The sixth battle was on the River Bassas which, like the Dubglas, is unknown. The two names rhyme, which might suggest Nennius found them together in a verse source. This in turn might imply that they are in the same area. Beyond that, we can only guess which of the riverine scenarios it represented.
The seventh battle was in the Wood of Celidon. This is the only battle of whose general location we can be sure. It was somewhere just north of Hadrian’s Wall. We assume that the Britons are on the defensive here, as the wood is inland from any potential Saxon settlements. Such lack of settlement is good evidence for a British victory against the Saxons. The area was the focus of English activity from the mid-sixth century and there is no reason to think they simply avoided the area fifty years previously.
The next battle gives us unequivocal evidence that Arthur was a Christian. He carries the image of the Virgin Mary on his shoulders. The scene of the battle, Castellum Guinnion, is assumed to be a Roman fortification. As the result of the battle is that the pagans are put to flight, we have to understand that the Britons are defenders. If the Saxons were in the castellum, they had nowhere to flee. The best guess is that Guinnion is Vinovium (Binchester, Country Durham) hence part of the north-eastern milieu. It is close to Catterick/Catraeth.
The ninth battle, in the City of the Legion, must also be interpreted as a British defence. On the evidence of the city list, this was most likely Chester, which only fell to the Saxons in the seventh century. Caerleon is a viable alternative. Either would derive from a South Welsh source. None of the rest of the battles is demonstrably in the north, and it may be that Nennius has eight named battles from a north-eastern source, followed by four from southern sources. The tenth battle, on the shore of the river which is called Tribuit, and the eleventh, on the hill which is called Agned are, with the City of the Legion, the only ones which a tenth-century South Welsh writer felt confident to name in Welsh.
Agned may or may not have been the same as Bregomion. A northern location for this has been suggested, although Branogenium (Leintwardine in Herefordshire) would fit a South Welsh pattern better. If Agned is in the South Welsh area, we assume that the Britons are defending it. Unless Arthur made a habit of being trapped in such situations, the most sensible inference is that he led relieving forces to rescue the besieged Britons.
None of the battles has a demonstrably Kentish location. We could assign the unknown locations to Kent, but that would be sheer guesswork. On the evidence, the Kentish source may have done no more than refer to the wars of Octha, without giving the battle names.
The twelfth battle was on the hill of Badon, a famous victory remembered as having secured peace and a virtual end to Saxon attacks. The Annales use this battle to report Arthur’s Christian affiliation. Here we are told that Arthur carried the cross three days and nights, so a siege seems intended, as described by Gildas.
This story of Arthur, Warleader of the united Britons c. 500, is internally consistent and consistent with other sources. There is nothing inherently implausible about it. Someone led the British at the real siege of Mons Badonicus. Historia Brittonum and Annales Cambriae both independently said his name was Arthur. We have to assume that Arthur did not perform this as a one-off feat, but that he had a military career in the wars between Britons and Saxons which preceded it. The battles ascribed to him are not (apart from Badon) famous ones in search of a named leader.
The Historia and the Annales both tell us that Arthur was at the British victory at Mount Badon. They have nothing else in common. There is every reason to suppose that they were composed independently. Neither the Annales nor the Historia has detailed knowledge of Gildas. They are most unlikely to have derived their versions of Badon from it. They were both probably loosely familiar with Bede’s History, but Bede thought that Ambrosius was the victor. The Annales had not even heard of Ambrosius, at least in any context after 455. The Historia had and, logically if Bede was the only source for Badon, would have linked the great battle to Ambrosius. That Nennius did not is a powerful argument in favour of his independent use of a pre-existing tradition.
Where was Mons Badonis?
Is Gildas’s siege of Mons Badonicus really the same as the battle of (Mons) Badonis in the Historia and Annales? We need to tackle this question directly, as it is the touchstone for the existence of Arthur as a historical character. We will look at Gildas’s description of the Saxon Wars later. Suffice for the moment that he says they culminated in the siege of Mons Badonicus. Mons means a hill or mountain. Gildas uses collis which specifically means hill in other contexts, so is probably thinking of Mons as something on a larger scale. Some writers, usually with special pleading for a favoured location, translate it ‘hill-country’. On its own, this might be so, but it is hard to imagine how someone might besiege a hill-country!
