Empires and Barbarians

by Peter Heather

  The second problem is more fundamental. These early Anglo-Saxon burials take two basic forms. In central and southern England, archaeologists have uncovered a large number of quite small inhumation cemeteries, some of the burials being richly furnished with gravegoods. Further east, in parts of East Anglia and along the northeast coast, a smaller number of much larger cremation cemeteries have been excavated (Map 11). The cremation cemeteries raise few problems of attribution. A cremation habit was entirely foreign to late Roman Britain, and both the burial form and the identifiable objects that survived the cremation process have direct antecedents in fourth- and early fifth-century materials from south-eastern Jutland. There is not much doubt, then, that Germanic-speaking immigrants from the Jutland region generated the cremation cemeteries of eastern England.12

  The inhumation cemeteries are more problematic. For one thing, they contain a large number of burials without gravegoods. Who were these people? Were they poorer Anglo-Saxons, left-over Romano-Britains, whose standard burial rite had indeed been unfurnished inhumation, or people who just chose not to bury their dead with gravegoods? Likewise, while there is little doubt that many of the items found in the furnished graves (brooches, sleeve fasteners, weapons and so on) were first made and used by continental Germanic-speaking populations, this is not true of them all, and, more generally, it can be argued that their appearance and spread in England is not a safe guide to the extent of Anglo-Saxon immigration. Unlike the cremation rite of eastern England, the dress items found in the inhumation cemeteries were not lifted lock, stock and barrel from one particular corner of the Germanic-speaking continent. Particular combinations of items eventually became confined to specific corners of England, but many of these items originated in disparate areas of Germania. Sleeve fasteners, for instance, became a distinctive element in the dress of early Anglo-Saxons living just inland from the Wash, but whereas most of what they wore had disparate origins the fasteners themselves had only been found earlier in parts of western Norway.13 It looks, in other words, as if the process that unfolded in lowland Britain was similar to that underlying the so-called Danubian style of Attila’s Empire (Chapter 5). New and distinct Anglo-Saxon dress combinations coalesced out of a wide variety of sources in lowland Britain in the fifth century.

  If dress items and habits were being passed around between different groups of Germanic immigrants, there is no obvious reason why they could not also have been passing from Anglo-Saxon immigrants to Romano-British natives. We are all now comfortable with the idea that under certain conditions new identities can be adopted. This tends to happen particularly when old identities are in flux, which was true both for Anglo-Saxon immigrants and Romano-British natives in the fifth and sixth centuries. The boundaries of the new Anglo-Saxon kingdoms differed from the pre-existing political structures of Britain, and there is no reason to suppose, either, that they had been transposed from the continent. Famously, individuals with British names – Cerdic and Cynric – appear in the ancestry of the ruling house of Wessex, and the late seventh-century law code from this kingdom, known as Ine’s Law, explicitly mentions the presence within it of substantial landowners of indigenous non-Anglo-Saxon descent. These are strong indications that Wessex may have been created by a complex Anglo-Saxon/Romano-British double act, rather than a simple Germanic conquest. The cemetery of Warperton in Warwickshire also provides a so far unique example of a chronological progression from late Roman to Saxon-style burials within the single graveyard. This too is suggestive of processes of cultural assimilation. Particularly since many of the inhumation cemeteries continued in use for two hundred years, from the fifth into the sixth and seventh centuries, by which time there must have been much intermingling between immigrant and native, it is perfectly reasonable to question whether continental dress accessories can really be used as a guide to the origins of the corpse found displaying them.14

  Neither the available archaeological nor the available literary evidence answers in any straightforward way, then, the key questions about the extent and nature of Anglo-Saxon immigration. Nor, unfortunately, is there any immediate prospect of new methodologies or sources of information filling the gap. DNA testing and isotope analysis have both come on stream in the last decade or so to offer new paths forward, but neither looks likely to provide easy answers. So far, it is unclear whether ancient DNA can really be extracted from fifth- and sixth-century skeletons preserved in typically damp British conditions. The jury remains firmly out, and while it is deciding, less direct routes have been explored. The most important concerns the distribution of particular gene combinations within the male Y chromosome among modern English males. This is potentially very useful. The Y chromosome is handed down unchanged over time from father to son through the male line, and there is one gene combination which can with some plausibility be linked to an intrusive population group of males moving from northern continental Europe into lowland Britain in the middle of the first millennium AD. This gene combination is now very widely distributed among modern Englishmen, being found in 75 per cent or more of those sampled.

