The Reign of Arthur

by Christopher Gidlow

  Arthur is a figure of unparalleled importance in the Annales. No other secular figures or events are recorded in the first hundred years of entries. The entries in this section, excepting the first about Pope Leo changing Easter, refer to the births and deaths of Irish ecclesiastics, derived from the framework the Annalist was using at this point. In the later entries, northern Wales and North Britain become more prominent, but South and Central Wales are sparsely dealt with. Here we seem to have a confirmation of the Historia, that Arthur’s fame was not confined to South Wales.

  There is absolutely nothing objectionable about the references to Arthur in the Annales. Everyone else in the Annales is a real historical character. The style of the entries about Arthur is no different from those, for example, concerning the wars of Cadwallon. One cannot help thinking that, if the Annales were the only source other than Bede and Gildas, his existence as the victor of Badon would be taken for granted.

  Received wisdom has it that the Annales had their origin in marginal notes in a table of Easter dates based on a great cycle of 532 years. As we have it, scribal defects would make any such computations impossible. Although the years are marked out in decades from the beginning of the cycle, in their current form some decades have eleven years, some nine and some ten. Easter is mentioned in the first entry and again, 220 or so years later, when it is first celebrated among the Saxons. In this year, a second battle of Badon was fought. Some entries are in the nature of ‘St Columba born’ (five years after the first battle of Badon), showing that they were not compiled, diary fashion, in the back of some monk’s service book, but constructed after the events, like the early entries of the Anglo-Saxon Chronicle. As the entries approach better-recorded times, we can fix the otherwise unlocated chronology at various points. These are not always what we expect, and are in some cases very defective. The battle of Chester, which we would have expected to find c. 603, seems to be given ten years later. This is an easy mistake to explain away, caused by placing the event in the next numbered decade. What, however, are we to make of entries like the first Saxon Easter, apparently in 665, or the conversion of Constantine to the Lord in c. 587?

  The battle of Badon is noted around 516. It is placed fifty-nine years after the death of St Patrick, drawn from the Irish Annals. This shows once again the Annales’ independence from the Historia. Nennius had used the phrase ‘At that time’ linking St Patrick’s story to Arthur’s. By the time the Annales were composed, some versions of the Historia, including the Vatican Recension, had moved the St Patrick section to after the battle of Badon. Equally, there is nothing at all about the Saxon wars leading up to Badon, nor about Ambrosius or Vortigern. As the Annales stand in the Harleian Manuscript, it is possible to read the unfinished chronographical material at the end of the Historia, covering for instance Ambrosius and Guitolin, as being connected with them. If so, the writer was unable, despite these computations, to work out a position for Ambrosius relative to the other annals. There are no mentions of the Saxons at all until a hundred years after Arthur’s time.

  Modern translations of the Annales give the impression that dates such as Badon 516 or Camlann 537 are actually to be found there. In fact, there is quite a lot of variation between possible dates for those events. We could start with one of the later dates, such as the death of King Edmund (year 503), reported in the various Anglo-Saxon Chronicle manuscripts to 947, give or take a year, then count backwards. This would indeed place the battle of Badon (year 72) in 516, again give or take a year. Over this span of time, six phantom years have crept into the Annales, due to some decades being given eleven years. If the decade counts are wrong, but the number of individual years is right, then Badon would be dated 510.

  If, however, Badon is 516, then the battle of Chester (year 169) is placed ten years later than expected from Bede. If we fixed the early Annales by making the battle of Chester AD 603 as Bede does, this would put Badon at 506. On the other hand, if the Death of Edwin in year 186 is linked to Bede’s AD 633, Badon would have been fought in 519. Alternatively, we could work forward from Pope Leo’s changing of Easter in 455, bringing us to a battle of Badon in 518.

  All the possible Annales dates seem later than we would expect from the Historia, where Arthur fights Hengist’s son soon after the time of St Patrick. They fit better with the idea that he was an adversary of Octha, Hengist’s grandson, and that his career ended just before the arrival of Ida of Bernicia in the mid-sixth century.