Badonicus is an adjective describing what kind of mountain it is – a Badonic, Badonish or Badonian mountain. Gildas uses this adjectival form only once elsewhere, when he describes the area across the seas from Britain as Gallia Belgica – Belgic Gaul. The reasonable understanding is that this hill is in a region called Badon or at a place called Badon. It is a highly unusual construction if the name of the hill itself is ‘Badon’.
More inferences can be drawn from Gildas’s text. He has previously told us how the Britons had fled to the hills (using the form colles). We understand, therefore, that the Britons are those being besieged at the Badonic Hill. Although it is possible that an unsuccessful siege was the undoing of the Saxons, it is unlikely that this would have the catastrophic results Gildas describes. Given that Gildas thought that retreating to the fortifications was in itself misguided, we have to conclude it was not a tactic used by the victorious Britons in this case. The obvious inference is that the victors were a relieving force which broke the siege of a strategic British position in the Badonian area.
It is unlikely that Gildas expected us to infer that the hill at/in Badon was itself an important centre or fortification. He had a range of words such as urbs (town/city), castellum or receptaculum (fortress) which he could have used if that had been his intention, rather than the neutral mons. Archaeological evidence shows that the refortified hillforts of the period had usually stood without residential or military use for centuries. They might no longer have had names of their own.
In Historia Brittonum, Bellum means a battle, rather than its classical meaning of war. Badonis means of Bado or of Badon, understood as the name of a person or place. Compare it with Celidonis earlier in the list – the forest of Celidon. The writer is not telling that the hill was called Badon, any more than that the forest was called Celidon. He was perfectly capable of describing what a hill was actually called. The previous battle was ‘on the hill which is called Agned’.
The final piece of evidence that the hill is at Badon not called Badon comes in the Annales, where the engagement is ‘Bellum Badonis’ – the battle of Badon, with no mention of its hill. This battle seems to take three days and nights, indicating it is most likely a siege, as Gildas said. There is nothing to make us think that the sources are not all referring to the same battle.
As for its actual location, probably we should look in the south-western part of Britain, east of Dumnonia and where Gildas and Nennius share a geographical interest. Nennius is likely to have found information on the battle either in a South Welsh source, which would come within Gildas’s area of interest, or in the Kentish material, essentially a romanticised version of the story given by Gildas.
Within these areas, we are looking for a place or region called Badon with a hill, probably a large hill, fortifiable if not with a fifth/sixth-century fortification present. We are not looking for a hill called Badbury. Badbury does not mean a hill near Badon. It means a fortification named after (for argument’s sake) Badon. It would be the English translation of a British Din Badon or a Latin Castellum Badon, not Mons Badonis/Badonicus.
We do have one significant pointer to the location of the battl
e of Badon, or at least where the writer of Annales Cambriae believed it to be. Approximately 150 years after the Arthur victory (the round figure may be the result of a deliberate synchronism) c. 665, is recorded ‘Bellum Badonis Secundo’ – the second battle of Badon. The battles at this period in the Annales are fought by the North Welsh and the Mercians, and the Northumbrians.
Bede presents the circumstances of these battles. His people, the Northumbrian Angles, were struggling against the Mercians who had not yet converted to Christianity, and their North Welsh allies. Although there is much confusion about the names of the battles and how those in Bede relate to those in the Historia and the Annales, Bede records nothing which would equate to a second battle of Badon around 665. Bede knew from Gildas that there had been a first battle of Badon, and might have mentioned if he knew about a second. The Annales record the first Saxon Easter in the same year as Badon II. This could be a mistaken impression of the Synod of Whitby, on the Easter controversy, in 664.
