TWO

KING ARTHUR

Then Arthur fought against them in those days with the kings of the Britons, and it was he who led their battles.

Nennius History of the Britons

THERE IS NO Myth in British History, and few in the world, to match the story of King Arthur: the knights of the Round Table, Guinevere, Lancelot, the quest for the Holy Grail. Since the Middle Ages these tales have exerted their fascination and continue to do so today, particularly in a declining Britain where the myth of a golden age has obvious attractions. But beneath the legend lie the real events of the fifth century; the period of the fall of the Roman Empire. This time remains the darkest, the least documented, in British history. Yet through the shadows modern historians think they have distinguished a remote war leader, a real-life British hero who is said to have died around 500, and whose fame, it is thought, grew from successful battles against the Anglo-Saxon invaders who poured into Britain from northern Europe and Scandinavia after the fall of Rome, the ancestors of today’s English. But are the historians right?

In spite of their obscurity the years following 500 were some of the most important in the 2000 years of recorded British history. It was then that the key racial and linguistic alignments of Britain were defined. The Celtic inhabitants of Britain were driven into Cornwall, Wales, Strathclyde and Scotland by the Anglo-Saxon newcomers who settled in the east and south in what was later to be called England. The dispossession of the Celtic people of Britain by the Anglo-Saxon invaders was the subject of a huge literature in the Middle Ages, and in Arthur it had its greatest hero.

The purpose of this chapter is to look at the way in which Roman civilisation in Britain came to an end, and to see what justification there is to link a historical Arthur to these events. Go into any bookshop today and you will find yourself in no doubt that he existed. A mass of Arthurian books is available, from reputable academic works to theories from the lunatic fringe. In them Arthur emerges as everything from cult hero to guerrilla generalissimo, Dark Age Superman to Dark Age Che Guevara. They remind us that every age makes of Arthur what it will. In the eighteenth century Gibbon contented himself with noting that ‘the severity of the present age is inclined to question the existence of Arthur’ and wisely avoided conjecturing a career for him. Macaulay in the early nineteenth century thought him ‘no more worthy of belief than Hercules’. It has been the twentieth century which has sought to corroborate the details of the myth and find a historical Arthur. The Victorians perhaps paved the way for this. Books like Tennyson’s Idylls of the King, which were based on the legend, had a fantastic emotional impact on the late Victorians, with its stress on chivalry, heroism, and nationalism, and its dark strain of sexual jealousy and betrayal. All this appealed to the Victorians’ nostalgia for a lost golden age.

But Victorian England was also the era of the birth of scientific archaeology. Remarkable discoveries were made in Greece, Crete, and Turkey by the German, Heinrich Schliemann, which were thought to have proved that the Homeric legends were ‘true’, that Agamemnon and Achilles had existed and that Troy was sacked by the Greeks just as had been described in Homer’s poems. We now know that Schliemann was wrong in most of his conjectures about Troy and Mycenae, but it is perhaps not surprising that British scholars of the early part of this century, brought up in this intellectual milieu, should have been the first to try to reconcile the legend of Arthur with historical fact, to work out a rational account of the Arthurian story and to suggest that behind the tale of Arthur and his knights lay a historical armoured cavalry leader with a brief to drive back the Anglo-Saxon hordes as Roman power in Britain ebbed.

As at Mycenae and Troy, finds of the right period have been pressed into service. Since the Second World War historians have not only gone back to Glastonbury and Cadbury and ‘found’ Avalon and Camelot, but have built up a detailed history of Arthur’s empire and its political ideologies. We shall look at the sources for these conjectures later. Let us simply note now that it is a natural impulse for societies to construct a golden age retrospectively, and for the most part such speculations are no more than that. An example of this was the discovery of ‘Arthur’s tomb’ at Glastonbury. In the mid twelfth century Geoffrey of Monmouth’s fanciful book, History of the Britons, fired the imagination of the credulous intelligentsia of the time. In 1191, six years after the destruction of their abbey by fire, and with their restoration fund badly needing a boost, the monks of Glastonbury dug secretly in their old cemetery and ‘discovered’ the remains of a large man buried in a tree trunk. This, they claimed, was King Arthur himself, and they produced a forged inscription to prove it. Eyewitnesses had not mentioned a woman, but it was soon said that Guinevere had also been found. Given the medieval propensity to manufacture relics, no disinterested observer would think this find had any more relation to King Arthur than the subsequent discoveries in the same place of the sword Excalibur and the grave of Joseph of Arimathea. It is interesting to note that the great twelfth-century expert on Glastonbury, William of Malmesbury, who wrote an official history of the monastery just before Geoffrey of Monmouth wrote his book, specifically denies that the burial place of Arthur was known. Nevertheless the 1191 discovery has been accepted as the grave of King Arthur by a number of scholars in recent years.