  The Arthur entries are two of three in the early part of the Annales which relate to secular events in Britain. These three entries were placed at ten- or twenty-year intervals. The slight misplacing of Badon relative to Camlann (21 instead of 20) is explained by the fact that the previous decade has been given eleven years by accident. These regular spacings suggest how the Annalist fixed them to the rest of the chronology. The link is the beginning of the reign of King Edwin of Deira. This comes ‘101’ years (which without the phantom year is exactly 100 years) after the battle of Badon. North-eastern Arthurian material could have passed on this synchronism, or it could just be a rough estimate (‘one hundred years before the reign of Edwin’) originating with the Annalist. The reign of Edwin, when Christianity was first brought to Northumbria, starts the historical material with which we are familiar from Bede, which in turn links to the career of Edwin’s North Welsh opponent, Cadwallon, who figures prominently in the Annales.

  The link with the reign of Edwin is easy for the Annalist to set down. He simply noted the related British events two years after the numbered decade counts. This in itself led to scribal confusion, where the copyist was inconsistent as to whether the numbered decade counts should themselves be counted as years.

  The third of the secular entries in the Badon–Camlann sequence is recorded ten years after Camlann c. 547 (537 at the earliest and 550 at the latest): ‘There was a great plague, in which Mailcun King of Genedota (Gwynedd/North Wales) passed away’. This king appears twice more in the Harleian Manuscript. In the Historia, immediately after the passage on Outigirn and the British poets, Neirin, Taliessin and the others, we read ‘Mailcunus reigned as great king among the Britons, that is in the Guenedota region’. His descent from Cunedag of the Gododdin tribe is then given.

  Included in the Harleian Manuscript is a series of genealogies. These date from the same period as the Annales, and may be related to them, though they do not appear together in any other manuscripts. The first of the genealogies is of Ouen, tracing his descent through his father Higuel (Hywel the Good) and the Kings of Gwynedd, back to Mailcun, Cuneda(g) and beyond. In the Annales, Higuel’s death is recorded three years after Edmund of England, that is in 950. Mailcun is shown as the ancestor, five generations back, of the King Catgollaun (Cadwallon) of Gwynedd, the adversary of the Northumbrian kings. In the Annales, his death is recorded as c. 631, and is dated by Bede to 634. This seems a plausible span of time.

  The linkage of these characters to Higuel’s dynastic line is uncertain. Our first impression is that there are too many generations. Death dates for Higuel’s ancestors are given in the Annales six generations back to Rotri, who died 196 years previously – a reasonable average of 32.66 years per generation. Rotri, however, is given as the great-grandson of Cadwallon. If the same average is continued back seven generations beyond Rotri to Mailcun, we would expect to find Mailcun dying around 470, much earlier than his position in the Historia or Annales.

  The genealogy of Higuel seems to have been constructed by combining his well-attested ancestors back to Rotri with some famous early figures from the Annales: Cadwallon and Catgualart, Iacob son of Beli and Mailcun. Where there is a long gap between Annales records, another figure (Iutguaul, Catman and Run) not mentioned in the Annales is inserted in the genealogy, a process which has apparently inflated the number of generations. The genealogy creates the illusion that a single hereditary dynasty has been ruling Gwynedd since at least the time of Mailcun and that Higuel and Ouen are its lineal descen
dants. We know this is not actually true. After the death of Rotri, a Caratauc was king of the region. A different genealogy is given for him, tracing his lineage back to Mailcun’s grandfather.

  Disjunctures like this make it difficult to place too much value on the genealogies. Some are demonstrably false. It is extremely unlikely that Hywel’s wife was descended from an unattested Dimet (‘Dyfed man’), son of the Emperor Maximus, and totally untrue that Maximus was the descendant in eight generations from Constantine the Great.

  Much weight has been put on them, because of the importance of genealogy in medieval Wales. Then, Welshmen were usually named with their father’s name used as a surname. However, this does not seem to have happened in the sixth century. None of the tyrants Gildas denounces are given patronymics, nor does one feature on Voteporix’s memorial stone. In the Gododdin, many of the warriors do not have patronymics. In the Historia only Vortimer, ‘son of Vortigern’, is named in this way. In the Annales, the first Briton to be given a patronymic is Selim, son of Cinan, in year 169. Before him only the two Irish leaders, Gabran, son of Dungart, and Aidan map Gabran have them. Even the seventh-century Cadwallon’s father’s name is not given in early sources.

