`The barbarians over the Rhine ... reduced the inhabitants of Britain and some of the Gallic peoples to such straits that they revolted from the Roman empire, no longer submitted to Roman law, and reverted to their native customs. The Britons, therefore, armed themselves and ran many risks to ensure their own safety and free their cities from the attacking barbarians.' - Zosimus, late fifth century.'
`However, the Romans never succeeded in recovering Britain, but it remained from that time on under tyrants.' - Procopius, mid late sixth century.
Sometime after 446 Aetius was said to have received an appeal for aid from the Britons who were under attack from the Picts, Scots and other barbarians. They complained that `the barbarians push us back into the sea, the sea pushes us back into the barbarians; between these two kinds of death we are either drowned or slaughtered'. There had not been Roman governors in Britain for a generation, but the island was still clearly considered part of the empire in a general sense. Leading Britons obviously felt the same way, hence their appeal to the military commander of the west. In the event, Aetius had other priorities and sent no aid. Left to their own devices the British `councillors' agreed with a local warlord or king - literally a `proud tyrant' - to hire Saxon mercenaries. These beat back the northern barbarians, but then turned against their employers, sacking towns and forcing many to flee across the sea. Eventually, a nobleman named Ambrosius Aurelianus, described as the `last of the Roman race', emerged as leader of the survivors. The Britons won some victories, culminating in a great triumph at Badon Hill.'
The story is first told by the British cleric Gildas sometime in the sixth century. He provides no dates, although the impression is that this account covered a considerable period - decades at the very least. He does mention that the initial appeal was to a man who was three times consul, and this was only true of Aetius from 446-454. Actually, Gildas calls him Agitio, but a later version of the story was surely right to correct this to Aetius. Yet the mistake does raise the question of just how much Gildas actually knew of events a century or more before his lifetime. He was also not writing a history and this passage comes from the introduction to a bitter attack on the `tyrants' and priests of his own day. Literary sources for fifth-century Britain are very sparse and were almost all written long after the events they describe. Some facts may be accurate, others confused and perhaps merged with myth, or deliberately distorted by later propaganda. Separating these strands is certainly not easy, and a few scholars would say that it was altogether impossible.4
Yet there is nothing inherently implausible about Gildas' account. The Saxons - a term that was used at the time to embrace a range of different groups including the Angles, Jutes and Frisians - did end up dominating much of what would become England by the late sixth century. Other sources suggest that there was serious conflict with them in the middle years of the fifth century. That some Saxons were hired to fight against other `barbarians' and later came into conflict with their employers was a familiar enough tale in the Roman world at this period. We also know that significant numbers of Britons fled to north-western Gaul, so that in time Armorica became known as Brittany.'
Dates are rare in the sources, and often both they and some of the accompanying details must be suspect. There is archaeological evidence for the period, but even by normal standards this presents considerable problems of interpretation. As a result, radically different accounts of life and politics in fifth-century Britain continue to be produced. On the fringes of this there is the constant flow of material about Arthur, much of it aimed at the more popular end of the market and extending into fiction and film. This varies from quite serious history to highly fanciful studies. Gildas never mentions Arthur and it is a later source that associates him with the victory at Badon Hill. It is better to begin trying to understand the Britain of this era before any attempt to describe an `historical' Arthur. Recently, there has been much more emphasis by academics on placing events in Britain within the wider context of the history of the Western Empire. This has proved fruitful, although if anything has added to the variety of interpretations of the same evidence. Here the emphasis will be the other way around, looking at what the experience in Britain tells us about the last years of the Western Empire.'
Britain was one of the last major additions to the Roman Empire. Julius Caesar had landed in the south-east in 55 Bc and returned with a larger force in the following year. There was no permanent occupation. The expeditions were huge propaganda successes, but achieved little in practical terms and did not result in the creation of a province. Trade with Britain increased massively in the following decades and there was some diplomatic contact. A string of royal refugees fleeing the power struggles within and between the tribes of south-eastern Britain arrived at the imperial court seeking support. Augustus decided against intervening, feeling that the cost of occupation would be greater than any likely profit.
