British History, 407-597, by Fabio P. Barbieri

Novels I Classical Literature I (Pseudo)-Scientific I Children's Books I Comics I Music & Movies I Various stuff
What's New I Sitemap I Arthurian articles I History of Britain, 407-597 I View guestbook I Sign guestbook I Poll I About me I Links I Search

  Vortigern Studies > Faces of Arthur > British History > Book VII > chapter 7.1

Faces of Arthur Index

Vortigen Studies Index

The Arthurian Collection is a part of Vortigern Studies



British History
click here

Chapter 7.1: The house of Vortigern

Fabio P. Barbieri

Let me sum up what we have found out about the enormously complex prehistory of the legend of Vortigern. We have a Vortigernid dynastic fiction - O - in which Vortigern and Vortimer play parts comparable with those of Conn, Art mac Conn and Cormac mac Art in the great chain of legends of the dominant Irish dynasty; it has survived in two separate versions, that of Nennius clearly slanted in Vortigern's favour, that of Geoffrey clearly negative. However, while the overall result of our analysis has been that the original must be regarded as pro-Vortigern and actually a Vortigernid family legend, there is evidence that Nennius used a source (concerned to explain the establishment of the Vortigernid fortresss in Gwynessi) which had suppressed episodes discreditable to Vortigern, preserved by Geoffrey and paralleled in Ireland.

Quite separately, we have N, an alternative account in which a different and much more political reason for the summoning of the Saxons is given - namely, that they were to serve as support for an otherwise unsupported usurper suspected of doing away with his predecessor. In the version we have, this is followed by the account of the meeting with Ronwein and Vortigern's disastrous love for her; but while the two reasons for the call to the Saxons are not wholly alternative - and Geoffrey was able to use them both, not only one after the other, but to fine dramatic effect - they certainly do not need each other, and the Irish parallel shows that the arrival of the alien woman was the beginning of the whole cycle of woes, rather than a stage of it. In other words, we have no reason to connect N and O in any way; we can read them as originally quite separate and independent, with only the central issue - the call to the barbarians - in common.

N is not only about its division between Good Guithelinus and Bad Vortigern; it is a pseudo-history of Britain. Its Guithelinus episode is the pro-Vortigernid (and pro-Maximid) reworking of an originally anti-Vortigern historical fiction, N1, in which Vortigern was blamed for every evil in the fifth century, beginning - both in time and in order - with the seduction of Constans out of his monastic vows (an act which, historically, seems to have had more resonance than anything else in the miserable career of his father Constantine III). N1 also exculpated the Senatorial nobility and even the Picts, in the tale of the murder of Constans, to lay the blame on Vortigern alone.

N also incorporates N2, a separate source welded into the longer cycle: an account of Ambrosius which seems to have at least some historical features, including possibly even the story of his death and one credible item about the historical Vitalinus, and which carries a message of acceptance of, perhaps even reconciliation with, the Saxons, which cannot be dated to any known period of Gildasian or Welsh culture. Any such account of Ambrosius (Gildas, in the sixth century, had one) must have dated close enough to his lifetime to be historical; and Ambrosius can hardly have lived beyond the end of the fifth century. If he died before his time, of illness or of poison - as hinted by the Eapa story - even this must be substantially reduced.

The difference in character with N1 is worth underlining: N1 begins with a clearly unhistorical account of an event - the seduction and murder of Constans - which happened between 407 and 411, but which is written from the point of view of someone to whom the threat of war from the Picts and the settlement and revolt of the Saxons (432-442) are already ancient history, and who does not know the relative chronology of these kings, making Vortigern succeed Constans. In other words, it can hardly be earlier than the sixth century. However, N1 is written from the viewpoint of someone who takes Ambrosiad rule as a permanent reality, and therefore cannot be later than the fall of Britain (about 570-620).

These elements are welded together to form a wholly artificial and remarkably compact narrative unity stretching from the end of Roman power in Britain to the enthronement of King Arthur, whose underlying ideology shows a strong kinship with the mind of the murderous Cadwallon of Gwynedd, recklessly and confidently trampling over the more humane views of barbarians expressed in both N1's story of the bamboozling of Constans' Picts by Vortigern and the merciful approach of N2's victorious Ambrosius. At a bare minimum, N must logically be later than N1, but earlier than P, the *Gesta Germani, which shows a quite changed mental world and set of priorities.