There is even more doubt about the Arthurian associations with such tourist spots as Tintagel, Amesbury, Winchester and Cadbury. In the main they are the inventions of twelfth-century romantic poets. If we wish to uncover real events behind the stories we must do as William of Malmesbury enjoined us all those years ago: ‘Throw out such dubious stuff and gird ourselves for a factual narrative.’ The question is, what happened in Britain at the fall of the Roman empire? Bound up with that, there is a second question: did there exist at this time a war leader called Arthur? In much of this chapter we will be examining the historical background to the late fifth century. It is only when we understand the nature of that time that we will be in a position to evaluate the evidence for Arthur’s existence.

A WORLD IN DECLINE?

Portchester on the south coast of England. This massive Roman fortress, the best preserved in northern Europe, was one of a dozen built around the year 300 to defend Britain against the incursions of Anglo-Saxon invaders. The society it was to protect was already under stress, with class divisions, a decline in trade and depopulation of towns, a falling birthrate, high inflation, the gradual collapse of a money economy, and the growth of great private rural estates.

In 410 came the end of 350 years of Roman colonial rule; a period as long as that in which the Portuguese ruled over Angola, longer than the British supremacy in India. The Romans did not simply abandon England and sail back to Italy. Their armies had been gradually withdrawing in the preceding years, and before that there were several periods when the Britons had elected their own emperors and cut their links with the Roman government on the continent. What happened in 410 was a formal severance of responsibility for defence. But Britain had been a Roman province for so long that the Roman influence remained long after the Roman departure. The problem of how long Romanitas continued here is one of the cruxes of early English history and archaeology. Historians now recognise that the western provinces of the Roman empire remained a recognisably sub-Roman civilisation for centuries under the barbarian Germanic tribes who took over.

The first few years of the fifth century were critical for the western Roman empire, threatened on all sides by barbarian peoples breaking through the frontiers. One barbarian chief, Alaric, King of the Visigoths, appeared in Italy in 401 and besieged Rome in 408, finally entering it in 410. Other Germanic peoples, the Vandals, Suevi and Burgundians, struck into Gaul in 407. It was to meet this challenge that the Romans were forced to withdraw troops from Britain. The Britons themselves were under increasing threat from Scottish, Pictish and Anglo-Saxon raiders, and with no help forthcoming from Rome, elected their own leaders, one of whom, Constantine, held power from 407 to 411 and led troops into Gaul against the Roman government there. According to the Gallic Chronicle, Britain, deprived of fighting men, then suffered a large-scale invasion in 408. ‘The provinces of Britain laid waste by the Saxons: in Gaul the barbarians prevailed and Roman power diminished.’ What happened next is the subject of argument. The general view is that though their leader Constantine was fighting in Gaul and called himself an emperor, the Britons did not consider that they had ceased to be part of the Roman Empire. In 410 they appealed for help to Emperor Honorius in Rome but he could do nothing except order the local communities to arrange for their own defence. Rome had troubles of its own: in the same year it was sacked by the Visigoth Alaric.

A different perspective can be seen in the works of the late fifth-century Byzantine writer Zosimus, which explains how the Britons were able to resist the Anglo-Saxon invaders successfully. According to Zosimus it was the Saxon ravages (presumably culminating in the 408 attack) which forced the inhabitants of Britain to secede from the Roman Empire in a kind of UDI, and become once again ‘another world’. They organised their own defence, took up arms, and ‘braving every danger freed their cities from the invading barbarians’.

Some historians have suggested that this revolt went hand in hand with a social revolution, a peasant revolt like the Bacaudae in Gaul, and that the peasantry successfully opposed the Romano-British upper class and defeated the Anglo-Saxons. The appeal to Honorius in Rome would then have been a last plea from the threatened landed class. It has even been proposed that this revolution went hand in hand with a radical religious movement, Pelagianism, a new kind of puritanism born out of a time of stress. However, it is now thought that this theory is unlikely to be true and that there was no movement for social reform. In fact it is probable that the people who reorganised Britain’s defence in the early fifth century were the Romanised urban upper classes, the curiales, in the areas where town life still functioned.

Two contemporary sources throw some light on this dark period of British history: a biography of St Germanus and St Patrick’s Confessions. In 429 St Germanus came to the island as an agent of the Catholic Church in Rome to combat the spread of the heretical Pelagianism, and though there were serious incursions in the south by Saxon and Pictish pirates, according to the biography, organised Roman town life still continued. Local magistrates were still in charge in the cities and there was obviously nothing unusual in a bishop from the continent travelling through the province to correct ecclesiastical observance. By the time of St Germanus’ second visit in 447 the island was still holding out against the Saxons, if the author of the saint’s biography is to be believed: writing in the 480s (which was precisely the supposed period of the Arthurian wars) he speaks of Britain as essentially Roman in administration and orthodox in worship, and most remarkably, ‘a very wealthy island’.