In 43, the emperor Claudius was desperate for military glory to cement his tenuous hold on power. Therefore he ordered a massive expedition to invade Britain and even travelled to the island himself. The tribes of the south-east were quickly overcome or surrendered. Progress elsewhere was slower, and it is not at all clear just how much of Britain the Romans planned to conquer. In 6o they came close to losing the territory they did control. Queen Boudica of the Iceni rebelled and was joined by many other previously pro-Roman tribes. The three largest cities of the province - Londinium (London), Camulodunum (Colchester) and Verulamium (St Albans) - were all sacked. Finally a decisive battle, followed by ruthless punitive action, broke the back of the rebellion and it was never repeated. Over the next decades more conquests were made in the west and north. What would become Wales and the north of England were occupied only after heavy fighting. In 84 a Roman army won a victory somewhere in Scotland, while elements of the fleet circumnavigated Britain and proved that it was an island.7
Claudius sent four legions and a strong force of auxiliaries to invade Britain. A generation later the garrison was reduced to three legions, although the number of auxiliaries seems to have increased. One estimate of the garrison of Britain in the middle of the second century places this as high as 50,000 men, although this rather assumes that all units attested in the province were there simultaneously. Even if the actual garrison was smaller, it was certainly a substantial part of the entire Roman army - somewhere between one-tenth and one-eighth. Some troops were stationed in the west, especially in Wales, but the bulk of the provincial garrison was deployed to the north. It was also in the north that a series of frontier defences were created, before the main line settled permanently on Hadrian's Wall. All of this was immensely costly. It was also dangerous for such a large army to be given to a single governor, and it was no coincidence that one of the challengers for the throne in 193 was the legate of Britain. Septimius Severus' campaigns against the Caledonians may have permitted a substantial reduction in troop numbers. Barrack blocks built in the forts of Hadrian's Wall in the following years all appear to have been about half the size of those in earlier periods. It is quite likely that centuries in these units were halved in size from eighty to forty men, although they still remained under the command of a centurion. If this occurred more generally, then the size of the army in Britain may have been reduced by as much as 50 per cent. The provincial command was also divided.'
Even if the garrison in the third century was substantially smaller than in earlier periods, it was still large and costly. Britain's mineral resources were exploited from very soon after the conquest. The island also produced a substantial surplus of grain, and a large part of this was either from imperial estates or taken by the state as tax, so that British wheat helped to feed the troops stationed in the Rhineland. Even so, it is doubtful that the profits of occupying Britain ever covered the expense of maintaining government and garrison there. Over time just under thirty cities were created. Most were local capitals and administrative centres for groups based around the old tribes. Some of the cities, notably London, Cirencester, Silchester and St Albans (Verulamium) were large and in due course gained basilicas, theatres, amphitheatres and bath houses. No circus with chariot racing was known in Britain, until one was discovered in 2004 at St Albans. In the fourth century all major cities almost certainly did build large churches - a basilica-type building that may well be a cathedral has been identified in London. Other capitals were more modest and it is fair to say that none acquired the splendour of so many cities in other provinces, especially those nearer the Mediterranean. In many areas there were no cities and only what are known as `small towns' by archaeologists. Usually lying on the main roads, these communities acted as market towns and housed various local industries.'
Many British aristocrats sided with the Romans from the beginning and did very well out of the conquest. The grand villa or palace complex at Fishbourne was built in the first century, most probably as a residence for the client ruler Tiberius Claudius Togidubnus, who was described as a `great king'. It was normal Roman practice to win over the leaders of conquered peoples, and prominent British families soon gained citizenship and a Latin education. However, it does seem to have taken the British nobility a very long time to break into the higher levels of imperial service. Many built villas and grand town houses, but the ethos of civic life so typical of other provinces was less developed in Britain and it seems to have been rarer for them to spend money on major endowments for their communities. Britain has produced markedly fewer inscriptions than most other sizeable provinces, and a large part of those that do survive are military. In many areas the focus of life remained essentially rural. Much of the population continued to live on farmsteads and in small villages. Some changed to more Roman styles of buildings, but others continued to live in the traditional - and highly functional - roundhouses familiar from the Iron Age.'°
Geographically, Britain was on the very margins of the empire - indeed, of the world as far as Greeks and Romans were concerned. There was contact with Ireland, but the Roman official line was that it would not be worth the expense of conquest and occupation. Britain required a large garrison and never quite developed as much or in the same way as some of the provinces nearer the heart of the empire. It might therefore be tempting to see its conquest as a failure, a costly burden imposed on the empire by the vanity and urgent need for glory of Claudius. This would be a mistake. Not all provinces developed identically, and Britain remained under Roman rule for three and a half centuries during which time life there changed profoundly. If there was a significant drop in troop numbers in the third century, then this may have brought the province closer to making a profit for the empire. If it was not the wealthiest province in the empire, there was still considerable prosperity for a wide section of the population. The recent claim that `for every winner under Roman rule, there were a hundred losers' is very hard to substantiate, and the same author himself notes that `it does not follow that life would have been any better without Rome.' It does seem true that the gap between richest and poorest widened. Then as now the fact that some of the population became substantially wealthier does not automatically mean that the lifestyle of the remainder declined or that they became poorer in real terms. The sheer quantity of finds on virtually every British site from the Roman period compared to Iron Age or postRoman occupation makes it clear that many objects were far more readily available to the general population."
