The second consists of legendary material of local interest. It is dominated by the story of Ambrosius, who comes from Mais Elleti in Glywysing. He subsequently rules ‘all the kingdoms of the western part of Britain’, including Builth and Gwerthrynion (which Vortigern’s descendant still rules). Although it has a strong South Welsh flavour, the Ambrosius material has a scope encompassing ‘the whole of Britain’ and the fate of the British people. It starts in Gwynedd in Snowdonia and sees Vortigern consigned to Caer Gurtheyrn in the region called Gwynessi, somewhere in the north.
Surprisingly, Arthur in the Historia shows more affinity to St Germanus than to Ambrosius. The legendary, magical and prophetic features of Ambrosius’s story are completely absent from the battle-list and only slightly more apparent in the Mirabilia. The battle-list has a strong Christian tone. Arthur carries an image of Holy Mary and by her power and that of ‘our Lord Jesus Christ’ overcomes the ‘pagans’ at Castellum Guinnion. Germanus uses the power of prayer to destroy fortresses. There are other similarities of language, too.
The Mirabilia support a South Welsh location for Arthur. The writer, although supposedly covering the whole British Isles, shows his parochial vision by locating his wonders mostly between Builth in the north, Ceredigion in the west, Wye and Hwicce in the east and the Severn in the south. We cannot be absolutely sure that the writer of the Mirabilia and the Historia are the same person (Arthur is the only figure common to both works), but even if they are different, there is no doubt the works were combined at an early stage of the manuscript tradition. A shared interest in south-east Wales is one of the most likely reasons for the works, if distinct, to have been combined.
The alternative, that the Mirabilia were written later, by an author mining the Historia for an interesting name to attach to a wonderful cairn and tomb in his own land, is untenable. There is no reason for him to latch arbitrarily on Arthur the Saxon-killer. The figure of Ambrosius, already associated with magical landscape phenomena and located in South Wales, would have been an obvious candidate. The ruler of Builth’s own ancestors, named in the Historia, could have been the magical huntsmen who left the dog’s footprint at Carn Cabal. There is absolutely nothing in the Historia to lead a writer to think that Arthur had hunted a famous boar or killed his own son. There is only one conceivable reason for reporting that Arthur the soldier was responsible for the wonders of Carn Cabal and Licat Anir – local traditions that said that he was.
Gildas was familiar with the broad area. He knew of the ruler of Dyfed, for instance. His Mons Badonicus could well be within the horizons of a South Welsh writer. Both Gildas and the battle-list refer to the city of the Legion/Legions. One possible location, Caerleon, is in South Wales. The more plausible contender, Chester, is not far away. It may even, as some historians argue, have been part of the kingdom of Powys at the time.
What, then, of those stylistic features which made a connection between the Arthurian and Kentish materials so plausible? If the author knew that Arthur came from South Wales, seemingly his own home area, why did he not mention this? On the contrary, while Vortigern and Ambrosius are linked specifically to South Welsh locations, the Arthurian material is clearly written as if it is linked to Kent.
Integrating regional materials –
Nennius and Vortigern
It is quite possible that ‘Nennius’ found Arthur in sources from both areas. He knew of Vortigern from separate sources and tried to integrate the information into a single story. Vortigern was a character in at least three of the sources used, the Kentish material (paralleling his role in Bede), the Life of St Germanus and the legend of Ambrosius the prophetic child. Vortigern probably also appeared independently in other minor sources, such as the family tree of Fernmail of Builth. Evidence of other material comes in a later passage, where the Historia calculates there were twelve years from the reign of Vortigern to the discord between Ambrosius and Guitolin, a discord known as Guolloppum or Catguoloph (the battle of Wallop in Welsh). The genealogy of Vortigern has Guitolin as his grandfather, though this contradicts the story that Ambrosius was a child in Vortigern’s reign.
The contrary position, that Nennius took Vortigern from one source (the Kentish, Bede-influenced one) but inserted him in unrelated material about tyrants, cannot be supported. If that were the case, we would expect to find a single narrative thread, based on Bede, with no need for contradictory stories to be reconciled with it. Instead, we find that each source supplies a similar, but not identical, story of wicked fifth-century tyrant, Vortigern.