ST PATRICK

The most interesting source for this period is the Confessions of St Patrick. Patrick, the patron saint of the Irish, was in fact a mainland Roman Briton taken into slavery in Ireland at the age of sixteen in one of the raids around 400; one of thousands seized by pirates. His father owned a small villa in the west (perhaps in the region of Carlisle), was a local town councillor, and a church deacon. Escaping from captivity in Ireland the young Patrick took passage on a trading ship to Gaul, which was then being devastated by far-reaching barbarian raids. The key point about Patrick’s narrative is that when he returned to Britain in c. 415, five years after the Romans had left Britain, there is no suggestion of anarchy, and when he wrote his account in the middle years of the century the imperial Roman system of local government was intact. The local town councils, for example, were still responsible for raising taxes for the government. It was still a world where professional rhetoricians could earn a living as they could in Rome; a world where a letter writer could address the British dynasty of Strathclyde as ‘fellow citizens’. We can therefore assume the continuation of a feeling of identity with Rome in the Romano-British ruling class, the senatorial aristocracy and the local landowners. This is the kind of background we would expect for an historical Arthur.

VORTIGERN

When did Britain fall? Several sources indicate that the crucial breach took place in a period ten years on either side of 450. The Gallic Chronicle says the island fell under Saxon domination in 441 or 442. According to the Anglo-Saxon historian Bede the ‘coming of the Saxons’ took place in 448 or 449. This last date was derived from the most important source for the fifth century, the British cleric Gildas. Gildas wrote his book, On the Ruin and Conquest of Britain, in the 540s and his account is generally interpreted as meaning that in 446, 36 years after they had applied for help to fight the Anglo-Saxon invaders, the British government made an appeal to the Roman consul, Aetius, for military aid against Pictish and Scottish invasions. When this aid was not forthcoming, Anglo-Saxons were introduced as friendly mercenaries, and this was, according to Gildas, the fatal step which led to the collapse of British rule in the east of the island. Most important, it was not self-governing cities or local oligarchies who brought in these troops, but a dictator. According to all later sources, by the 430s a large part of Britain had fallen under the sway of a British leader called Vortigern (‘Great King’), and according to British and Anglo-Saxon tradition it was he who invited increasingly large numbers of Anglo-Saxon mercenaries from Germany and Denmark to fight for him. What little we know of Vortigern shows that for a time he had something approaching absolute power. He was even able to arrange the migration of a whole people, the Votadini under their leader Cunedda, who were forced to leave Lothian and made to settle in North Wales, in order to resist Irish incursions.

Like many dictatorships, the new order of Vortigern was strong, unscrupulous and efficient. Although Gildas was writing in the 540s he must have spoken to men who remembered Vortigern’s rule. He says that this was a time of prosperity and that the people were successful against the attacks from the Picts and Scots. This view, though, may only reflect the experience of the ruling party, because, as in many periods of decolonisation, there were many rival factions. These came to a head in civil warfare which coincided with two shattering blows. In 443 the whole Roman world was swept by a plague, the severity of which has been compared with the Black Death, and which must have hit Britain around 446. At the same time the Anglo-Saxon mercenaries settled by Vortigern in Kent, led by Hengist and Horsa, revolted.

THE SAXON REVOLT

Until recently historians believed that the Romans had been employing Germanic mercenaries in Britain for over fifty years before they departed in 410, and that many of the mercenaries had settled around the cities, were used to Roman life in towns, and had perhaps married Romano-British women. This familiarity with the Roman towns, it was argued, facilitated the changeover to Anglo-Saxon rule in the eastern parts of Britain. Recently, however, the archaeological evidence for this picture, namely the date of Anglo-Saxon military cemeteries which had been assigned to the late fourth century, has been seriously challenged. At present it seems best to follow the traditional scenario for the coming of the Anglo-Saxons as outlined by Gildas and Bede. The wars which precipitated the fall of Britain began as a struggle between various Romano-British parties, some or all of whom may have hired Anglo-Saxon mercenaries. Our sources for instance suggest that Vortigern had opponents who had purely Roman names; one, Ambrosius Aurelianus, is said by Gildas to have been born into a family who had been emperors in Britain; another opponent, Vitalinos, is recorded fighting a battle against Ambrosius. The break-up of Britain in the fifth century, then, reminds us of twentieth-century liberation wars fought, for example, in former Portuguese or Belgian colonies in Africa, with several opposed factions fighting each other, with perhaps Vortigern’s ‘British Patriotic Front’ fighting against Ambrosius’ Roman party which opposed his nationalist dictatorship. Both sides may have employed Anglo-Saxon mercenaries.

