The Arthurian Collection is a part of Vortigern Studies

Against everyone who denies the existence of a historical King Arthur - and, till I undertook this investigation in earnest, that included me; the more so for the experience of reading John Morris - is cast Chesterton's wonderful poetic curse:

O learned man who never learned to learn

Save to deduce, by timid steps and small,

From towering smoke that fire can never burn

And from tall tales that men were never tall;

Say, have you thought what manner of man it is

Of whom men say, He could strike giants down?

Or what strong memories, over time's abyss

Bore up the pomp of Camelot and the crown?

And why one banner all the background fills

Beyond the pageants of so many spears,

And by what witchery in the western hills

A throne stands empty for a thousand years?

- Who hold, unheeding of this immense impact,

Immortal story for a mortal sin,

Lest human fable touch historic fact

Chase myths like moths, and fight them with a pin.

Take comfort, rest; there needs not this ado;

You shall not be a myth, I promise you.

As with everything that Chesterton wrote, here is a bewildering mixture of keen and astonishingly penetrating intelligence and of ill-judged, unwise, indeed dangerous populism. The statement of what Dumville called the "no-smoke-without-fire theory" is as bald as it gets, and there are any amount of reasons for historians to treat it with anything from nervousness to irritation; but once all the allowances have been made, and all the ruinous effects of the no-smoke-without-fire theory on every historical writer from Geoffrey to Morris pointed out, there still remains one basic fact, which all my research has done nothing but throw into starker and starker relief: that one name, and one name alone, emerges from all the lost history between 470 and 540, between Ambrosius and Gildas: and that name is Arthur.

They certainly haven’t waited for me to make this observation; and it would not be important at all, if we had gone on believing that no credible record of the period ever existed. But now that we have reason so to believe - that an abundance of lost sources has been teased out of Gildas, Nennius and Geoffrey, some of which show good reason to be seen as contemporary - the persistence of the name Arthur, of the galaxy of surrounding figures - Cei, Bedwyr, Gwalchmai, Gwennhwyvar, Uther Pendragon, and so on - and of no unrelated figures, is a significant fact. Also, there is a considerable amount of overlapping. L is still known to Nennius, but the only thing he places where L should be is a list of Arthur's twelve battles. Two centuries earlier, in the time of Cadwallon of Gwynedd, the inventor of the genealogical fiction of the "House of Constantine" sees Constantine (I)II, Ambrosius, and Arthur, as the peaks of post-Roman British history; in other words, N's author regarded Arthur and his father Uther as historical figures on the same grounds as Constantine and Ambrosius.

Dating from Cadwallon’s time, N is contemporary, perhaps some decades earlier, than the first composition of the Gododdin^[1], with its one mention of the hero; it antedates by more than fifty years the insertion of Arthur’s names in a praise-poem for Geraint of Dumnonia (died 711), where he already appears as an "Emperor, ruler of our labour". What is much more, N is barely 100 literate years after Arthur’s own supposed lifetime. If a seventh-century Welsh savant knew of the historical Constantine, Constans and Ambrosius, then he might a fortiori know of a still later Arthur! And “…a further allusion to Arthur in an early poem… would be almost as significant as that in the Gododdin if only its textual authority were more securely established. An elegy for Cynddylan, a seventh-century prince of Powys, has survived only in copies from the seventeenth century and later, but it is composed in a language, idiom and metre closely resembling that of the Gododdin, while its text has orthographical features which suggest that it was copied from a manuscript as old as the thirteenth century Book of Aneirin [the chief Gododdin manuscript].  One of its lines describes Cynddylan and his brothers as canawon Artur fras dinas dengin, “whelps of stout Arthur, a strong fortress.”^[2]  Personally, I am not as cautious as the authors of this passage: if waddles like a duck and it quacks like a duck, then by Heaven it is a duck.

Again, Nora K.Chadwick, Rachel Bromwich and John Morris – quite a diverse crowd! - agree that, as Morris puts it, "The name of Arthur was given to the sons of Aidan of Dal Riata and his son Conang; Peter of Demetia; Pabo of the Pennines; by Coscrach of Leinster; and by Bicoir the Briton, probably of the Clyde... all these children were born and named in the mid or late 6th century; no other child is known to have been named Arthur for 500 years, until after the diffusion of the... romances...".  This relies on documents of various dates, some of which may have used the name Arthur - as the Fenian cycle indubitably did^[3] - as a typical British name, and therefore be under the influence of unrecorded but clearly burgeoning British legend.  Nevertheless, it is a fact that the Arthur who was a son of Aidan was recorded by Adomnan in the early seven hundreds, in an account generally taken to be founded in history.  There may, perhaps, be smoke without fire; but there certainly is an awful lot of smoke.