  But how should this exciting new evidence be interpreted? The researchers initially argued that their findings confirmed what the Victorians had always thought, that something akin to ethnic cleansing took place during the Anglo-Saxon invasions with the 75 per cent distribution among the modern population reflecting a 75 per cent replacement of males in the fifth and sixth centuries. Given, however, that arriviste Anglo-Saxon males formed – on any estimate – a new elite in the land, and had therefore greater access both to food and to females, you have to figure that they had a bigger chance of passing on their genes to the next generation than the indigenous Romano-British. And more recent mathematical modelling by the same researchers has shown that you don’t have to make that breeding advantage very large for the 75 per cent result among the modern English population to have been generated from an intrusive male group that was originally no larger than 10–15 per cent of its fifth- and sixth-century counterpart. Self-evidently, therefore, the modern DNA evidence is not going to settle the quarrel between those favouring mass Anglo-Saxon migration and those persuaded by elite transfer and emulation.15

  Nor does isotope analysis look any more promising in overall terms, although it generates fascinating individual results. This technique works on the basis that the mineral contents of an individual’s teeth carry the signature of where they grew up, transmitted into the chemical composition of dental remains by the water drunk as a child or teenager, depending on whether you’re looking at baby or adult teeth. Some of these chemical signatures can be recognized as belonging to particular regions, where those regions are geologically distinct. Potentially, therefore, you might be able to show whether an individual buried in Anglo-Saxon clothing really came from the continent, or was an identity-swapping Romano-British wolf. The problem, however, is that the technique will work only for first-generation immigrants. The child of two echt Jutlanders, if born after they crossed the North Sea, will have grown up with absolutely East Anglian teeth. Much expensive sampling and a lot of very precise chronological identification would be required, therefore, for isotope analysis to produce any broad conclusions. And given that the offspring even of first-generation Anglo-Saxon immigrants will have had British teeth, I doubt that it ever will. For the moment, therefore, neither isotope nor DNA analysis offers us an obvious way past the intellectual impasse between mass migration and elite transfer originally generated by the limitations of the traditional historical and archaeological evidence.16

  So by themselves, the available bodies of source material set up an intellectual problem, but offer no obvious solution to it. The historical evidence is much too thin to provide a convincing picture of what went on in lowland Britain in the fifth and sixth centuries, while the major transformations in material culture can be explained either in terms of mass invasion or mass cultural emulation. The new understanding of late Roman Britain, likewise, shows that th
e province’s population was much too large to make ethnic cleansing even the remotest possibility, but the linguistic evidence from post-600 AD shows precious little sign of indigenous influence on the Germanic tongue of the Anglo-Saxon world which emerged from Britain’s late antique Dark Age. The argument is more than a little stuck, but if we think first about the evidence of the Anglo-Saxon migration flow itself rather than its effects upon Britain, and then reconsider the issue of mass migration versus elite transfer from a more comparative perspective, it does become possible to move the argument past its traditional roadblocks.