In the middle of this internecine strife, Britain was further disabled by far-reaching Germanic raids described by Gildas in graphic terms. When supplies and money for payment ran out, Vortigern’s federates devastated with fire ‘all the neighbouring cities and lands … until it burnt nearly the whole surface of the island, and licked the western Ocean with its red and savage tongue’. Some Britons were enslaved by the Anglo-Saxons, others fled overseas, others retreated to forests, offshore islands, and most of all ‘to the high hills, steep and fortified’, the old Iron Age hillforts. This was not an organised campaign of conquest, but a violent raid, for Gildas describes the Anglo-Saxon mercenary armies retiring to the lands they had been given by Vortigern in the east, presumably Kent and East Anglia. Gildas’ impression of continuous, destructive raids may be exaggerated. Rather, the atmosphere may have been like a Dark Age gold rush of impoverished immigrants from the then ‘third world’, the underdeveloped lands of the Germanic north, into the rich agricultural provinces of the Western Empire. These new ‘barbarian’ settlers – though initially invited – found themselves militarily strong and politically and socially unabsorbable. Despite Gildas’ dramatic tale of devastation, they may not have been the only, or even the chief threat to internal security in late fifth-century Britain. But for thirty years between the 460s and the 490s they provided the ideal enemy in a prolonged war organised by the surviving senatorial aristocracy of Roman Britain. Nothing provides a better stimulus to preserving one’s identity as a ruling class than an external foe, especially one comprising ‘uncivilised’ immigrants. According to later traditions about this ‘patriotic war’, the army leader at the climax of the struggle was Arthur.

Gildas, however, is our only reliable source for these events. With Vortigern dead, the British organised resistance against the invaders. Under the leadership of Ambrosius they fought a number of successful battles culminating in a great victory in the 490s, at a place called Badon Hill. This battle, says Gildas, gave forty years of peace to Britain, though, as he wrote in the 530s or 540s, ‘not even at the present day are the cities of our country inhabited as formerly; deserted and dismantled they lie neglected until now, because although wars with foreigners have ceased, domestic wars continue’. Gildas does not name the British leader at Badon; as we shall see, it is considerably later traditions which insist that he was Arthur.

WROXETER: ‘THIS IS HOW IT ENDS: NOT WITH A BANG BUT A WHIMPER’

Archaeologists have been able to corroborate Gildas’ picture. Although the Roman cities were not deserted everywhere, some were abandoned at this time, or their populations shrank dramatically. In Cirencester, for example, the second city of Roman Britain, archaeologists have established that civic life continued into the 440s; the defences were repaired, flood prevention work carried out at one of the gates, and the piazza of the forum kept clean. But soon after that time, whether caused by the great plague or by the Saxon revolt, unburied bodies were found in the streets and the town seems to have contracted to a few wooden huts inside the amphitheatre.

The most vivid picture we yet possess of declining late Roman city life comes from Wroxeter near Shrewsbury. Unlike most Roman towns, Wroxeter did not become a modern city; it still lies under farmland, and is now being painstakingly uncovered. The present excavation is around the basilica of the baths complex, formerly a great brick hall the size of a cathedral nave. This centrepiece of Roman civic pride fell into disuse around 350, and was demolished to be succeeded by shanties.

To the great surprise of the excavators, however, a later phase has been discovered which shows that the area was rebuilt. The basilica area was levelled, covered with thousands of tons of carefully laid rubble, and on this base a large number of timber buildings were erected including a massive wooden hall laid on beams, 125 feet long and 52 feet wide with a narrow extension 80 feet long. This hall, with its porticoed façade, wings and steps, was the central structure of a complex of related timber buildings. South of it were rows of timber booths separated by a finely sifted gravel street roofed like a pedestrian precinct. At the upper end of the street was a series of large wooden buildings with classical façades, ‘perhaps the last classically inspired buildings in Britain until Wren and the eighteenth-century revival’, as the excavator has called them.

Who can have been the initiator of this drastic reorganisation of a whole city centre? It needed wealth, a high degree of organisation, and strong motivation. It was certainly not the work of demoralised peasant villagers, nor was it effected by Irish or Anglo-Saxon invaders. It has the hallmarks of Roman public works, only constructed with timber: we must surely be looking here at a complex of religious or public buildings or the private domain of some great man.

The end of this phase, the last occupation of the main area of the city, is equally intriguing. These halls were not sacked or hurriedly abandoned. They were deliberately dismantled and all useful materials taken away. When? The excavators are not sure, though a date towards the end of the fifth century is the present thinking. Why? This may be easier. Wroxeter is a large town, 200 acres with two miles of walls, and thus difficult to defend without a large fighting force. The likelihood is that the city was abandoned for a more defensible site. And if the princes of Powys had Wroxeter as their main centre up till around 500, could the city have been the base of Vortigern, who appears in the genealogies of Powys? Or could it possibly have been Arthur’s base? We shall probably never know, but this massive injection of energy, capital and manpower into what was evidently a declining town suggests the influence of one of the powerful leaders struggling for control in sub-Roman Britain, a man who wished to restore something of the grandeur of Rome, albeit in timber.