  The genealogies of Higuel, his wife Elen and Caratauc have similar plans. Mailcun is given in the same generation as Cincar son of Guortepir (Vortiporius) and of Cinglas (Cuneglassus). This information, that Mailcun and Cuneglassus are contemporaries and that Vortiporius is a generation older, harmonises with what is recorded by Gildas, and could derive from him. The names ‘Arthur’ and ‘Outigirn’ do appear in the genealogies, but in contexts which make it clear that these are not the same as the warleaders of the late fifth and sixth centuries. Arthur is listed much more recently than Mailcun, and Outigirn much earlier.

  The importance of this information is that Mailcun was indisputably a real person from the generation following Mount Badon. He was a contemporary denounced by Gildas, using the sixth-century version of his name, Maglocunus.

  Maglocunus and Maelgwn Gwynedd –

  Double Standards in the Dark Ages

  Practically every historian studying the period, no matter how sceptical about Arthur, takes it for granted that Maglocunus is Mailcun or (in modern Welsh) Maelgwn Gwynedd. That is, they accept that Gildas’s Maglocunus was the sixth-century ruler of Gwynedd, and probably an ancestor of the Gwynedd dynasty. Even the most sober historian is prepared to construct complex arguments about Gildas’s location or the government of sub-Roman Britain based on that equation.

  Unequivocally, Maelgwn Gwynedd is a figure of ninth- and tenth-century historical writing, exactly as Arthur the warleader is. He is found in exactly the same sources, Historia Brittonum and Annales Cambriae, with all their limitations. Gildas does not mention the Kingdom of Gwynedd at all; still less does he say that Maglocunus is its king. That information is derived from exactly the same sources that tell us Arthur was the leader of the Britons at Mount Badon. In the case of one, Gildas names the man, Maglocunus, without naming the place, in the other he names the place, Mons Badonicus, without naming the man. The logic – that Maglocunus must have been king of somewhere, and that Gwynedd must have had a king, therefore there is no reason not to accept the ninth-century tradition that Maelgwn was king of Gwynedd – can be applied with equal force to Arthur. Somebody led the united Britons at the siege of Mons Badonicus. The only person the Britons said was the leader was Arthur, and we have no reason not to accept that tradition either. On the contrary, the arguments in favour of Arthur leading the victorious Britons are far stronger than those which make Maelgwn Gwynedd Maglocunus.

  The fact that Maglocunus is named in Gildas while the leader at Badon is not adds nothing to the force of the argument. Ambrosius is named by Gildas too, but that does not allow us to infer that he really was a fatherless prophetic boy who predicted magical worms beneath Vortigern’s fortress. Although Gildas has much to say about Maglocunus, that material does not appear in the Historia or the Annales. It is simply his name which is used in those later sources.

  In the ninth and tenth centuries, Arthur served no obvious political purpose beyond offering Britons the comforting idea that one of their leaders had fought successfully against the English. No contemporary dynasty claimed to be descended from him, or acknowledged him as part of a collateral line. No Welsh state of the period owned him as a native son or used him to justify their territorial claims. His status and battle-sites bore no relationship to the political realities of the time.

  Not so Maelgwn Gwynedd. By the time of the Historia, and even more so the Annales and Harleian Genealogies, the rulers of Gwynedd were undeniably the most important British rulers. It is hardly surprising they should adopt Gildas’s foremost tyrant, Maglocunus, as their ancestor. After all, the crimes Gildas accused him of – fighting rivals, listening to praise singers and rewarding his warriors – were likely to endear him to a Dark Age audience.

  It is quite conceivable that references to Maelgwn might have been altered to fit a contemporary North Welsh agenda. It is clear that Nennius tries rather awkwardly to place Maelgwn in a North Welsh context. While providing British synchronisms for the reign of Ida, he writes ‘Then at that time Outigirn was fighting bravely . . . then Talhearn Father of Inspiration was famous for poetry and Neirin and Taliessin and Bluchbard and Cian at the same time were famous for British poetry, Maelgwn, Great King of the Britons was reigning.’ Then is tacked on the transparent gloss ‘that is in the Gwynedd region’. It is transparently an addition not just because it is clumsy, but because it is immediately contradicted by the rest of the passage. Maelgwn is a descendant of Cunedag who 146 years earlier had come down from the lands of the Gododdin to expel the Irish from ‘these regions’. What regions were these? ‘The sons of Liathan prevailed in the country of the Demetians and in other regions, that is Guir Cetgueli (Gower Kidwelly), until they were expelled by Cuneda and by his sons from all the British Regions’; South Wales, that is, not North Wales at all.

