more difficult and more costly than those of the fifth and sixth centuries that brought Anglo-Saxons to southern Britain. It is also the case that a substantial female presence is required to explain the degree of language change that followed in southern Britain. We will return to this issue in more detail in a moment, but the dominance of the immigrants’ Germanic language, which was essentially untouched by native British Celtic tongues, could never have occurred, in the absence of Saxon grammar schools, if Germanic mothers had not been teaching it to their children.30

  If the overall scale of Anglo-Saxon migration into lowland Britain in the fifth and sixth centuries is highly debatable, we can make some headway with its nature: namely, an extended flow of population over time which included women and children. Thanks to the dearth of historical sources, we are again short of explicit information about its causes, but it is a pretty safe bet that an overriding motivation among the migrants was the wealth of the relatively developed Romano-British ag
ricultural economy. Gildas’ account indicates as much. In his view, it was the prospect of good pay that brought the original Saxon mercenaries to England, and their subsequent revolt focused on ransacking the island for everything they could find once they couldn’t extract any more cash.31 Having made themselves supreme, they then took control of the landscape – the primary means of wealth production in this fundamentally agricultural world – to ensure their prosperity in the longer term. By 400 AD, the Romano-British economy may or may not have been past its mid-fourth century peak, but either way it was still much more developed than the rural world inhabited by the Anglo-Saxons on the other side of the North Sea. And in fact we have unimpeachable evidence that this greater wealth had long held an attraction for Germanic populations from less developed landscapes over the water.

  Saxon pirates had been finding their way across the North Sea to lowland Britain from at least the mid-third century. And although narrative accounts to this effect are lacking – one major sea-borne Saxon raid on northern Gaul is reported in some detail by Ammianus, but none on Britain – we have impressive indirect evidence that Saxon sea raiders had remained a threat to Romano-British landowners throughout the fourth century. From the late third, the central Roman authorities operated a unified military command which encompassed both sides of the Channel and the eastern shoreline of Britain. Its commander disposed of naval flotillas and garrisons, and a string of powerful fortifications some of which survived the Saxon takeover. The massive fortifications of Portchester (just outside modern Portsmouth) were formidable enough, indeed, to fulfil many functions right through the medieval period and down to the Napoleonic War, when it served as a prison camp for French sailors. This whole collection of men and materiel was designated the litus Saxonum – ‘the Saxon Shore’ – leaving no doubt as to the threat it was designed to counter (Plate 15). That the Romans should have bothered to maintain a military investment on this scale suggests that Saxon sea raids, if usually small, were nevertheless an endemic problem.32

  In the drawn-out migration flow of the fifth and sixth centuries, there was plenty of time for subsidiary motives to come into play as well. Sea levels were rising in the North Sea, to the extent that some continental communities may have been more ready to move because their traditional way of life was under threat. More than a few longstanding coastal villages – including many of the terpen we met in Chapter 2 – came to an end in this period. Abandonment stretched, in fact, over a very wide area: from the Frisian coast to the Elbe–Weser region, all the way to Schleswig-Holstein. In the past, this phenomenon led some to suggest that rising sea levels were the fundamental cause of Anglo-Saxon migration, but this is overstating the case. Eastern England, where many of the migrants ended up, was also affected by rising sea levels, and eventually land was abandoned well beyond the coastal regions of Saxony as the fifth century turned into the sixth. At most, then, rising sea levels can have been only a secondary factor. By the sixth century too, Merovingian Frankish power began to intrude aggressively into the Saxon homeland. It was this political factor that prompted the exodus of those twenty thousand Saxons who eventually joined the Lombards, and there is no reason why Frankish pressure should not have led others to join their peers across the North Sea. Nonetheless, a broadly voluntary, economic motivation was probably the main cause of the Anglo-Saxon migration flow, since it began long before the Franks became a factor. This is also suggested by its basic nature.33 A drawn-out process, as opposed to a sudden surge of migrants, does more suggest the steady pull of economic attraction than the impact of major political crisis, such as that which propelled the Goths across the Danube in 376.