‘TRUSTING THEIR LIVES TO THE HILLS … PRECIPITOUS AND FORTIFIED’

As the cities declined, many of the warlords went back to the hills, renovating the Celtic Iron Age hillforts which had been abandoned when the Romans first conquered Britain. These citadels were easier to defend than long and badly maintained city walls. They suggest an atmosphere of retreat and fear, such as Gildas describes (‘terrified by the wolfish villains’). Stand inside them and images are evoked of refugee compounds, robber barons, warlords surrounded by their armed followings, private armies. This is the background modern historians have seen as Arthur’s, in the period when Anglo-Saxon mercenaries settled in Kent, Sussex, East Anglia and the Thames valley and spread their incursions deeper into southern and western Britain.

A late local tradition connected Arthur with one of these hillforts, South Cadbury in Somerset, ‘that is Camelotte’, and when the Camelot Research Committee dug there between 1966 and 1972, they caused a sensation. On top of the fort they found the 18-acre area had been refortified with a drystone wall, inside which had been timber buildings including the feasting hall of a Dark Age warlord. Elements of the refortification strongly recalled Roman military architecture; imported pottery from the Mediterranean gave a hint of aristocratic luxury and showed the buildings were occupied in the last quarter of the fifth century – precisely the time at which Arthur is supposed to have flourished.

Although the name Cadbury-Camelot has stuck, the excavators did not, in fact, find Camelot, for that name is the invention of a French poet who wrote in the twelfth century and is therefore of no value to the historian save as a symbol. Nor was anything turned up to connect the place specifically with King Arthur. What the dig did prove was that in the later fifth century (c. 470–500) someone was powerful enough to wall this hillfort, erect buildings and build gates; someone whose retinue was large enough to need such an extensive site; someone who built in a hybrid Roman-British style.

At the time it was thought Cadbury was exceptional and must have been the fortress of a particularly great leader. Now archaeologists know it was not unique, for many other Iron Age hillforts were refortified at this time in the south-west and elsewhere. Indeed such reoccupation seems to have been the rule rather than the exception: over a dozen instances have been found in Somerset alone, and forty in the south-west as a whole. Many others await investigation, such as the fort hidden under woods at Amesbury in Wiltshire, a particularly interesting early Christian site which etymologists connect with Ambrosius Aurelianus and which was taken from the Britons by the West-Saxon kings relatively early in Anglo-Saxon times.

The significance of these hillforts is not yet clear. For instance, do they represent centralised control of a dictator like Vortigern, a generalissimo like Ambrosius, or are they local defences as Gildas implies? Tempting as it is to associate such impressive works with tyrants and their military élites, we cannot be certain that the men who rebuilt them commanded more than local allegiance; we do not even know whether such men were rulers at all (the forts could have been organised by confederacies of local peasants, farmers or aristocrats). But dating at a number of sites now seems to link them with the specific situation of the war with the Anglo-Saxons in the period c. 470–500, which is exactly the period when Arthur is thought to have lived. Evidence from one site examined in detail, Cadbury-Congresbury in Somerset, shows that it was reoccupied in the mid to late fifth century, but within half a century the timber and stone ramparts had begun to collapse and the ditch had filled up with silt and stones almost to the top. Only then was the imported pottery arriving at the site: a hint that the temporary crisis had passed, the new phase defences were no longer needed, and imported goods were now coming in: calm after the storm?

THE SIEGE OF BADON HILL

The storm in question was the series of battles between the Britons and the Anglo-Saxons mentioned by Gildas which took place in the last quarter of the fifth century. The war culminated in the siege of Badon Hill – Mons Badonicus – which took place perhaps a little before 500. According to Gildas, who wrote his account 43 years after the battle, this was the ‘last great victory of the fatherland’. As we have seen he tells how, nearly a century after the Roman departure, Romano-British armies led by Ambrosius Aurelianus beat back the Anglo-Saxons and won a peace which lasted up to the time that Gildas was writing. His story is confirmed by archaeology: the British recovery after Badon is recognised in the lack of sixth-century Anglo-Saxon pottery in such areas as Sussex, Essex and Hertfordshire which have all yielded fifth-century material; similarly, judging by modern finds of their grave goods, the Anglo-Saxon expansion in the Upper Thames ceased for fifty years after Badon. Badon is surrounded by controversy. Gildas names no leader of the Britons for this battle, nor do we know the leader of the Saxons, though modern writers conjecture a joint force from Kent, Sussex and Wessex under the South Saxon Aelle, who Bede and the Anglo-Saxon Chronicle say was overking of the settlers at this time.