Willingly or not, the population of Roman Britain accepted Roman rule. There was some armed resistance, particularly in northern Britain, but nothing to unite the communities elsewhere in opposition to imperial government. Britain seems to have been spared the worst of the disruptions of the third century, if only because it was physically harder for large armies to cross the seas to reach it. Even if numbers were less than at their height in the second century, there were still significant forces stationed on the island. This, and the tendency for emperors to see Britain's problems as distant and rarely urgent, made it a fertile ground for usurpers. Constantine the Great was by far the most successful, but all of the others also had ambitions beyond simply controlling Britain itself and all took troops across the Channel. This weakened the provincial garrison, but did mean that, with the exception of the suppression of Allectus, all of the campaigns in the resulting civil wars were fought outside Britain.
By the end of the fourth century Britain formed a diocese under a vicarius based in London and responsible to the praetorian prefect. The diocese was subdivided into either four or five provinces - the existence of a fifth is uncertain. Valentinian I formed a province named Valentia after himself, but it is unclear whether this involved the creation of a new province or the renaming of an existing one. The Notitia Dignitatum lists three military commands in Britain. The Comes Britanniae commanded a force of comitatenses consisting of three infantry and six cavalry units. The proportion of foot to mounted regiments is unusual, even if the former were normally larger than the latter. It suggests a force tailored to chase small bands of raiders rather than fight massed battles. This small field force is usually seen as a late creation, perhaps by Stilicho, given recorded cases of despatching field army units from Gaul to deal with problems in Britain during the fourth century. The Dux Britanniarum commanded units of the limitanei, mostly stationed in the north and including the garrisons of some named forts on Hadrian's Wall. Finally, there was the `Count of the Saxon Shore' (Comes Litoris Saxonici per Britannias) controlling limitanei based around the east and south coasts from Brancaster, near the Wash, to Portchester, not far from modern Portsmouth. One survey estimated the maximum total strength of these troops as 20,000, but guessed that it was actually lower at nearer 12,000. As always, the real number of effectives at any one time is likely to have been substantially less than the army's paper strength.12
The garrison of Britain created three usurpers in 406-407. There were clearly still enough troops for the last of these, Constantine III, to cross into Gaul and gain control of a large part of the Western Empire. He must have taken some, perhaps most, of the British army with him and it is unlikely that any of these troops returned. This can only have weakened the defences of Britain, just like previous, ultimately unsuccessful usurpations. The garrison was weaker, but opinion is divided over precisely what threats it faced. No one would dispute that there were enemies to the north. The Picts - the name was most likely derived from `picti' or `painted men' because it was believed that they commonly wore tattoos - seem to have emerged when the old Caledonian tribes became a little more united. To their west were the Scotti, who are likely to have migrated from Ireland and would eventually give their name to Scotland. Both peoples had launched serious raids into the British provinces during the fourth century. Many of the attacks came by sea along the coast, and there seems also to have been some forays made by the tribes still living in Ireland. The sixteen-year-old St Patrick was taken as a slave by just such a group of raiders, although it is unclear whether this took place before or after the end of direct Roman rule."