Nennius links the stories in two framing passages. At the start, all Vortigern’s major enemies, Scots, Picts, Romans and Ambrosius, are listed together. At the end, all his sons are given in a single list.
It is just as plausible that Nennius knew Arthur from both a South Welsh and a Kentish source and linked the two strands in a similar way. But what of the Caledonian wood, which is no more in Wales than it was in Kent? The starting point for our investigation of Arthur as a historical figure was the reference to him in a north-eastern source, the Gododdin.
THE NORTH-EAST
Rather than being linked to the southern material which precedes it, Arthur’s battle-list could be linked to the paragraph which follows: ‘When they were defeated in all these battles [the Saxons] sought help from Germany . . . until the time when Ida reigned, who was the son of Eobba. He was the first King in Bernicia, that is Berneich.’
Although I have followed modern editors in imagining a new paragraph after Arthur’s victory at Badon, Alcock points out (Alcock 1971) that this is not the case in the Harleian Manuscript. Although the initial letters of each section, intended to be filled in later in red, are not present, it is easy to see where the section breaks were intended. One was at the start of the renewed wars with the Kentish kings. The next begins ‘Then Arthur . . .’, running through without a break to the end of the first Ida passage.
If the scribe was faithfully following his original, this clearly links Arthur to the Northumbrian wars. Unfortunately, we cannot be certain of this. Arthur was a more famous figure at the time the manuscript was copied, and the scribe might well think his introduction merited a section break and decorated capital letter. After the Ida section quoted, the narrative flow ends abruptly and a scribe could reasonably decide on stylistic grounds to start a new section there.
In fact, at this point in the text, the Irish, Kentish and South Welsh material previously used in the Historia run out. Nennius uses different sources to continue his work. The new material, of northern English origin, continues the history to the reign of Egfrith of Northumbria, who died c. 678 or 682. This northern material covers the same ground as Bede and could derive from a commentary on his work. It consists of English genealogies and Northumbrian material based on kings’ lists. Incomplete chronographical material, tying Ambrosius, the battle of Wallop, Vortigern and the coming of the English to Roman Consul lists, follows this but is not connected to it.
Although the genealogical material is English, it has been annotated by a British speaker. A battle with the Picts is called Gueith Lin Garan, King Penda’s rich gifts to the British Kings is recalled as ‘Atbret Iudeu’(the distribution of Iudeu). A British pun is made on the name of Welsh King Catgabail (‘and so he was called Catgabail Catguommed [the Battle-shirker]’). The passages concerning Outigirn, Mailcunus and the British poets, must be British in origin. These additions are from a very late stage in compilation. It is simplest to see them as the work of the author himself.
Material concerning other British territories in this section is linked tenuously with the north-east. A synchronism with ‘Mailcunus, Great King among the Britons, that is in Gwynedd’, is taken as an opportunity to tell how his ancestors came from the Gododdin region. None of Outigirn’s battles is named, but it is unlikely that his exploits alone in this section derive from a Kentish or South Welsh rather than a north-eastern source. And, as we have seen, the style makes a link to the Arthur battle-li
st almost inevitable.
This northern milieu is exactly where the Gododdin leads us to expect to find sources dealing with Arthur. Neirin, its supposed author, is one of the bards named in this section. Arthur’s victory in the Caledonian wood must be somewhere in the region of Bernicia and Gododdin. Binchester in Deira could be Castellum Guinnion. There is a river Glen in Bernicia. It flows by Yeavering, a sub-Roman location which was a home of the Northumbrian Kings by the time of Bede, who mentions the river by name.
On the other hand, we have equally compelling arguments for Kentish and South Welsh sources for the battle-list. Gildas knows nothing of the north after the building of the Roman walls, but makes the siege of Mount Badon the crucial event of the British resistance. If Mount Badon was actually in the north, it would be the only northern location named in the whole of Gildas’s work. We might also wonder how a campaign against the Saxons waged exclusively in the north would have decisively turned back the invaders and secured peace for a generation. All Saxon sources saw the southern kings as spearheading the attack and archaeology bears out that this was the most heavily settled area.