The story would be clearer if we knew where Badon Hill was. Gildas implies that the battle was in the south-west and the fact that it was a siege of a hill – a mons – strongly suggests that this was another reoccupied hillfort. Several sixth-century battles took place at hillforts: Old Sarum and Barbury Castle in Wiltshire, Dyrham Camp in Gloucestershire. We do not even know whether the Britons were the defenders or the besiegers at Badon. There are several possibilities for the battle site, but the best would seem to be Liddington Castle, a prominent Iron Age hillfort near Swindon in Wiltshire. Next to it is a village called Badbury which philologists say could have come from a Celtic Badon. It lies in a central position between the main Anglo-Saxon settlements as they stood in the year 500, the Romano-British areas controlled by the cities of Gloucester, Cirencester and Bath and the zone of reoccupied Iron Age forts which stretches through Gloucestershire, Wiltshire and Somerset. On a good day Liddington can be seen from Cirencester, fifteen miles away, its scarped western side standing out clearly: a separate, isolated hill. Most important, Liddington marks one of the great Dark Age road junctions, at the intersection of Ermine Street, another Roman road going due south, and the Great Ridgeway, which cuts across central England and which runs right underneath the ramparts of the castle. In the early medieval period these roads were still in use, and therefore Liddington was a key site. Also, an excavation on the fort has revealed reoccupation and refortification, and pottery which was imported at the time the battle took place. It seems likely that this was the site of the battle which saved Romano-British life in the Cotswolds and the south-west for fifty years, but our knowledge may be increased and it would be unwise to build too much on a conjectural identification.

There is no disputing the historicity of Badon. Gildas places it securely at this time. But what justification is there for accepting the accounts of Arthur’s leadership at the battle?

THE TWELFTH BATTLE WAS ON BADON HILL WHERE 960 MEN FELL IN ONE DAY AT A SINGLE ONSET OF ARTHUR; AND NO ONE KILLED THEM BUT HE ALONE, AND IN ALL THE BATTLES HE CAME OUT VICTORIOUS.

That quotation appears in one of the most famous passages in British historiography, the tale of the twelve battles of Arthur and his leadership at Badon, and it is found in Nennius’ History of the Britons.

We must now examine the sources of the Arthurian myth, sources which many historians believe to contain a hard core of truth which proves the existence of Arthur. There are two key texts: the Annals of Wales, and Nennius’ History of the Britons, and both are thought to contain authentic survivals from fifth- and sixth-century history. On these two books the evidence for a historical Arthur rests, and because they are so important we must look at them as they appear in the composite Welsh historical manuscript, Harleian 3859 in the British Library. The book is a miscellaneous collection which includes the Annals of Wales and the History of the Britons. Though now in the same book, these are two quite distinct items. The book was written in Britain in the early 1100s, that is 600 years after the events it describes.

The annals and history are very different in origin. The material in the annals dates from long before the 1100s. The last entry, dated 954, is followed by the family trees of the South Welsh kings of the tenth century, so it is likely that our text is at one or more removes from a document compiled soon after 954. The content of the annals, however, goes back to the mid fifth century. There are two famous entries, one relating Arthur’s death:

(490–516). Battle of Badon in which Arthur carried the cross of our Lord Jesus Christ on his shoulders for three days and nights and the Britons were victorious.

(c.511–537). The fight at Camlann in which Arthur and Medraut were killed.

However, these annals were only kept as a contemporary record from round the year 800, and the spelling of the names in the earlier sections indicates that they were originally written only in the eighth or ninth century. This warns the historian that the annals may be unreliable as evidence for reconstructing fifth-century history; they are not a contemporary record. The monk who wrote the Badon annal did not receive the news from a messenger hotfoot from Arthur’s campaign HQ: it is a later scholar’s reconstruction, and that means we have no way of knowing what was originally written down here. Moreover, most of the entries in this section of the annals are short and laconic in the extreme, which suggests that there is a prima facie case for viewing as late additions the apparently mythic details of the ‘three days and nights’ and the reference to the cross. The eminent Arthurian scholar, Thomas Jones, considered that originally the annal read simply: ‘Battle of Badon in which the Britons were victorious.’ We will discuss the significance of the annal referring to the death of Arthur later.

NENNIUS: ‘I MADE A HEAP OF ALL I COULD FIND’

We can see why the legendary details should have been added to the account of Badon, some time after 800, by turning to Nennius’ History of the Britons. This contains the most famous piece of Arthuriana of all, but as a source the history is even less reliable than the Annals of Wales. Written around 830, that is over 300 years after Arthur’s supposed death, the history survives in manuscripts written in the tenth century and later. By the ninth century Arthur had become a folk hero, and Nennius credits him with many miraculous deeds. There is no evidence that Nennius had a reliable source for the events of the fifth century. For instance he gives us the famous list of the twelve battles of Arthur, a list of strange and obscure names and, at first sight, curiously circumstantial detail.