  Adventus Saxonum

  The historical work of the Venerable Bede, written in his Jarrow monastery in the early eighth century, provides two dates for the Adventus Saxonum: the arrival of Saxon invaders in Britain. The first is 446, based on Gildas’ record that the British appealed for imperial assistance to Aetius, de facto ruler of the Roman west in the mid-fifth century, when he was ‘consul for the third time’. Gildas did not date this appeal, but Bede had access to Roman consular lists which told him that Aetius’ third consulship fell in the year 446. His other date is c.450–5, derived from a Kentish dynastic tradition of his own time that its founders had established themselves in that south-eastern corner of Britain during the joint reign of the Emperors Marcian and Valentinian III.17 Among more modern scholars, however, there is widespread agreement that, whatever its scale, the flow of Anglo-Saxon migration into lowland Britain was not a one-off event, but a long-extended process.

  The only account of its origins is to be found in Gildas’ Ruin of Britain. This tells us that the Anglo-Saxon takeover stemmed from extensive attacks made by Picts and Scots (from Ireland and Scotland respectively) upon the British provinces after they had dropped out of the Roman imperial system. Plenty of controversy surrounds the details, but other contemporary sources tell us that in 406 or thereabouts the Roman army in Britain had put up the usurper Constantine III, who shifted his command to Gaul to fight the Rhine invaders. Eventually, perhaps in 409, the British provinces revolted again, breaking away – it seems – from the usurper’s control. Shortly afterwards, they may or may not have received a letter from the western Emperor Honorius telling them that they would have to look after their own defence. At this point, however, such a letter would have been no more than de jure recognition of the de facto situation. Honorius could do nothing to help, and the Britons found themselves in a decidedly sub- or post-Roman situation.18

  It is at this point that Gildas apparently picks up the story. The difficulties facing the Romano-British, now independent, eventually became so severe, that

  they convened a council to decide the best and soundest way to counter the brutal and repeated invasions and plunderings . . . Then all the members of the council, together with the proud tyrant, were struck blind; the guard . . . they devised for our land was . . . the ferocious Saxons . . . A pack of cubs burst forth from the lair of the barbarian lioness, coming in three keels, as they call warships in their language . . . On the orders of the tyrant they first of all fixed their dreadful claws on the east side of the island, ostensibly to fight for our country, in fact to fight against it. The mother lioness learnt that her first contingent had prospered, and she sent a second and larger troop of satellite dogs . . . [Eventually the Saxons] complained that their monthly allowance was insufficient . . . and swore that they would break their agreement and plunder the whole island unless more lavish payment were heaped upon them. There was no delay, they put their threats into immediate effect . . . A fire heaped up and nurtured by the hand of the impious easterners spread from sea to sea. It devastated town and country round about, and, once it was alight, it did not die down until it had burned almost the whole surface of the island and was licking the western ocean with its fierce red tongue.

  Despite all these defeats, which caused many British, Gildas tells us, either to surrender themselves into slavery under the invaders, or flee overseas, the Romano-British were not finished. Even when Aetius turned down their final appeal for imperial assistance, they continued to resist. One of their number, the famous Aurelius Ambrosius, historical prototype of the mythical Arthur, organized a counterattack which culminated in a great British victory at the siege of the unidentified Badon Hill. After this, prosperity returned to the island, a happy state of affairs that lasted for the entire and considerable intervening period down to the moment that Gildas wrote.19

  One major problem with Gildas’ account from the modern historian’s perspective is chronological imprecision. When did the events he describes begin? Gildas gives no indication at all of when the council might have issued its original and ill-fated invitation to the Saxon mercenaries. Bede clearly supposed that all the action from invitation to revolt and beyond unfolded in quick succession, and hence dated the arrival of the Saxons to 446, on the basis of the appeal made in the middle of the mayhem to Aetius when he was consul for the third time. Most modern scholars would argue that the action was much more drawn out, on the basis of contemporary sources of reasonable quality which record some major Saxon attacks on Britain already in c.410. This makes it likely that the original invitation for mercenary assistance would have been issued a generation or so before Bede supposed, which is compatible with Gildas’ actual wording. Gildas’ account then becomes a brief summary of a lengthier sequence. A more extended chronology also fits well with the fact that the earliest datable Saxon remains in England belong to the 430s.20