  That is the evidence for Maelgwn Gwynedd – hardly compelling. The Historia later deals with the deeds of Cadwallon, King of Gwynedd, with no indication that he is a descendant of Mailcun. That information is only given in the Harleian Genealogies, where not just those two kings but also Cuneglassus, Vortiporius, Magnus Maximus, Constantine the Great and many other figures of history are recruited to the family trees of Higuel the Good and the royal family of Gwynedd.

  The Annales reinforce the view that Maelgwn is King of Gwynedd: ‘Great Plague, in which Mailcun King of Gwynedd passed away.’ It is hardly surprising to find him here, given the prominence of the kings of Gwynedd in the other entries. Once again, an identical source is used by historians to confirm that ‘Maelgwn Gwynedd’ is the tyrant Maglocunus, to that we are using to identify Arthur as the leader of the Britons at Mons Badonicus. While the evidence for Maelgwn Gwynedd is equivocal, that relating to Arthur gives useful and plausible evidence supporting what we know from Gildas. After the Annales, Welsh legends and Saints’ Lives would give similar treatment to Maelgwn Gwynedd as to Arthur, which should not lessen our belief in the historicity of either.

  I do not argue that the Historia is wrong in connecting Maglocunus with North Wales. I believe, rather, that the case for ascribing the victory of Mount Badon to Arthur is much stronger, not being tainted by obvious dynastic interests. There is no reason why both pieces of information should not have surfaced in the written record of the ninth-century Historia Brittonum after having been preserved since the sixth century. However, historians cannot have it both ways. If Arthur must face blanket challenges to his existence, then so should Maelgwn Gwynedd, and if Maelgwn Gwynedd can be accepted on a balance of probabilities, then so should Arthur.

  The Battle-list

  The list seems intended to show that Arthur fought across Britain. We can infer that it combines locations in the north-east, the Kent area and in the Severn Valley and adjacent regions at least. Most of the battle sites are unknown, suggest
ing that they are now in England, with English names. Some clues can, however, be drawn from the little information given.

  Many of the battles are on rivers. Logically, the battles would either have the rivers across them as a barrier, or they would follow the line of the river as an invasion route. In either case, Arthur and the Britons could be attackers or defenders. If the campaign is attempting to cross the river, then Arthur would either be attacking into Saxon territory or holding the river to prevent a Saxon crossing. Alternatively, he could be using the river valley as a line of advance downstream into Saxon territory, or blocking the Saxon advance upstream. The Saxon presence in coastal areas, with Britons in the highlands, makes it inevitable the rivers were used in this way.

  The battle on the River Glein is fought near its ‘mouth’. The Welsh used the same word for mouth or confluence, making the latter a possible reading. It is unlikely that either side is trying to force a crossing at such a site, so this must be a battle along the line of the river, hence a thrust into Saxon territory on the east coast. There are two existing River Glens in England, one in Northumberland, one in Lincolnshire. Both are in plausible war zones, with the balance in favour of the Northumbrian Glen, which is named by Bede and runs by the formerly British and then Anglian royal centre at Yeavering. Neither Glen has a mouth, being tributaries of larger rivers. Perhaps the Glen name was originally carried by the main branch down to the sea, or the confluence was indeed intended.

  The case for the Lincolnshire Glen is bolstered by the description of the next battles as being ‘in the Linnuis region’. This is taken, on the slight similarity of names, as the Lindsey region of Lincolnshire, although no river Dubglas can be found there. Since Arthur was victorious in all his battles, the four battles on the Dubglas must have been defensive, preventing a Saxon crossing, or part of a campaign downstream. If Arthur was trying to cross the river, then by definition all the battles could not have been victorious. A slightly better location, the Lindinis region of Somerset, is examined below.