The name `Saxon Shore' is only attested in the Notitia Dignitatum. Ammianus mentions the Saxons launching raids against Britain in 367, as part of a simultaneous onslaught by the Picts, Scots and also the Franks. Otherwise there is little explicit evidence for Saxon raiding on the coast of Britain, unlike frequent mention of their attacks on the northern coast of Gaul. This may simply be because the sources are much better for Gaul and extremely sparse for Britain. A literal reading of the accounts of Carausius' operations against pirates in the English Channel would suggest only attacks on Gaul, although raids on the British coast are usually assumed. If the Saxon Shore was named after the enemy it was supposed to defend against, then this would make it unique in Roman history. On the other hand, the suggestion that it was named because large numbers of allied Saxon troops were stationed or settled there is even less convincing and unsupported by any evidence.14
Some have questioned the ability of raiders from what is now northern Germany and Denmark to reach Britain. Far less is known about the sea-going vessels of the tribes in this period than the ships of the Viking age. The few to survive archaeologically may only have been intended for inland waterways. None have sails and it has been claimed that their keels were too small to have supported a strong enough mast to mount one. It would have been possible for a group of warriors to row to Britain and return with any spoils, but it would certainly have been difficult. More probably, we have simply not yet discovered an example of a sailing ship intended for longer voyages. A carving of a boat with a sail has been found in Denmark and it does seem inherently unlikely that the peoples of the area never adopted this technology. It is harder to say how well a shallow-keeled boat would perform under sail, but it is important to remember that fighting at sea was always rare. These vessels were only ever meant to deliver and carry off a group of warriors."
It is doubtful that any such vessel was especially large, and later - admittedly somewhat questionable - sources tend to talk of between one and four boats in any force. A raiding group numbering hundreds of warriors was probably exceptionally large and most parties would have been smaller. Normally they would only strike at targets near the coast or reached along an easily navigable river. Other than in their means of transport these attacks were essentially the same as other barbarian raids. The late fourth-century military writer Vegetius mentioned that the English Channel was patrolled by small warships, whose sails, rigging and even the sailors' uniforms were coloured to blend in with the sea. Although this may well have made them harder to see at a distance, it can rarely have been possible to intercept the raiders on their way in. They were more likely to be caught before or after they landed, and most of all as they retreated. Again, the pattern was much like raiding on land and, in just the same way, successful expeditions would encourage more attacks. Maritime raiding does seem to have been fairly common throughout the Roman period. It may well have occurred before this period and would certainly persist for many centuries after the Western Empire had gone, most famously in the Viking Age.'6
The Saxon Shore forts varied in design. They were probably built over a long period and not as a single planned system. Other third-century forts, such as the one on the Taff at Cardiff, which were not included in the later command, seem to share many common features. All were strongly fortified and situated on a navigable river, usually at its mouth. Little is known of their internal arrangements. The purpose of the Saxon Shore forts has been as hotly disputed as their name. One suggestion has been to see their role as primarily logistic. This assumes that grain from imperial estates or gathered as tax was transported along the rivers to be massed in a supply dump within the Saxon Shore fort. It could then be transported by the Roman navy to the Rhineland or wherever else it was required. There is no evidence at all to support this, and it has largely been based on the mistaken assumption that there was no significant Saxon raiding.'
The Saxon Shore forts could not have prevented every attack, but did restrict access to the major rivers. This made it harder for raiders to reach deep into the countryside. We do not know whether by the start of the fifth century the Romans still maintained naval squadrons in Britain as they had done in earlier years. None are mentioned in the Notitia Dignitatum. Seaborne patrols would have been useful in making raiders more wary and might on occasions have caught them on shore or returning home. It is impossible to know when Vegetius' camouflaged patrol boats ceased to operate. As a general rule naval forces were and are more expensive to operate than troops on land and so are especially vulnerable to shortages of funds. The garrisons of the forts would have been capable of patrolling the land around their bases. Like anywhere else, there were not enough troops to keep the entire frontier perfectly secure. The best the Romans could hope for was to make it dangerous for raiders to operate, intercepting enough groups to prevent the coast of Britain becoming seen as an easy target. Their presence may have done a little to make the local population feel that they were protected, and the larger forts such as Porchester may also have served as places of refuge.