Let us therefore look at the possibility that L may have had a historical Arthur as its protagonist, and that it may be found in Geoffrey's account of Arthur, as N2 turns up in his account of Ambrosius.  We can begin with Nennius; who, writing over a century after Adomnan and the Geraint poet, only presents Arthur - apart from a couple of allusions, in his final chapters, to the marks left by the hero on the very landscape of the country - in a list of twelve battles, all fought against the Saxons and fought, so far as anyone can make out (the place-names are very obscure) in Britain only.  We have seen that Nennius’ purpose is to collect, in a coherent chronological framework, obscure historical notices (written and in Latin) from outside what I have called the Welsh historical collection.  Now it is quite clear that Nennius knew a good deal more about Arthur than he said; it obviously follows that - as with the Nennian allusions to L, to Gildas, to Bede, to the Discordia Guitolini et Ambrosii - Arthur featured prominently somewhere in the Welsh historical collection.

Now Geoffrey's account of Arthur’s youthful deeds covers some of the same ground as Nennius' battle-list, but some of the battles are omitted or left unnamed – Geoffrey seems to mention only nine - and we notice that the hyperbolic number of fallen at Arthur's hands at Badon is different in the two writers - 960 in Nennius, but only 470 in Geoffrey.  In other words, neither the list of Nennius’ twelve battles nor the description of its climax at Badon Hill can be directly connected with Geoffrey’s account.  Geoffrey did not copy the sequence from Nennius; he had an independent source – obviously from the “Welsh historical collection” – and I think it is a fair inference, given that Nennius’ purpose was to collect obscure references, that he produced his chapter 56 just because the data it featured were different from the standard account.  To say, therefore, that it is the standard account that turns up in Geoffrey’s account of Arthur’s early days, is perhaps too long a leap, but not an implausible one.

The differences do not stop there.  Geoffrey also agrees with Gildas, and disagrees with Nennius, that Mount Badon was not actually the last battle in the cycle of wars.  Gildas called it almost the last, and hardly the least, of the furciferi's defeats, while Geoffrey follows it up with clashes in Scotland, after which come twelve years of peace.  Gildas clearly means that Badon was not quite the last battle against the Saxons; in Geoffrey, on the other hand, it is exactly that, and those that follow are against Picts and Scots.  Except for the barbarian component of Mordred's faithless army, Geoffrey's Arthur never has to fight a Saxon again.  This may reflect another historical tradition, speaking of some extra fighting after Badon, misunderstood by Geoffrey - or at some point before him - to mean fighting against other barbarians^[4].  There is no trace of it in Nennius 56, where Badon is the twelfth, last and greatest; but then, scholars believe the sequence of twelve battles to be formulaic, and the chapter itself to be a Latin translation of a Welsh praise poem like Taliesin's praise of Gwallawg, which also mentions exactly twelve battles.

This leads to the interesting consideration that this praise poem stops, according to Geoffrey, at the first stage of Arthur's career.  Geoffrey ascribes all the battles named by Nennius exclusively to Arthur's youth, within a broader chronology, crisply set out, and shared, to the best of my knowledge, by none of the romances.  There are three periods of war, each for substantially different objectives, and each divided by a stated number of years of peace.  Only the first of these periods has any similarity with the Nennian battle list, and in it, as in Nennius, Arthur is concerned only with Britain, which he clears of her traditional enemies, Picts and Saxons.  He does indulge in a few foreign ventures, which however do not go much beyond the natural limits of Britain: he conquers Ireland, and peacefully receives the submission of the rulers of the nearest isles^[5].  There then are twelve years of peace.  Then he sets out to conquer Norway and Gaul; the campaign climaxes in an epic duel with the heroic Roman Frollo, described as a giant, who falls at Arthur's hands.  This takes nine years; there then are five years of peace, until one Lucius, emissary of an Emperor Leo, claims tribute from the British and reparation for the invasion of Roman Gaul.  Arthur reacts as we might expect, and a great battle takes place in Gaul, in which the British triumph; but just as Arthur is about to enter the by now defenceless Empire and take it for himself, the news of Mordred’s treachery reaches him - he goes back to Britain, and the battle of Camlann takes place.  No duration is given for these final wars.