These accounts resurface after half a millennium in the glorious synthesis of Geoffrey of Monmouth, whose legend of Ambrosius includes plenty of quite unhistorical features of almost certainly later origin, such as the story of Tremorinus, Merlin and Stonehenge. He still has a solid chronological framework (probably derived from N), but a completely unacceptable genealogy; and he probably has grace notes of his own, which may or may not include the whole story of Eldadus and Eldol as well as the inclusion of Hengist as king of the English.

What all these Vortigern legends have in common is the calling of the Saxons, where O and N meet but for which they give different explanations. However, there are two later legends, very different from each other but both of ecclesiastical origin, which have largely or entirely forgotten the nature of Vortigern's sin. The *Gesta Germani, a much later legend of Vortigern, showing in every word and feature the fingerprints of eighth- or ninth-century Powys concerns, is interested - from an ecclesiastical point of view - in removing from the lineage of Builth and Gwrtheyrnion the stain placed on them by Vortigern's sin, symbolized by his incest (a sin which affects the purity of the bloodline more than any other), and placing it on Vortigern alone. From Brittany we have the cult of St. Gurthiern, developed from a long-lived legend of Vortigern as a saintly repentant, which might, in a sense, be seen as the other side of the coin - the protagonist expiating, alone, all the sins he has committed, whatever they were, and positively refusing any further contact with his family. This cult started in Britain, possibly in present-day Bradford-upon-Avon, but was almost certainly extinct in the island by the time Nennius wrote, and only some obscure allusions to Vortigern wandering off alone shunned by everyone seem to conserve a memory of its myth. It was however, preserved in a small abbey in Brittany, along with a text whose earliest part seems to be roughly contemporaneous with one of Nennius' own sources, the Annales Romanorum; that is, to be probably older than Nennius himself. It was, however, written in Brittany, when the legend of Gurthiern was already so rooted in the country that the saint was believed to have died in Quimperlé.

Both legends preserve fragments of a genealogy of Vortigern, which Nennius believes to stretch to the then king of Builth and Gwrtheyrnion - almost certainly his own kinsman - Fernmail. Its first names are attractive, suggesting late-Roman realities of which Nennius could not have been aware. They are specifically late-Roman; Uitalis, Uitalinus and Mauricius are typical of the later Empire and early Byzantine period, Paulus (though a good Roman name) is likeliest to be Christian in origin, and so is Pascentius, "son of the shepherd" - the latter probably a reflection of the royal rank of his father, "shepherd" of his people. And by a coincidence, the Briton Bonus of Ausonius' poem has a date not unsuitable for a man who was Vortigern's great-uncle. Ausonius' epigram is dated about 382[1]. Vortigern came to power about 428, a space of two generations.The odds are not in favour of Silvius Bonus being the Bonus of Vortigern's family tree, but it is possible: any man of wealth in late Roman Britain would also be educated and very likely to dabble in poetry and criticism - Ausonius himself was immensely rich.

It must however be admitted that the pedigree is dubious. The term for any reckoning of the Vortigernid dynasty must be not Vortigern himself, who was enthroned shortly before 429 but whose date of death is very uncertain, but Pascentius, who received his land from Ambrosius - that is, he accepted Ambrosius' overlordship, which cannot have been before 452, and probably after 465. By 834 or so, Fernmail, claiming to be his descendant, was king of Builth and Gwrtheyrnion. I do not know whether he was young or old, just come into his inheritance or long on the throne; but a simple calculation will tell us that there is a problem. Fernmail's predecessors are listed as Teudibir, Pascent II, Guoidcant, Moriud, Eldad, Edoc[2], Paul, Mepurit, Briacat and Pascent I.Taking the usual estimate of 25 years or so to a generation, this gives us about 225 years between the end of Pascent I's reign and the beginning of Fernmail's; which, even taking Pascent to have reigned to the end of the fifth century, would only get us to 725. Either some names have dropped out; or some of the kings listed had an exceptionally long reign (a British king-list of the last three centuries would have to show the enormous reigns of George III, Victoria and Elizabeth II); or the connection of this dynasty to Vortigern is adventitious.