All of Britain's cities were strongly walled by the fourth century. Opinion is divided whether they were in decline by this period. There are cases of public buildings such as basilicas and theatres falling into ruin. Significant numbers of large town houses were also abandoned. Only a minority of communities seem to have had a working public bath house by 400. Yet cities were clearly not abandoned. There was still some building, although more often in timber than stone, and several former public buildings were turned into workshops or factories. More problematic is what is known by archaeologists as `dark earth', a thick layer of dark grey soil often containing plant remains, animal bones and charcoal, which is found on top of earlier Roman buildings on many urban sites. Although some have seen this as traces of less well-constructed timber structures, more probably it represents the abandonment of these parts of a city for building. It may have been cultivated or perhaps no more than a convenient dumping ground. Perhaps the population of many cities and of the small towns declined. Certainly, they were less grand than they had once been, but this in itself does not mean that they ceased to function altogether."
The fourth century saw the construction of many of the largest and most luxurious villas ever built in Roman Britain. It is possible that some aristocratic families chose in these years to spend more time on their estates than in their town houses, but equally, completely different factors may have caused this phenomenon. By the end of the century very few new villas were being built on such a scale and some existing ones fell into disuse. As with the decay of major buildings in the towns, this did not in itself mean that the lands were abandoned. The estate may have continued to function as a unit based around a rather humbler dwelling and we really cannot say very much at all about the state of the rural economy. There is also little direct archaeological evidence for Christianity in Roman Britain. Pagan temples did continue in use especially in the rural areas, but suggestions of a pagan revival in the countryside are unconvincing. A better case can be made for the majority of the urban and rural population being at least nominally Christian by the start of the fifth century."
Whatever the gradual changes in the fortunes of towns, villages and country estates, the end of formal Roman rule in Britain was both abrupt and unexpected. Constantine III flourished for four years before finally suffering defeat. Many of the prominent men in Britain had already lost enthusiasm for his rule. As far as we can tell he had shown no real interest in Britain or its problems once he crossed to the Continent. His resources were stretched very thin and the success he enjoyed was really just a reflection of the weakness of central government. Around 407-408 some British leaders rebelled and expelled Constantine's officials. Zosimus tells us that in about 410 the rebels appealed to Emperor Honorius, who replied from Ravenna `urging them to fend for themselves'. Doubt has been raised about this passage and some have suggested that a copyist's error has changed Bruttium in Italy to Britain. This is not especially convincing and raises problems of its own. In the end, it probably does not matter whether or not Honorius actually did instruct the leaders in Britain in this way. Direct Roman rule certainly ended at about this time. The government at Ravenna was simply incapable of reasserting rule in such a distant province. Even when Constantine was finally defeated there were too many other problems to deal with and its resources were inadequate.2O
A[fler the End
Roman rule in Britain was ended by a rebellion against Constantine III. As far as we can tell it was not a revolt against Rome or the empire itself. For at least the next century the educated inhabitants of the island still seem to have referred to themselves freely as Romans or Britons. In a way it was unusual that the rebels did not proclaim a new emperor. By this time the army left in Britain must have been small, probably no more than skeleton units of limitanei dotted around the frontier outposts. They were neither numerous nor united enough to impose a single ruler, whether as emperor or representative of Honorius' government. No one had the power or money to hold the diocese or even the individual provinces together. New coinage stopped reaching Britain in significant quantities after 402. None of the communities or leaders to emerge in the fifth century minted their own currency. This lack of new coinage makes it much harder to date sites from this period. It does not mean that the economy entirely ceased to be monetary, and money may still have been used for some exchanges for a considerable time. It is an indication, however, that there were no more professional, salaried soldiers. The imperial taxation system also ceased, and gold or grain or other levies no longer had to be gathered and transported on such a vast scale."
Britain broke up into many separate communities. It was not simply a reversion to the old tribes pre-dating Roman rule. Too much time had passed for these to have great meaning and, instead, the administrative states created by the Romans had more significance. Even so, the powers that would emerge did not follow these boundaries very precisely. Instead, new states or kingdoms were created. Most, if not all, were ruled by kings - or tyrants, as Gildas and other sources tend to dub them. They may not have been the only authorities, and some civic leaders seem to have continued to exist, but such warlords were undoubtedly stronger than any other powers to emerge. Central imperial power had gone and in its place anyone capable of controlling enough force, influence and wealth was able to carve out a kingdom.22
A source written in Gaul in the middle of the fifth century talks of Britain being `devastated by a Saxon invasion' in 410. There is no archaeological evidence for this attack, but then the same is true of most barbarian raids on Britain and other parts of the empire. Certainly, settlement by Saxons or other north German peoples in early fifthcentury Britain seems to have been limited to a few small communities in the south-east. These may as easily have been mercenaries brought in by British leaders - or before that by the imperial authorities - as settlers who seized territory by force. The example of Alaric's Goths shows that the same group could easily appear in both guises over the course of just a few years. The attacks in 410 were most likely heavy raids and need not have involved huge numbers of warriors or any attempt at permanent occupation. Some might prefer to date the attacks earlier and associate them with the ones that are supposed to have provoked the rebellion against Constantine. Alternatively, Saxon attacks may have become heavier to exploit the weakness in Britain following the expulsion of the imperial authorities."