This suggests that we must look outside Bede’s area of interest for this second battle. The Anglo-Saxon Chronicle covers wars of which Bede had little or no knowledge, those of the West Saxons. Bede reports how Wulfhere of Mercia established his hegemony over the South Saxons and the Isle of Wight. The Anglo-Saxon Chronicle continues the story with Wulfhere’s battle with Aescwine, King of Wessex. It dates this to 675, the year that Wulfhere died according to Bede. However, these early dates for West Saxon history can hardly be treated with precision. The Chronicle spreads them out regularly, perhaps to give the West Saxons a more consistent presence in the early entries. It could be that Wulfhere’s battle with the West Saxons was actually part of his southern campaign ten years before this.
Searching for Mount Badon.
The Mercians fought the West Saxons at Bedanheafod, meaning Bedan-head. This seems more than coincidentally similar to the Annales second battle of Badon. ‘Head’ could convincingly refer to a hill or mountain. Could we in this entry be looking at the English name for Mons Badonicus?
Where then was Bedanheafod? Logic dictates it was somewhere on the borders of the Mercians or the West Saxons, or within Wessex, given the circumstances of the battle. Furthermore, for the battle between two Saxon kings to be at the same place as a previous battle between the Saxons and the Britons, it would have to be somewhere which in 500 was in British or disputed territory, but by 675 was outside the British sphere.
In Bede’s time, the West Saxons bordered on the Hwicce in Gloucestershire and Somerset in the west, but that is all the evidence we have. The Hwicce did not found a kingdom lasting into the eighth century, and their history is thus unknown.
As the Anglo-Saxon Chronicle relays it, during the first half of the sixth century, West Saxons had spread as far as Netley and Charford. In 552 they were fighting the Britons west of this frontier, taking first Old Sarum, then Barbury in Wiltshire. Next they moved north to Bedford, moving into the Thames Valley in 571. Then, 577 saw the momentous victory at Dyrham which secured Gloucester, Cirencester and Bath. The frontier was further defined by battles at Alton Priors in Wiltshire and Stoke Lyne in Oxfordshire.
Approaching our date (665/675) we see the 614 battle at Beandun (unknown), 648 hegemony over Ashdown (Berkshire), and 652 fighting at Bradford-on-Avon. In 658 victory over the Welsh gave the West Saxons Penselwood as far as the Parret in Somerset. By 709, their westward expansion had led to the creation of two dioceses among them, divided east–west by the Selwood forest. Selwood has Barbury and Old Sarum on one side and all the post-Dyrham expansion sites on the other. It is tempting to see it as marking a division between old and new West Saxon territories. Britons held on to the rest of Somerset until the fall of Somerton in 733.
There is nothing absolutely certain about this chronology or even the identity of the belligerents. We would expect other Saxon peoples such as the Hwicce to be those extending the frontiers, not the West Saxons. The impression, however, is that the area which changed from British to Saxon control over the 150 years is Wiltshire, north Somerset, Gloucestershire, as well as the Thames Valley from Berkshire to Oxfordshire. This is very much where we would expect to find Mons Badonicus, as we shall see from Gildas. It is not unlikely that Mons Badonis and Bedanheafod were fought in the same place. Plummer suggested (1892–9) that Bedanheafod was (Great) Bedwyn, Wiltshire. Although there is no continuity of names – Bedanheafod is not found anywhere other than in this chronicle entry – the site is a surprisingly good one. Bedwyn itself is in the river valley, but its head or hill is immediately to the north and west of it, capped by the hillfort now called Chisbury. Chisbury dominates all approach routes and would, for instance, constitute a barrier to Saxon expansion. It is, moreover, on the extreme east of the Savernake Forest, a perennial obstacle to movement. Fortified hills and dense forests are exactly the defensive positions Gildas tells us the Britons adopted. Great Bedwyn also happens to be extremely close to the cluster of those Bad- names, Badbury and Baydon, used to support the case for Liddington.
The Death of Arthur
It is rather too much to expect that Arthur the Warleader would die in his bed. Gildas does not say what happened to the victor of Badon, but we can imagine a likely scenario. Although the external wars ceased, civil wars continued to be waged down to Gildas’s own time. Historia Brittonum hints that Arthur was involved in civil strife, relaying the tradition in Ercing that he had slain his own son.