The latter, however, seems to me rather hard to believe. Why would any later dynasty want to claim descent from the most hated man in Welsh legend? I cannot, of course, be certain that the dynasty of Gwrtheyrnion in Nennius' time was descended, father to son, from Vortigern. Nobody can prove that. But its claim to Vortigernid descent was firm and clear; more than one of its members, including Nennius himself, had tenaciously worked at whitewashing the ancestor or the line; and it had carried on into the Carolingian age a number of legends and other ideological wreckage from ancient Vortigernid claims (while the lack of any discernible properly Ambrosiad item - except for the single stray monument of Gildas - is the most remarkable feature of the transmission of culture in early Wales). Vortigern never lacked for advocates: when the dynastic legend of the femme fatale Ronwein, absolving Vortigern of blame, was darkened by the whole country's abiding hate of the man who had let in the Saxons, there was not wanting a Nennius to insist that, whatever he may have been, Vortigern was the legitimate king of Britain, and that nobody could call him an usurper, since the truthful young Ambrosius had himself acknowledged that his own father was only a consul. It is only after the extinction of the kings of Builth and Gwrtheyrnion that the anti-Vortigern line finally triumphed in the writings of Geoffrey, of William of Malmesbury, and in Arthurian romance.

It is likelier that the loss of a certain number of generations was due to a rewriting to do with dynastic struggle. Indeed, my look at the genealogy in the last book showed that it could be read as a claim to Gwrtheyrnion on behalf of the possibly intrusive son of a king of Builth, which might suggest that dynastic struggle was going on as Nennius wrote. Fernmail ipse est, qui regit modo in regionibus duabus Buelt et Guorthigirniaun - filius Teudibir. Teudibir ipse est rex Bueltiae regionis. Fernmail is the one who currently reigns in the kingdoms of Builth and Gwrtheyrnion, son of Teudibir. Teudibir is the one who is king of Builth - how can this be read except to mean that, while Fernmail was lord of Gwrtheyrnion, his father, Teudibir, still lived and ruled over Builth, perhaps jointly with his son? It was this kingdom that claimed its distinct identity - always connected, in the Celtic mind (and not only in that) with that of its founder - by Vortigern. The recital of the family descent seems therefore to have a political edge, showing that the son of the king of Builth, descended from Gwrtheyrn Gwrtheneu, is by right king of Gwrtheyrnion. It may also be for this reason that we are told that Pascent ruled over Builth and Gwrtheyrnion by Ambrosius' gift: a double validation, with Ambrosius, the most illustrious king of old, granting the Pascentiads, not just to find land of their own - as Vortigern's search for Gwynessi seems to say - but a specific territory, and that territory none other than Builth and Gwrtheyrnion, to which the Pascentiads of Nennius' time seemed so eager to lay claim.

A great break in the fortunes of the house took place with Vortigern. This may seem a facile assertion, but it is borne out by the nature of the Nennian genealogy, not unsupported by the shreds preserved in Quimperlé. Only two generations are registered before Vortigern: his father Vitalis and his grandfather Vitalinus, with his three brothers Bonus, Mauron and Paul. This is as much as may be remembered by the living memory of an uprooted family with no preserved written records, as ordinary daily experience will show: in other words, the memory of the father, grandfather and great-uncles of the usurper are the sort of thing that would be handed down by word of mouth once a great disaster has removed the records. We may identify this great disaster either with the revolt of the Saxons, during which towns and villas indubitably went up in smoke along with the records they contained; or with the return of Ambrosius, whom we strongly suspect of having burned Vitalinus/Vortigern to death in his fortress - of course, any such conflagration would be likely to destroy much of the usurper's papers. It would be especially ruinous if it was followed by the expulsion of the usurper's survivors, scattering to the four winds with little or nothing to show for their association with the fallen king. This agrees with the psychology we have seen both in Gildas' reference to the superbus tyrannus and in the survival of the cult of the hidden, penitent Gurthiern, both of them speaking of an enormous and sudden mental dislocation associated with his fall, such as left both the educated classes and the populace struggling to comprehend what had happened; it is at the same juncture that we find a Roman family which had produced an emperor - whether or not usurping - left with nothing but the remembered names of the previous two generations.