Saxon raids posed a problem, especially to those communities in vulnerable areas. The same was true of plundering bands of Picts, Scots and Irish. All were likely to have been quite small-scale, especially when the attackers came by sea. Roman rule in Britain was not ended by outside attacks, nor were the British powers that emerged rapidly overrun by these foreign enemies. There is some sign of the Britons organising to combat their foes, especially on Hadrian's Wall where several forts were reoccupied in the fifth century. Sometimes the evidence of activity is slight, but at Birdoswald a large timber hall was built on the foundations of the Roman granary. Someone also repaired the defences at Housesteads, although in earth rather than stone. At the very least this suggests local war leaders with warbands were based in partially restored former army bases. One scholar would even see this as the sign that a leader emerged able to revive something of the old military command of the Dux Britanniarum, albeit doubtless on a more modest scale .14
Britain's kings and warlords most likely fought each other as often as foreign enemies - the Romans had no monopoly on civil war - and the fragmentation of the provinces into many small kingdoms does not suggest harmony. Like the emperors, it would be surprising if they did not employ barbarians as allies or mercenaries to fight against their neighbours and rivals. For at least a few decades it was British leaders who remained in control throughout the old Roman diocese. No light was switched off, immediately extinguishing all aspects of culture and life from the Roman period. Most cities and towns continued to be occupied, as did many villas. Some substantial buildings were built within the old walls of towns, even if they were invariably of timber construction. Systems to supply water remained in use for most of the fifth century, in at least one case being repaired. Some baths continued to function, but in general these were one of the first things to decay and be abandoned both in cities and at villas. Very soon no one had the skill or wealth to maintain such sophisticated pieces of engineering, let alone build new ones. There were also more mundane changes. It quickly became rare to use pottery that was not made locally, and before long the potters ceased producing wheel-turned pottery."
Some things survived, but that is not to say that the changes were not major and fairly rapid - certainly within a generation - even if they were not instant. Life in Britain became less sophisticated, with few signs of prosperity comparable to the Roman period. The wealthiest were cushioned to some extent, and it was easier for them to leave and settle in Brittany, but their comforts were fewer both there and in Britain itself. Western Britain, notably Wales, Cornwall and Cumbria, had been amongst the least developed parts of the Roman province. Paradoxically, this may have changed in the century or so after Roman rule, with these areas becoming a little more `Roman' and almost certainly more thoroughly Christian. There is no good evidence for a substantial pagan community in Britain in the fifth and sixth centuries before the creation of the Saxon kingdoms.26
Britain was not cut off from all contact with the Roman empires after 410. Trade declined massively, and it was no longer part of the imperial bureaucratic and fiscal systems, but as far as both Romans and Britons were concerned it remained part of the Roman world. The church played a key role in maintaining this connection. Bishop Germanus of Auxerre in Gaul was later canonised and his biographer recorded two visits to Britain, the first in 429 and the second sometime in the next fifteen years. Travel to Britain was evidently still possible and not excessively dangerous. Nevertheless, it is hard to judge how much the biographer really knew of life on the island. Germanus seems to have visited St Albans (Verulamium) and went to the shrine of its famous martyr. In one city he healed the blind daughter of a local dignitary, called a tribune, but whether this was the correct title is questionable. He also rallied the locals to defeat a band of Saxons and Picts - in itself a fairly unlikely combination - teaching his men to raise a shout of `Alleluia!' This is said to have been enough to rout the enemy.27
Yet the main reason for both visits was to combat heretical Christians rather than foreign plunderers. Germanus held debates with priests adhering to a doctrine named Pelagianism after its founder. Pelagius was originally from Britain, although his preaching mainly attracted attention after he moved to Italy in 380. His particular brand of asceticism was moderate by the standard of the day, but his emphasis on the ability of individuals to become virtuous through effort and make themselves acceptable to God was far more controversial. Over time he attracted many prominent critics, including St Augustine, who accused him of effectively denying that salvation depended on grace alone. Pelagius was finally condemned as a heretic in 418. Germanus' biographer claims that the bishop easily confounded the British Pelagians in debate. He also characterises them as boastful and ostentatiously dressed, but this may just be conventional criticism. It is hard to say whether it can be used to show that there were substantial numbers of wealthy aristocrats and priests in the British towns.z8
Older books tend to depict the arrival of the Saxons, Angles, Jutes and other tribes as a massive invasion, which killed or drove off all of the British inhabitants of the south-east. Later these peoples would continue to expand, creating kingdoms and in time merging into the AngloSaxons, speaking their own Germanic language and with their own customs and laws uninfluenced by Roman or British ideas. The descendants of the population of Roman Britain were dubbed the `Welsh' or foreigners and forced into enclaves in Cornwall, Wales and the northwest. Thus was England created."