  As the existence of the litus Saxonum also shows us, the active field of information necessary for any migration flow already existed between lowland Britain and northern Germania by the year 400. Anglo-Saxon migration was exploiting known routes, and in some ways represented merely an extension of a pre-existing tendency towards Germanic expansion in this direction. The wealth of Roman Britain was well known to Saxon raiders of the third and fourth centuries, whose understanding no doubt included plenty of information about coastal and North Sea waters, and the best routes to take to target areas. This will also have included considerable knowledge of inland areas, since all rivers leading to the interior of Britain will have been part of the zone explored by ship-borne attack. First-millennium boats were small enough to go far inland along the rivers and were not restricted just to immediate coastal hinterlands. Gildas’ account suggests that this fund of information continued to expand as the fifth century progressed, very much as a developing knowledge base underpins modern migration flows. The first Saxons may well have been a mercenary outfit, as Gildas reports, hired by former or potential victims of previous predations to help in their defence. Such a move had been prefigured, it seems, at the end of the third century, when a usurping Roman commander in Britain, Carausius, initially appointed to fight Saxon and Frankish pirates, incorporated some of them into his forces. It was also fairly common in the later Viking period. Sea raiders were difficult for land-based forces to combat. News of the mercenaries’ prosperity led others to join them on the British side of the North Sea. This need not always have been the deep, dark plot that Gildas supposed. The original mercenaries may have signed up in good faith, but, as the situation developed – in other words as the information field expanded – their ambition increased, perhaps, or new Saxon groups with greater ambitions saw the opportunity for self-enrichment on a grand scale and moved in on the action, just as small-scale raiders in the Viking period were eventually supplanted by more important leaders with larger followings.34

  It was not new geographical information about lowland Britain, then, that transformed Saxon raiding into Saxon migration as the fifth century progressed, but an increased understanding that the previous political and strategic situation had been transformed out of all recognition. As long as Britain remained part of the Roman Empire, any serious attempts by continental Saxons to annex its landed assets were doomed to failure. The forces of the litus Saxonum were more than powerful enough to deal with raiders not smart enough to beat a hasty retreat: exactly the fate suffered by those Saxon raiders who turned up in Gaul in the time of Ammianus. Once Britain fell out of the Roman system, however, much more than the odd hit-and-run raid became possible, and not only for ambitious Saxons. As Gildas reports and other evidence confirms, raiders and even immigrants from Ireland and Scotland – Scots and Picts respectively – were also queuing up to feast on the remains.35 The pull of a developed, Roman-style economy underlay the entire sequence of events, and, as with the other migration flows we have examined, the intermixing of politics and economics could not be clearer. The wealth of lowland Britain could become available to Saxon migrants only if they took its political control into their own hands, and this only became possible when Britain lost its Roman umbrella. It took perhaps as much as a generation for the continental intruders to realize how vulnerable their preferred British targets now were. Raiding seems to have begun already, as we have seen, in c.410, but it was c.440 before the situation turned really nasty, at least according to continental observers. This is a plausible time lag for the Anglo-Saxons either to come to realize that the old blocks on full-scale expansion had been removed, or to develop a range of new ambitions that went beyond raiding to outright annexation.

  Prevailing political structures also shaped the action on another level. Compared with contemporary migratory phenomena on the continent, what’s striking about the Anglo-Saxon case is how much evidence there is for small-scale activity. By 600, the end result of the migration flow was, as we have seen, a series of relatively small Anglo-Saxon kingdoms. The same is true of the Romano-British world, which, in the era after Aurelius Ambrosius at least, fragmented in political terms. Looking at this pattern, one recent study has questioned why there were so few examples from fifth-century continental Europe of the kind of local takeover of power evident in the person o
f Aurelius Ambrosius, or in the quite small kingdoms which dominated Cornwall and Wales in Gildas’ day.36

  In part, this oddity was a result of transport logistics, whose impact went far beyond a possible skewing of gender ratios. Anglo-Saxon migration into Britain had to take the form of an extended flow of population rather than a single pulse of invasion because of the impossibility of transporting large numbers of people across the North Sea at one go. In one respect, the evidence is equivocal. We don’t know whether the population of Jutland customarily used ships with sails in the fifth century or not, and oar-powered vessels could have transported large numbers of passengers only on a one-way trip since there was little room in them for other than the rowers themselves. But sail-powered ships were available just a little way down the coast in Roman Channel ports, and there is no reason to suppose that their captains were not being paid to carry Saxons in larger numbers to Britain (as their counterparts elsewhere carried Goths and others across the Black Sea in the third century, and the Vandal–Alan coalition across the Straits of Gibraltar in 429). Other Saxons, as we know, were making their way as far as the Loire at the same time, which would have been one hell of a row. More important than the sail issue, therefore, is the fact that the available ships were all small, and limited in number. Just as more modern migration across the Atlantic was limited to a steady trickle until the advent of transatlantic liners in the later nineteenth century, so it was logistically impossible for large numbers of Anglo-Saxons to arrive en masse on the British coast.37