The history of the family really begins then.There is no obvious reason to doubt that Ambrosius had granted one of them, by name Pascentius, some lands in the north, in the region of Gwynessi, where their royal fortress was later known as Caer Gwrtheyrn. A gloss in some of the Cambridge group of manuscripts of Nennius identifies this with the hill-fort of Old Carlisle, on a Roman military road south of Wigton in Cumberland, and there is no need to doubt the identification, though already by Nennius' time the legend and place-name had been shifted - like many other northern legends - to Wales, and identified with a site on the Tovey, or, according to Geoffrey, with Little Doward in Monmouthshire. (Both these locations have to do with the particular concerns of their authors: the author of the *Gesta Germani, obsessed with Powysian politics, probably found it attractive to place the cursed spot of the wicked king's destruction in the unfriendly territories of Dyved, and Geoffrey, whose picture of Wales is for most practical purposes restricted to Gwent and Glamorgan, would of course place his fall there.) This agrees with the other notice, that Ambrosius executed the leader of the Saxons but then accepted submission of his followers, and with his traditionally “moderate”/”modest” character: it would seem to be his modus operandi to execute the principals but forgive their followers. Pascentius may or may not be the son of Vitalinus: the fact that he follows him in the pedigree only means that later ages regarded him as his legal heir.

Four centuries later, we find their supposed descendants in Powys. Powys as an entity is unknown to Roman Britain, and must have come into existence some time between the end of Roman power in Britain and its appearance in the *Gesta Germani in the eighth century, in which it is seen as an ancient and enduring reality. I think however that we can be a good deal more precise than that. Scholars agree that the name of Powys derives from a Latin word, pagenses, that is "men from the villages", from pagus, village. This is no more than the Latin translation of a well-known Celtic tribal name, Atrebates, which is etymologically close to the famous Breton, Welsh and Cornish root word tre- or tref, i.e. "village": the Atrebates, that is, are the "villagers". History records two tribes called Atrebates, one at the edge of Continental Belgic Gaul, placed between the Belgic Nervi (a famously aggressive and militant tribe) and non-Belgic tribes such as the Bellovaci and Suessiones; and the other in Britain, again placed in a border position between the conquering Catuvellauni and an unconquered area of Dobunni and Durotriges[3].

The suspicion that these kingdoms of "villagers" may have something to do with the borders of powerful conquering tribal confederacies is strengthened by an Irish fact: there was a tribe in Ulster called the Airgialla, which stood practically in the middle between two rival conquering confederacies - Ulster proper, the Ulaid, and the great Ui Neill group of kingdoms. The peculiarity of the Airgialla is that their name comes from their social position rather than from any legend of founding heroes: Airgialla means "hostage-givers", and, like "villagers", refers to a plural number of people in something of a subject position. A "villager" is any person residing in a settlement. This must imply an inferior position as compared with the Man in the Big House, the tigernos, let alone the Man of the Country, the gwledig or high king; and a tribal name referring to hostage-giving must place the tribe concerned in an inferior position as opposed, not only to a high king, but to any king - "he is no king who has no hostages in chains", says an Irish legal adage[4]. In other words, here we have two tribal names which imply inferiority to the whole royal level of society for all of its members; properly speaking, these tribes could have no king of their own race. (This raises an echo in the Gildasian-age idea of Britain as a land for Britons to live in but for Romans to rule.) And Atrebates and Airgialla were at the edge of powerful tribal confederacies.

What this suggests is that Powys came to exist as the result of the conquering activities of a Celtic tribal confederacy, and represented its effective outer border; but - and this is important - this Celtic confederacy must have used, not a Celtic language, but Latin, as the vehicle for its obviously Celtic social ideas, since the name of Powys is not Celtic but Latin in origin.