Multiple sources, one Arthur?
The only explanation which fits the facts is that Nennius has blended material from different sources to create a single Arthurian chapter in his chronological scheme. We can see the same method in chapter 38 or at the beginning of chapter 56 of the Historia where Hengist’s sons Octha and Ebissa, from the Kentish source, are linked to northern events and localities.
Two possible situations could have led Nennius to create a composite chapter on Arthur:
1. He found widespread traditions of British victories before the coming of Ida, but the name of Arthur in only one area, and attributed all the battles to him.
2. He found Arthurian material in all his source areas (South Wales, the north-east and Kent) and forged them into a single narrative
The first alternative has the advantage of caution. In this theory, for instance, Arthur could be a South Welsh hero credited by Nennius with victories at Badon Hill or the Caledonian wood. He might alternatively be a warrior of the Gododdin, relocated to the south to reflect Nennius’s geographical interests.
This limited explanation is less plausible. It does not explain why Nennius chose Arthur for this role. He was quite capable of writing about more than one character. He knew of Vortimer and Outigirn as anti-Saxon leaders. Both are included in the Historia without the need to fuse their battles in a single list. He has no difficulty recording Vortimer and Ambrosius or Outigirn and Mailcunus as contemporaries without combining them into a single character.
If Nennius was just searching for a name to attach to a list of battles, he had plenty to choose from. He knew the names of the famous kings Urbgen and Mailcunus. Embreis Guletic was a local South Welsh hero with a historiographical tradition, from Bede, that he fought the Saxons at Badon. If Nennius was concerned that his chronology involved a large gap between Hengist and Ida, he could have mined the genealogies for Fernmail’s ancestors or Catel’s descendants of the right period. He did not even need a name to link to the battles, which could just as easily be ascribed to the anonymous ‘Kings of the Britons’.
If Arthur was a significant local figure in any one of the source areas, we can hardly imagine why Nennius did not make this explicit in the text. The battle-list gives no idea of Arthur’s ‘home ground’. There is no mention of his descendants or ancestors, or his native land. Yet Nennius often links the figures of the fifth/sixth centuries to contemporary locations and dynasties. If he could make these links for Vortigern and Ambrosius, why not Arthur? In particular, why are the majority of the battles at obscure locations, rather than at local sites or ones already mentioned in the Historia?
The creation of the figure of Arthur to link previously unconnected traditions of the British resistance is motiveless and implausible. Elsewhere in the Historia, Nennius is happy to eke out sparse materials, including named individuals with only one incident connected to them, or events without named participants. If he had in mind the idea that a legendary superman must have led the Britons, his imagination has failed in the case of Arthur. The most legendary material in the whole work, with prophetic worms, fatherless boys and councils of wizards, concerns the undoubtedly historical Ambrosius. Similarly, the very real Germanus of Auxerre is shown destroying whole fortresses with fire from heaven. To turn an incredibly famous, non-historical Welsh culture-hero into a mundane Saxon-fighter is simply bathetic. The idea that Nennius did not have the inventiveness to parcel out victories in Kent to, say, Pascent son of Vortigern, near South Wales to Ambrosius and in the north-east to Urbgen or anyone else mentioned in his sources, beggars belief. There is no reason to ascribe all the victories to a single Arthur the Warleader, unless they were already ascribed to Arthur in his sources.
This leaves us with just one plausible explanation: Nennius found material relating to Arthur as a warleader in all three of his source areas. Just as he had with Vortigern, he worked these sources together into a single story. That he meant his readers to understand that Arthur had a wide-ranging military career is implicit in his ‘kings [plural] of the Britons’. Furthermore, Nennius used the exploits of Arthur across the country as a centrepiece to hold the clearly regional Vortimer and probably regional Outigirn into a framework of an extended pan-British war against the Saxons.