More recently, ideas about this and other ancient migrations have changed profoundly. Scholars have doubted the scale of any movements, suggesting that the invaders were far outnumbered by the indigenous population. At the same time, the violence of their arrival has often been played down, particularly by emphasising the idea that many arrived as mercenaries. The discovery of graveyards that appear to show Saxon and Briton being buried on the same site has been interpreted as demonstrating that the two groups could and did peacefully coexist. Others would see the spread of Saxon styles - again, largely through grave finds and mainly of metalwork such as brooches and belt buckles - no longer as an indication of the advance of these people. Instead, it has been suggested that the Britons deliberately aped these styles, willingly associating themselves with the Germanic peoples for political reasons.3°
As usual in such cases, the pendulum has swung too far and it is important to look again at the evidence. Saxon finds become markedly more common around the middle decades of the fifth century. Most are in eastern England, and the greatest single group are from burials. Initially, these take the form of cremations, but gradually inhumation becomes normal, with the body usually accompanied by grave goods. In roughly the same period, literary sources speak of a great war beginning when Saxon mercenaries rebelled against their British employers. In one tradition, the king responsible for enlisting the Saxons' services is called Vortigern. The names of the warriors' leaders, the two brothers Hengist and Horsa, mean literally `stallion' and `horse' and may well be later inventions."
The details and precise dates of this conflict are impossible now to reconstruct, but there does seem to have been an increase in the area dominated by Saxon groups at this time. Events elsewhere in the empire have shown that barbarian warbands did not need to be especially large to cause a fundamental shift in the local balance of power. The imperial government rarely had enough soldiers to defeat these groups. Usually this was only possible when they hired another set of barbarians to fight on their behalf. Such leaders as emerged in Britain can only have been massively weaker and so even less capable of dealing with barbarian groups. Given that there was no central authority as powerful as the empire at the start of the fifth century, even very small bands of warriors would have presented a major problem. On the other hand, it does seem that many settlements in northern Germany and Denmark were abandoned in the fifth century. In some areas sea levels rose and previously fertile fields were flooded or turned into salt marshes. A significant migration to better land in Britain is perfectly possible."
Mixed cemeteries apparently containing both Saxon and British burials are not straightforward to interpret. In the first place, considerable caution needs to be used before assuming that a particular object automatically denotes someone of a particular race. Brooches were both functional and valuable. They would not be discarded or remade simply because the design was not traditional to the owner's culture. In the end, brooches and belt buckles were there to hold up clothes more than to express identity. Any such item could as easily have been acquired through violent acquisition as peaceful trade. It is not impossible that the mixed cemeteries do indicate peaceful coexistence of two races within the same community. This does not mean that both sides lived willingly in this way. Many repressive regimes would not necessarily reveal themselves in the burial record. Britons apparently able to bury their dead according to their own customs and in the same general area as Saxons does not necessarily mean that they were not a more or less subject race.