In the light of all that we have learned so far, this absolutely demands to be read as Gildasian. It was in the age of Gildas, and in no other known period of Insular history, that Celtic political ideas were expressed in flowing Latin. Earlier, the Britanniae were simply a Roman country with Roman political ideas, who would not have found it easy to invent a kingdom of "villagers" - a word with no Roman political connotations; later, however much Latin the British and Irish monastic classes might know, their legal and political ideas came to be expressed in native idioms. Powys, therefore, is a Gildasian-age creation. Its founding hero Cadell was known as a historical figure to Taliesin, who claimed that his patron Cynan Garwyn was of his line[5]. He is not likely, in spite of his name Roman (Catellius) to have been a contemporary figures such as Vitalinus/Vortigern, Faustus and Germanus. Therefore he was inserted in the Vortigernid-Pascentiad legend of St.Germanus for a specific reason.

The Catellids themselves claimed descent from Vortigern through a quite different line: Cadell's father Britu was meant to be Vortigern's son from his wife Sevira (Seuera?), whose Roman name may be historical though the idea that she was Maximus' daughter certainly is not[6]. It simply represents the typical Welsh doctrine that, as Dumville puts it, "Roman power had ended with the death of Maximus in 388, and that from the descendants of this last British emperor all legitimate post-Roman power flowed."[7] In spite of the fact that "this gave some of the Northern heroes Maximid ancestry", there is no trace of this doctrine in the better Northern genealogies, who trace their ancestry either back to mysterious names such as Fer and Cursalem, or, through the much-loved Gododdin ancestor Coel, to fabulous antiquities and the inevitable Beli[8].

Now whatever the ancestry of Cadell, Taliesin’s mention - which is unlikely to be later than the 580s - shows that the Catellids were lords of Powys since before the fall of Britain. The Vortigernid-Pascentiads, on the other land, were certainly late arrivals. Their legend shows that their original lands were in the north, probably in Cumberland. The Vortigernid-Pascentiads did not claim Maximid ancestry; the Vortigernid-Brituid dynasty of Cadell, did.  ( did, for their Saint, the Gurthiern cult in Brittany.) This indicates that these two dynasties, in spite of the common ancestor, came to formulate their doctrines of kingship under different influences.

To judge by their dynastic legend O, the Pascentiads were already thoroughly Celtified by the sixth century, or at least they had loyal Celtic storytellers[9] at their court. Yet the political doctrine expressed by their Nennian pedigree does not include any descent from Maximus. This does not reflect the views of Nennius himself, who was quite convinced that "from the descendants of [Maximus] all legitimate post-Roman power flowed". As Dumville rightly observes, Nennius "ends his Roman history with Maximus [in fact, Maximianus, but the duplication is evident] and turns at once to his British successors... Cunedda [and] Gwrtheyrn” (Vortigern). As I am confident that Nennius himself was a Vortigernid - to be precise: a Vortigernid-Pascentiad - what this shows is that by his time the house of Builth and Gwrtheyrnion had completely accepted normal Welsh assumptions about the primacy of Maximus. The more interesting, then, that the pedigree trotted out by Nennius himself (and partially confirmed by the Quimperlé document and by one of the Morgan ab Owain genealogies) should not conform. Its arch-ancestor is one Gloiu, whom another Northern-originated notice (in Cullhwch and Olwen) describes as the son of a mysterious Casnar Wledig.

And just as the two dynastic legends bespeak separate theories of monarchy, O and N bespeak different narrative literary cultures. N, like its predecessor N1, is a brilliant historical romance, built on Celtic values and categories indeed, but presupposing and making inventive use of existing historical records. It is very free in its use of its material, and tends to form entirely new stories - like a modern historical novelist; and it is interesting that two writers out to prove exactly opposite points and, in fact, to contradict each other, should share a similar method and similar daring[10]. It is N, not O, that decidedly incorporates the Maximus-derived idea of British sovereignty, shifting the blame for the disasters that followed its end from Maxim(ian)us to Gratianus mancipalis; and, though he makes no use of his historical material, its author has read Gildas. O knows nothing of the Maximus theory; shows no clear evidence of knowing Gildas; and arises from a storytelling culture which applies a pre-existing account to explain historical events, even at the price of leaving in some incompatible material (as I pointed out in the case of the druids' supposed unwillingness to tell the king the truth).