  As for the Saxon revolt that spread its fire ‘from sea to sea’, the best chronological fix we have on this part of the action may come from a continental source, the so-called Gallic Chronicle of 452 (so called in a fit of wild scholarly fancy because it was put together in Gaul in 452), which reports that Britain fell to the Saxons in 441/2. This chronicle was composed only a decade after the reported events, and we know that considerable contact had continued between the Romano-British and Roman Gaul after 409 – on which more in a moment – so that it is actually a pretty decent piece of evidence. There are other possible ways of construing events, certainly, but it seems most natural to associate Gildas’ mercenary revolt with the mayhem of the early 440s. And by the 460s, at least one important British leader was operating in northern Gaul, in the Loire region, which would tie in with Gildas’ report that some British fled overseas. And even if you were to deny this specific association, one period of major Saxon invasion and British disaster would anyway have to be dated to the mid-fifth century on the basis of the Chronicle.21 That, however, wasn’t the end of the story. Gildas closed his historical excursus on a remarkably high note, which is one reason why his text has sometimes been considered a late fake. Thanks to Aurelius Ambrosius, the Romano-Britons were eventually successful, and, while detailed geography is non-existent, the general sense of Gildas’ account is that in the forty-year peace that followed Badon Hill the Saxons were confined, at most, to the eastern end of the island. Most historians would date this peace to sometime between 480 and 550 AD.22

  When Bede’s detailed historical narrative begins in earnest, however, with the arrival of the Roman mission in Kent in 597, virtually all of lowland Britain was firmly under Anglo-Saxon control. Either there had been further dramatic Anglo-Saxon advances in the mid-to late sixth century, or the level of Romano-British success implied by Gildas is massively misleading. The available evidence suggests the former. For what it’s worth, the Anglo-Saxon Chronicle places a major expansionary phase in the history of Wessex in the late sixth century when, under the leadership of one Ceawlin and his nephew Ceolwulf, great tracts of Devon and Somerset first passed into Anglo-Saxon hands. Despite the text’s problems, this may preserve echoes of an important later phase of expansion. Most of the royal dynasties that controlled the seventh-century Anglo-Saxon kingdoms which appear in Bede’s History, likewise, seem to have descended from an ancestor whose floruit dates to the last quarter of the sixth century rather than earlier.23 This again suggests that something important happened after Gil
das had finished writing.

  Continental evidence adds further weight to the argument, showing that Saxon populations remained highly mobile into the sixth century. One substantial group of Saxon migrants, reportedly twenty thousand strong, moved south in its middle years, eventually participating in the Lombard invasion of Italy. Another established an enclave at the mouth of the Loire at more or less the same time (the 560s).24 Such unimpeachable evidence of continued demographic upheaval in the Saxon homeland makes it entirely plausible to suppose that still more Saxons were at the same time following the route to Britain. Continental influence from a different direction also shows up in some of the later archaeological materials. Contacts of some kind between Scandinavia and East Anglia, quite possibly a new migration flow from Norway in particular, were established from the late fifth century, and there is some reason for thinking that the East Anglian royal dynasty had Scandinavian roots. Bede, in fact, generally agrees with the archaeology. He reports that Germanic immigration into Britain drew upon a very wide range of manpower: not just the Angles, Saxons and Jutes he mentions in the first book of his Ecclesiastical History, but also Frisians, Rugi, Danes and others besides.25 Burials in the Anglo-Saxon inhumation cemeteries increase in frequency, likewise, from the late fifth century onwards: from one in every four years in c.500 AD to one every two to three years by c.600.26 This can be explained in several ways: the conversion of Romano-British to Anglo-Saxon cultural norms, or a natural increase in numbers among the immigrants. All the same, something apparently tipped the balance of power established at Badon Hill firmly in favour of the Germanic-speaking immigrants – or at least the dominance of their cultural forms – in the mid-to late sixth century. In all probability, continued immigration from the continent played some part in the process.