There were significant numbers of Saxon war leaders in eastern and southern Britain by the second half of the fifth century, and they were strong and powerful. This was simply a reality that no one could afford to ignore, least of all the British leaders and communities closest to them. It is certainly quite plausible that some decided their best advantage lay in joining the new arrivals, hoping to benefit from their power. Plenty of British nobles had done much the same when Claudius invaded in 43. Some Britons may have tried to `become' Saxons, just as some of their ancestors had once been keen to `become' Romans. In neither case was this spontaneous, but simply a response to the arrival of a new power, which it seemed unwise or impossible to oppose.
One great difference was that the Saxons were no more united than the Britons. Apart from the term embracing groups from a range of different peoples, the Saxons themselves appear as disunited as other tribal groupings. It was not simply a question of allying with the Saxons, but finding a way to placate or defend against each of their war leaders within striking distance. Raiding is likely to have continued to be a normal part of life for the invaders. There is no particular reason to believe that rivalry and fighting between Briton and Briton or Saxon and Saxon ceased. There were also still other enemies, such as the Picts and Scots. Parts of western Britain seem to have been permanently occupied and settled by war leaders from Ireland. Nor was the conflict all in one direction. St Patrick wrote to the British King Coroticus condemning him for allowing his warriors to raid and take slaves from amongst Christian converts in Ireland. We also hear in 469 of a British war leader named Riothamus who had taken his band of warriors over to Gaul and become a local power. We do not know whether he had been forced to leave or simply scented better opportunities for profit and employment on the Continent.33
Given the thinness and questionable reliability of our sources, we cannot chart the wars of the fifth and sixth century in any detail. Yet the overall pattern was one of gradual expansion by the Germanic tribes. The Britons who migrated to Brittany or the western parts of Britain itself must have been fleeing from something. Doubtless conflict was not unceasing and there were periods of general peace as well as longer and more sustained local lulls. Perhaps the Britons did win some great victories as Gildas and others claimed, but in the sixth century Saxon power waxed stronger and began to shape Anglo-Saxon England. Not all Britons will have fled or died, but the survivors were absorbed by their conquerors. Celtic language was replaced by Saxon in a large part of the island and Latin for the moment was virtually or wholly abandoned. Until St Augustine of Canterbury's mission in 597, the Saxon kingdoms were all pagan, although it is impossible to know whether pockets of Christian belief remained within them. By this time the Western Empire was only a distant memory, but the Catholic Church preserved some of its international connections.34
Roman Britain had not fallen to outside pressure. The leaders who threw out Constantine III's governors quickly started to squabble amongst themselves. They were Romans as much as Britons and the rebellion was still essentially a civil war. Its result was to create many tyrants or kings instead of one imperial usurper. Constantine himself was too busy fighting for his own survival to try to regain control of Britain. Honorius and his successors lacked the power to do so. There were foreign enemies, and gradually some of these overran much of the island, but it is worth remembering how long this took. Everything, including power, trade and warfare, became far more local than had been true under the Romans. Some things survived, especially in the areas longest controlled by the Britons themselves. Christianity was one of the most important and mainly responsible for at least some continuation of literacy, both in Latin and eventually also in the Celtic languages. Debate continues to rage over how much, if any, continuity there was between Roman Britain and Anglo-Saxon England. Some towns were important to both, although whether occupation was unbroken is less easy to say. The evidence in favour of this is not good, but even the continuities that can be traced should never obscure the massive scale of the change. It would be five hundred years before even the greater part of the old Roman diocese would again be united under a single authority. The better part of seven centuries would pass before Norman cathedrals would match in scale the basilicas of Roman cities or the headquarters buildings of legionary fortresses.35
There is no firm evidence that Arthur actually existed and the mentions of him are all comparatively late. On the other hand, there is no reason that he could not have existed, and it is hard for anyone - perhaps especially anyone British - not to wish that he did. The fifth and sixth centuries were clearly years of frequent conflict, a time that produced many warriors and warlords. That one of them was especially successful and charismatic is plausible enough, although equally the later stories may be an amalgam of the deeds of many men generously fleshed out with myth. There is much more besides the reality or otherwise of Arthur about the history and society of Britain after the end of the Roman province that we simply do not know. In the long run there were probably fewer traces of the Roman presence left in Britain than almost any of the western provinces. Yet on the whole its experience may well have had more things in common with the wider experience, at least in the fifth century. It is now time to look at the final years of the Western Empire.36