Then Arthur fought against them (that is, the Anglo-Saxons) in those days with the kings of the Britons, but he himself was the leader of battles. The first battle was at the mouth of the river Glein. The second, third, fourth and fifth on another river called Dubglas in the district of Linnuis. The sixth battle on the river Bassas. The seventh battle was in the Caledonian forest, that is, Cat Coit Celidon. The eighth battle was in Fort Guinnion in which Arthur carried the image of St Mary, forever virgin, on his shoulders and that day the pagans were turned to flight and a great slaughter was made on them through the virtue of our Lord Jesus Christ and through the virtue his mother St Mary the Virgin. The ninth battle took place in the City of the Legion. The tenth battle he fought on the shore of the river which is called Tribruit. The eleventh battle took place on the mountain called Agned. The twelfth battle was on Badon Hill, in which nine hundred and sixty men fell in one day from one attack by Arthur, and no one killed (or overthrew) them but himself alone. And in all the battles he was the victor.

The list has been understood by some as a straight record of fifth-century wars. On inspection this proves not to be the case. It is generally agreed that the list has been taken from a Welsh battle poem of a kind fairly common in early Welsh literature. These poems have a tendency to ascribe to their heroes battles which they never fought in order to enhance their glory, and most scholars agree that even if Arthur did exist, he cannot have fought in all the battles that Nennius refers to. In two of them he certainly did not, and as we have seen there is no good evidence that he fought at Badon Hill.

Another objection is that although Gildas wrote in living memory of the battle, and although he mentions Ambrosius as the leader of the British resistance, he does not mention Arthur, or any leader, at Badon. Most suspicious of all are Nennius’ romantic details, the figure of 960 men (probably a Welsh poetic construction, ‘three three-hundreds and three score’) and the assertion that ‘no one killed them but he alone’. This is not an account which can be squared either with history or with common sense, and it seems most likely that Nennius gave the glory of Badon to Arthur because the leader at that battle was unknown. But if the Annals of Wales and the History of the Britons cannot be accepted as primary sources for the fifth century, do they offer us any clues as to where the story originated? For instance, where do the names in the battle list come from?

THE MEN OF THE NORTH

Whoever fought these battles, their names and the other early poetic references to Arthur (c. 900) surprisingly do not take us to the southwest or to Wales, but to Cumbria, southern Scotland, and the ancient kingdom of Rheged around the Solway. It would be fruitless even to attempt to identify most of the battle names, but one, Cat Coit Celidon, the battle of the Caledonian forest, is unequivocally northern and is usually taken to refer to the wooded country north of Carlisle. This could suggest that the poet’s milieu, and the background to the Arthur story might have been in this area. Other names bear this out. For instance the battleground named as Mount Agned in the British Museum manuscript has a second name, Bregomion or Breguoin, in a Vatican version which philologists have identified with Bremenium, the Roman fort at High Rochester in the Cheviots. A battle was indeed fought there (as we know from other sources) by Urien of Rheged in the later sixth century. The site is on the Roman road called Dere Street which runs south from Edinburgh, an ideal place for a clash between the warring tribes of southern Scotland, on the borders of the British tribes of the Votadini and of Rheged. Its gates still standing today to the height of a man, its artillery platforms overgrown, this windswept fort was abandoned by the Romans in the later fourth century. Urien’s battle there was known in northern poetry in the Dark Ages, but it was not fought by Arthur. It happened fifty years after his time when the Anglo-Saxons were penetrating the Cheviots. In the last quarter of the fifth century, the period which we associate with Arthur, they were not in this region.

‘The first battle was at the mouth of the river Glein.’ There is a river Glen in Northumberland, and here an Anglo-Saxon royal hall has been discovered with its associated buildings, including a fort-like enclosure taken over from the British predecessors on the site. Could this famous place, later chief residence of the Northumbrian royal family, have been Nennius’ Glein? It is not impossible that an early battle was fought here at a major Celtic site conquered by the Anglo-Saxons. But, again, a late fifth-century battle with the Anglo-Saxons is out of the question. They were not here until the mid sixth century.

Other speculations can be made about northern battles in the list, but cannot be proved. In the end, as with all the fictions in Nennius’ list, a case can be made to support almost any identification, but all evaporate on close inspection. All we can say is that they are not the battles of a fifth-century leader fighting the Anglo-Saxons. A gloomy conclusion, perhaps, but there does seem a case for thinking that the battles which form the background to the Arthur story were part of the internecine warfare of the northern British border tribes.

What, then, lies at the root of the stories? Could there even have been an early leader in the north-west whose local fame spread wide? If there were indeed battles in this region which were transformed into the list of battles by a later poet, in what social and political context did they take place?