These peculiar time-reckonings do not feel like legend.  I doubt that anyone could find any symbolic or structural significance in the sequence of an indefinite period of war, twelve years of peace, nine more years of war, five of peace, and another indefinite period of war ending at Camlann.  It seems more like a contribution, from previously known data, to the calculation of King Arthur's time according to the chronological concerns of mediaeval historians.  Nor is this the fabulously aged King Arthur of the romances, riding to his last battle at the age of 98: crowned at 15, beginning his anti-Saxon campaign some time later, he would, at a guess, be in his mid-twenties by the time of his great triumph at Mount Badon, and in his early fifties when he began his last cycle of wars; hardly very old, even by the standards of his time.

The Annales Kambriae, written probably some time in the tenth century (and therefore a good 200 years earlier than Geoffrey) might be expected to form one of his sources, but in fact he owes no more to them than to Nennius.  While the Annales date Arthur and Medraut’s death at Camlann at 537, Geoffrey firmly dates it at 542 – the only date in his book to be stated explicitly rather than implied by synchronisms.  On the other hand, his date and theirs for Badon Hill may coincide, but only if one reckons an incredibly (and I mean that literally) short time for the final term of war in Geoffrey’s account.  If we date Badon Hill, "almost the last of the furciferi's defeats, and certainly not the least", to the Annales’ date of 516, then the last cycle of wars, in which Arthur conquers the Roman emperor and then falls at Camlann, begins in 516+12+9+5= 542; but it is absolutely impossible that the colossal cycle of wars which is the climax of Geoffrey’s whole work, with Arthur defeating the Romans, going back to Britain and fighting three battles with Mordred, could have taken less than a year.

Now, we have seen that Gildas’ masterpiece may be dated by its clear allusion to the Byzantine-Persian ten-year peace of 561; and that he dates the siege of Badon Hill at forty-four years and one month before, that is at 516/17.  This agrees exactly with the Annales date; but it agrees with Geoffrey’s only if we attribute no duration at all to Arthur’s final wars.

And we have to.  The strength of the agreement between Gildas and the Annales – two documents with no connection, in one of which the date can only be reckoned by techniques of interpretation which the author of the Annales cannot possibly have had – is such that Geoffrey’s dating scheme can only be accepted if it agrees with them; everything that fits with dating Badon Hill at 516/517 must be considered right; everything that does not must be wrong.

A number of interesting corollaries follow.  First: granting the synchronism, the fact that the Annales and Geoffrey share a date for Badon Hill but not for Camlann must mean that they had an ultimate common source for the former, but not for the latter.  Second, there is an overwhelming presumption that this common source is L.  Third, that the fact that the Annales place Camlann where Geoffrey places the end of Arthur’s Gallic wars and the beginning of his last five years of peace must mean something.  Fourth, that Geoffrey describes this last dated period as five years of peace means in effect that he has nothing to say about them; that is, that he knew that Arthur lived for five years after 537, but that he could not attribute any deeds to him after that.  Fifth, if the discrepancy between Annales and Geoffrey means that the date 516 but not that of Camlann came from a common source, then Geoffrey and the author of the Annales both had at least two separate sources, of which the first but not the second were common, and of which the first (that is, on the hypothesis, L) contained an account of Badon Hill but no account of Arthur’s decline and fall.

I take Geoffrey’s date very seriously indeed.  There is special authority in the fact that it is the only explicit date in the whole Historia regum Britanniae – Geoffrey loves to invent synchronisms, but never supplies definite dates; no, not even for the coming of the Saxons.  And the fact that it actually conflicts with his own account, in the sense of allowing him no space for his great final cycle of glory and doom, must mean that he had it independently of that story itself.  (The fact, incidentally, that he kept the duration of his Arthur’s last wars vague, and that he did not actually date Badon Hill at all, might suggest that he was perfectly aware of the problem, and that he deliberately slipped around it.)  The Annales date, on the other hand, tells us that their author knew that Arthur and Medrawt had died in battle at a place called Camlann, but did not know exactly when; he decided to date it at the last year in which he knew Arthur to have been active.

We have therefore reason to believe that the source for the date of Badon Hill – that is, L – had no information on the decline and fall of Arthur.  And when we look at what Gildas and Zosimus tell us of L, explicitly or implicitly, everything supports this view.  Gildas sees L as the antithesis of his own book, that is, not as the account of a collapsing society ridden with rebellious, cowardly murderers, and riding headlong towards ruin, but of a triumphant world of patriotic heroes.  And if, in his moral world, Divine wrath was the suitable and expected reward of the moral horrors of his time, so, by the same token, success must deservedly have followed “the great dangers of most valiant heroes in grim war”; in the opposition he draws between a past of disciplined heroes and a present of backstabbing traitors, there is no space for a decline-and-fall account.  As for Zosimus, the British history he sketches is one of triumphant success, in which the Celticizing nativist party which expels the “Roman magistrates” goes on to defeat the “barbarians settled over the Rhine” not only in Britain, but also in Armorica (that is, not only Brittany but also the Loire valley from the great bend to the sea) and sundry unspecified other provinces; not only is there no hint of any decline-and-fall, but the whole context clearly excludes it, since the whole literary work he received must clearly have been celebratory rather than mournful in nature.