The evidence of Historia Brittonum is that sources referring to Arthur across Britain already existed before 830. The story, that Arthur was famous for fighting against the Saxons, that he fought at a Roman fortification and overthrew a large number of men in a single charge, is the same in both the Historia and the sixth-century Gododdin. The Gododdin is not one of Nennius’s sources, so the story must have been arrived at independently.
There is no need for a single Arthurian battle-list to be Nennius’s source. It is more likely that Nennius created his list by blending widespread material on Arthur the Warleader. This increases the value of the Historia, rather than diminishing it. It would clearly be more valuable to know that several sources contributed to Nennius’s picture of Arthur the fifth/sixth-century Warleader, than to imagine that he picked an old Welsh poem out of his heap of sources and set it, barely altered, in an historical scrapbook between St Patrick and Ida of Bernicia.
Wars across Britain
There are two possible ways that sources from different areas preserved accounts of the wars of Arthur:
1. Arthur fought widespread campaigns, winning battles across the country. These left memories of his exploits in various regions
or
2. Arthur fought in just one area, but stories of his exploits spread to other regions. Perhaps British exiles had physically moved to another area taking the stories with them. Alternatively, Arthur was perhaps such a famous character that other areas wished to claim association with him.
The first suggestion is the most straightforward. There is nothing implausible about Arthur fighting Saxons across the country. Writers often lose sight of the fact that Britain is a relatively small area. Many theories about Arthur have limited his activities to what are in fact tiny regions, as if a military commander were restricted to the range of a committed rambler. Hence we find Arthur’s campaigns located in, say, Somerset or Gwent. In reality, from the Roman conquest onwards, large conflicts have turned the whole country into a war zone. Gildas reports the Saxon invasion and British resistance as extending across the island.
Other writers have concocted ideas of Arthur as a cavalry leader to explain his ‘extraordinary mobility’. As we have no idea over what period these battles were fought, we have no idea how swift his forces would have to be to reach them. Even if they were fought in rapid succession, armies reliant on infantry, such as those of the Roman Agricola or the Saxon Harold in 1066, show us how easily they could move from one part of the country to another.
The Anglo-Saxon Chronicle provides numerous cases of wide-ranging conflict. Ceolwulf of Wessex is s
aid to have made war against the Angles, the Welsh, the Picts and the Scots. His reign, beginning in 597, could be classed as belonging to the same ‘legendary’ era as Arthur’s, but there are many examples from the later and undeniably historical portions of the Chronicle. The Vikings invaded the areas which had seen the first English settlements. They ravaged the island and were confronted by various English kings. These campaigns show what was possible in the Dark Ages. Between 872 and 877, the Viking ‘great army’ sallied out of East Anglia to camp at Torksey in Lincolnshire, on the Tyne, at Dumbarton, then down to Exeter and Gloucester. In 892 another great army landed in Kent. Three years later they were retreating from Chester, sweeping through Wales, then returning to East Anglia via York. In this campaign, they faced a united English defence under King Alfred. Arthur’s battle-list is thus not inherently implausible as a series of campaigns against the earlier invaders.
The alternative, that Arthur was a regional warleader whose fame spread to other areas, raises the question ‘why?’ This could only have happened if Arthur was already famous as a leader of battles, more famous than the ‘real’ leaders of the resistance. The story of the British resistance is not a legend. It is attested by the contemporary Gildas. It was brought about by the united efforts of the Britons. We would expect that, if all the kingdoms had different traditions, they would each have been able to name a local king who joined in the fight. As it is, none of these kings is named by Nennius. They are completely overshadowed by the warlord himself, Arthur.
Even if Arthur was a battle leader whose fame spread beyond his home area, he must still have been famous before the early ninth century. This is fundamental to understanding the Gododdin verse which began this investigation. In the Historia, Arthur is only famous for fighting the Saxons. The battle of Mount Badon was the memorable victory against them, as Gildas put it. If Arthur was not the commander at the battle, then the real victor of Mount Badon has been replaced by an even more famous Arthur. But what could Arthur possibly have done to become more famous than the victor in the greatest battle against the Saxons? In truth, there is only one thing Arthur can reasonably have been famous for. He must have been the victor at Mount Badon.
The Reign of Arthur Page 5