The main town of the border region in Roman times was Carlisle, and local redevelopment is now giving archaeologists the chance to explore a five-acre site within the old city. A late developer among Roman towns, Carlisle was probably raised to the status of one of Britain’s five provincial capitals in 369, and in the nineteenth century plentiful evidence was recovered to indicate that it had a rich urban life in the late Roman period. Columns, sculpture, coins, stone buildings and temples, and numerous inscriptions can be seen in Carlisle Museum. Its walls enclosed 70 acres, and urban life continued well after 400. After the Romans left Britain, Roman buildings were substantially rebuilt in timber, roads were maintained, and the aqueduct remained in use as late as 685. In the twelfth century William of Malmesbury mentioned an arched building which was still standing in his day and which had on it an inscription to Mars and Victory. Some answers to the problem of continuity between late Roman Britain and early Anglo-Saxon England may turn up here during the next few years. Already there are tantalising clues. The Life of St Cuthbert by Bede describes a settled Christian community there in the seventh century with a convent as well as a diocesan church. Had they been there before the advent of the Anglo-Saxons? A church which might have been built before the fifth century is the now demolished early church of St Alban, the British protomartyr. The Anglo-Saxon church of St Cuthbert (eighth–tenth century) was built into an existing Roman building. In 685 the citizens had enough respect for the Roman past to conduct the Anglo-Saxon monk Cuthbert ‘round the city walls to see a remarkable Roman fountain that was built into them’.

The town of Carlisle serves to remind us that the Roman Empire did not fall in one moment in history, and that a kind of Roman life may have lasted two centuries longer here than it did in, for example, Cirencester, or a century longer than at Wroxeter. In this border region British rulers in the Dark Ages claimed Roman descent through their genealogies. Perhaps they were Romano-British landed aristocrats, originally delegated power by the last official Roman governments, like some Third World rulers today who rise from obscurity and set up dynasties, monarchies, even ‘empires’ with the blessing of departing colonial powers. Could the background to the Arthur story – or at least, the milieu in which the story arose – be these petty chiefdoms, former town councillors of the north, who made their squalid timber Camelots in the temples and ruins of Roman Carlisle? One last piece of evidence may point that way.

‘THE LAST DIM WEIRD BATTLE OF THE WEST’

I have left till the end the entry quoted earlier from the Annals of Wales, which refers to Arthur’s death in a battle not mentioned by Nennius. This is the most intriguing of all pieces of Arthuriana, and the most difficult to interpret.

(c. 511). ‘The fight at Camlann in which Arthur and Medraut were killed.’ Here are the great figures of the story. Medraut-Mordred is traditionally the traitor, though in this brief statement we cannot tell whether he is Arthur’s friend or foe. Unlike the Badon annal the style is curt, more in line with the rest of the compilation. It uses a different word for ‘battle’, the Celtic gueith instead of the Latin bellum. We must remember that the language shows that it was written a good deal later than the fifth century: like the Badon annal this was not written down in its present form until 800–1100. Were the names added then? Or could the entry derive from an early record and therefore give us a definite testimony for Arthur’s existence?

As it happens, one of the Roman forts on Hadrian’s Wall bore a name, Camboglanna, which philologists think could be represented in a late form in the annals’ Camlann. Until recently this fort was identified with that of Birdoswald which stands over a great sweep of the river Irthing east of Carlisle. Now, however, scholarly opinion tends to support a new identification with the fort at Castle steads which is close to Birdoswald, and which is also situated on a sharp curve of the Irthing, as is implied in the name Camboglanna, meaning ‘crooked glen’. The site is certainly an appropriate one for an epic finale. We should not dismiss the possibility that a battle at Camlann took place in the Dark Ages: it is mentioned in the annals and it became far better known than Badon. Indeed it became a byword for a tragic, irretrievable disaster. Could this be the one genuine Arthurian reference? The site of the ‘last dim weird battle of the west’, as Tennyson put it? Is it possible that Arthur existed as a chieftain and warleader in the Solway region, not fighting heroic warfare against the Anglo-Saxon invaders, but engaged in a desperate dogfight between rival British dynasties, battling it out in the sub-Roman twilight? The idea has a certain appeal. Arthur as a kind of anti-hero, an interpretation which suits later twentieth-century taste, and is as typical of our preoccupations as Tennyson’s were of his. Did the deaths of two obscure leaders of unknown tribes give birth to the whole story, give birth to one of the greatest figures in the literature of the world? The supernatural magus of folklore? It is possible. Yet, reluctantly we must conclude that there is no definite evidence that Arthur ever existed.

Like Schliemann’s discoveries at Troy and Mycenae the modern search for King Arthur has stimulated exciting finds and theories which are changing our view of the end of the Roman world in Britain. But no more than with the siege of Troy is there convincing evidence that King Arthur’s wars actually took place.

After the Fall of Rome Celtic Britain sank into its Dark Age. Then, faced with the rising aggressive power of the Anglo-Saxon imperialists – Offa, Alfred, Athelstan – the British needed a hero. It didn’t matter whether he had ever existed in the flesh; the hope of his return was enough. As Malory put it in Morte Darthur a thousand years on:

Yet some men say King Arthur is not dead but had by the will of our Lord Jesus Christ into another place: and men say he shall come again and win the holy cross. I will not say it shall be so: but many men say that there is written on his tomb this verse: ‘Here lies Arthur: once and future king.’

The reader, however, might prefer to keep in mind the advice of John Ford’s newspaper editor in The Man Who Shot Liberty Valance: ‘When the legend becomes a fact, print the legend.’