The items of information in Geoffrey’s account tend, it seems clear, to fall into certain groups which belong together and contradict others. The 542 date must belong together with the five final years of peace, since neither appears in the Annales: they are peculiar to Geoffrey. (Let us call this group 2.) The whole account of Arthur’s last wars against the Romans and Mordred is a separate group of items of information (let us call it group 3) which, as a whole, contradicts group 2. Therefore they are from a separate source. Geoffrey, therefore, had at least three sources: L with its account of Arthur’s wars and triumphs, the item about five years of peace before Camlann, which sounds annalistic but features a definite date unknown to the author of the Annales Kambriae; and the elaborate narrative about Arthur’s final wars. It was the insertion of this third group which made trouble – whether he knew it or not – for his chronology, by making it impossible to date Badon Hill at its correct date of 516/7. This, indeed, may well be why Geoffrey was so careful to avoid definite Anno Domini dates.

Notes

^[1]The Gododdin is a collection of elegies attributed to the bard Aneirin, lamenting a number of heroes fallen in a raid against the English of Bernicia in about 600AD. A variety of scholars and historians have argued for it being historical and possibly contemporary; take for instance T.M.Charles-Edwards’ very interesting article The authenticity of the Gododdin: a historian’s view (in BROMWICH and BRINLEY JONES, eds., Astudiaethau ar hyr hengerdd, Cardiff 1978). In one of its two manuscript versions, Arthur is mentioned as the paragon of heroic bravery, especially perhaps in defensive fighting (the warrior compared to him is also compared to “an alder palisade”). The authenticity of the line has been doubted, but, so far as I can see, on no very good ground, save that it only exists in one of the two versions.

^[2]BROMWICH, JARMAN, ROBERTS and HUWS, The Arthur of the Welsh, Cardiff 1991, introduction.

^[3]At some point in the development of the Fianna cycle, the Irish invented an Artúr son of Beine Brit (the victor of Mag Mucrama). Artúr was a Fianna member who stole Finn's wonderful dogs out of impulse; the Fianna then pursued him, beat the nonsense out of him, and from then on he remained Finn's faithful follower. The story is distantly involved in the destinies of the Fianna: it is when they pursue him to Britain that the originally horseless Fianna infantry capture the one horse and mare from whom all their 150 steeds were born. His insertion shows that the Irish knew of a Welsh hero called Arthur, and his character as an incompetent and luckless thief is probably a comment on some such story as Arthur’s invasion of Ireland in search of a magic cauldron in Cullhwch and Olwen, or his rather more this-worldly campaign in Geoffrey. That the hero is burlesqued in the most popular Irish heroic cycle shows that his name and reputation among the Welsh were well known in Ireland; and the fact that, unlike his “father” Beine Brit, he is not an object of terror but of mockery, shows that the image of the British in Ireland had radically declined between the creation of the story of the battle of Mag Mucrama and that of Artúr himself - they are no longer the gigantic power that can overrun Ireland and destroy a High King, but the thieving "Taffy" of the rhyme, who "stole a side of beef". In other words, the legend of Mag Mucrama cannot have taken form after the fall of Britain to the English, nor the legend of Artúr before.

^[4]On the other hand, this agrees with Gildas in that Geoffrey’s Arthur enjoys a long career after Badon, in which the Saxons do not feature; just as Gildas regards the Saxons as completely defeated and pacified shortly after Badon and up to his own day. We must realize that nobody inventing a romance about sixth-century Britain, one or more centuries after the events, would conceive of the Saxons, the island's once and future conquerors, being defeated and inactive for decades.

^[5]Orkney, Iceland, and Gotland. The latter two are obviously Geoffrey’s own exaggerations; at the time, no human being had ever even set foot in Iceland. He is probably expanding the traditional description of a king of all Britain reported by Nennius: iudicabit Britanniam cum tribus insulis, he judged [ruled over] Britain with the three islands. Mention of Gotland, however, may be significant in other ways, as we will see.

History of Britain, 407-597 is copyright © 2002, Fabio P. Barbieri. Used with permission.

Comments to: Fabio P. Barbieri