Annales Cambriae say that Arthur was slain at Gueith Camlann twenty or twenty-one years after Badon, enough time for him to have a grown son to fight against. Alongside Arthur fell Medraut. All subsequent versions of the death of Arthur have made the two adversaries. It seems likely that this is what the annalist intended.
Where was this Camlann? We would expect it to be in a British area. Although we have used the grave of Anir in Ercing to locate Arthur, we have to acknowledge that civil strife need not be carried out in either of the participants’ backyard. The tyrants pursued thieves all over the country, if need be, as Gildas tells us.
Camlann could be in one of the South Welsh, north-eastern or Kentish areas where Arthur’s Saxon-fighting activities have been placed. I am more inclined to think that the battle was fought elsewhere, precisely because Nennius does not know about it. Its name, Gueith Camlann, as opposed to Bellum Badonis, might indicate a Welsh/British source of continuing tradition. While Mount Badon was more or less unknown in Welsh tradition, there is a rich vein of legendary material relating to the battle of Camlann. Badon, most probably, passed out of British control by 665. We can suppose that Camlann, on the other hand, was still a living British location at this time.
Although the name Camlann actually means ‘crooked enclosure’, there is a consensus that it really derives from the British ‘Camboglanna’ – crooked stream/valley (Alcock 1971). There is one place known to have borne that name in Roman times, the fort of Castlesteads by Hadrian’s Wall. Camboglanna was, in fact, maintained during the sixth century. It is possible that Arthur’s heroic death at a Roman fort in the north is one of the features which he had in common with Guaurthur of the Gododdin. The northern Camboglanna is thus a distinct possibility for Arthur’s last battle.
There are, though, some alternatives. There is a Camlan in modern Wales, the side of the valley above a sharp bend in the River Dyfi. This lies on the main route between Gwynedd and Powys. Above it is a second hillside Camlan and tributary Afon Gamlan (River Camlan), a suggestive cluster of names. Gildas testifies to the civil wars raging beyond Dyfed, presumably either in North Wales or Powys, in which the tyrant Maglocunus took the leading part. It is also close enough to Ercing to consider that it was part of the Welsh conflict which had previously seen Arthur kill his son, and is just beyond Carn Cabal. There is a distinct North Welsh predominance in the Annales, which would add to the plausibility of this suggestion.
There are plenty of other Cam/Camel streams to support regional theories. Geoffrey of Monmouth was the first to specify a location f
or Camlann, as a stream in Cornwall, the modern Camel. Tradition places the battle at Slaughter Bridge, Camelford. This is the first suggestion that Arthur might have fought in Dumnonia. Another suggestion is the Camel by South Cadbury hillfort, one reason why Leland identified it with Camelot. Clearly, this method, based on the similarity of names, is too imprecise. More precision can be gained by re-examining the Annales entries.
The early annals are based on Irish Annals composed in 741. Eleven British entries are added to these annals, giving events up to c. 613, the first of which are the two Arthurian ones. After this, the focus of the Annales is obvious. The entries concentrate first on the wars of Cadwallon of North Wales and the Northumbrians, before shifting to South Wales. From the early ninth century, at least, the Annales seem to be written in St David’s. Kathleen Hughes (1973) identified the first stage of composition as being between 741 and 769. If the Arthurian entries date from that phase, they would be earlier than Historia Brittonum. Unfortunately, their early placing in the Annales is not necessarily indicative of their early composition.
The eleven entries are located as follows: two unlocated (Arthurian), four northern, three or four North Welsh and two or three South Welsh (depending on whether we consider Urbs Legion (Chester) a northern outpost of Powys or part of Gwynedd). One of the South Welsh entries, on the death of Bishop Dubric, actually seems to derive from the St David’s phase, grafted on to an originally North British entry.
The balance is, therefore, in favour of Camlann being in North Wales or the north, but assigning the Arthurian locations to any of the three regions would create an imbalance in its favour. We need another form of analysis to be certain.
The Reign of Arthur Page 8