The author of O does not seem to have been naive: to the contrary - as I pointed out - comparison between the divergent versions of Geoffrey and Nennius shows that there was a very early written Latin version, making sophisticated and inventive use of two important Gospel quotations, Mark 6.23 and John 13.27. The difference is therefore not in literary sophistication, but in different attitudes to the past and to the art of storytelling. N and N1 live in a world of written history, of argument and counter-argument, in which data can be interpreted and reinterpreted according to different ideological or dynastic views; O lives in a world which sees the past in terms of a certain number of accepted truths, expressed in stories which can be used to understand otherwise incomprehensible events. The culture of O is not in the habit of considering historical data as independent items to be drawn into a web of interpretation; in other words, in my view the culture of O is oral at heart. It has not really come to terms with the implications of written preservation and manipulation of stories. (P, the *Gesta Germani, has a great deal more in common with O than with N and N1, even though it does seem to have integrated a certain amount of historical information within its basic mythological shape.) Neither N nor O are very close to what we would today call history, but in the case of O the reason is the habit of applying more or less ready-made categories, expressed in the form of wholly unhistorical narratives, to known personages; in the case of N and N1, it is the grave shortage of proper and connected historical accounts (when N had one at hand, that of Ambrosius, he made use of it), that left too much space to inventive rewriting, argument and point-scoring. Apart from our great historical novelists, such as Mary Renault, there is a close parallel to this sort of thing in China, where the mighty Romance of the three kingdoms is generally taken to be a largely fictional reconstruction of largely lost history, incorporating a certain amount of fortuitously surviving data.

It follows that the two different narratives arise from two different environments; and we can identify them with some confidence - the northern world of Pascent's kingdom of Gwynessi, and the Welsh one of Cadell's Powys (which, before the fall of Britain in 570-620, looked to the lowlands through the valley of the Severn). We can take the association of Pascent with Gwrtheyrnion to be late; the dynasty spent some time in Gwynessi in the North. This could also explain the mess in the pedigree: if we take some at least of the Gwynessi generations to have been forgotten, with only the earliest generations from Gloiu to Pascent being remembered with any precision, then the "generation gap" no longer looms so large.

My point is that there is nothing impossible or unlikely in the theory that Pascent's descendants, after having been ensconced in Gwynessi for a few generations - the date of their establishment there is beyond conjecture, but they evidently were there when the first version of O was written down - eventually ended up in a remote Welsh highland tract, still cherishing the title of kings and their royal descent, still determined - as the writers of the *Gesta Germani and the Historia Brittonum show - to whitewash, if not Vortigern's moral character (which was by then impossible) at least his legitimacy. The dynasty, which fragmentary Nennian allusions and glosses place in the north (ad sinistralem plagam) and near Carlisle, turns up after the fall of the north (though in a time when Strathclyde still held on) in a tiny kingdom subsidiary to Powys, which is ruled by another dynasty claiming Vortigernid descent. It looks as though the two royal houses, beleaguered by the overwhelming power of Northumbria, had managed to appeal to the better natures of their distant kinsmen in their still-defended mountain fastnesses.

Now, there are a number of reasons to believe that, in the wake of Saxon conquest, Wales may have become the refuge of a number of dispossessed northern dynasties.

I have already suggested that the house of Maelgwn ended up in Gwynedd not because it was their ancestral property but because they were driven from Maelgwn's conquests in the North; but among their dynastic fictions there nests another probable case of a refugee dynasty, that of Ceredig. Dumville has analyzed the king-lists of Cunedda's supposed descendants, and found that the list which begins with his supposed son Ceredig is missing at least two, perhaps three generations. His conclusions are eloquent: "On the evidence available, Ceredig of Ceredigion may with modest confidence be assigned to the sixth century... It is clear from the comparison of the pedigree of king Gwgon [a descendant of Ceredig] with those for the other kingdoms of Greater Gwynedd, that it has a different origin and history". In other words, the relationship of Cunedda with Ceredig is a genealogical fiction, intended to legitimize a dynasty intruded into Greater Gwynedd, where legitimate royal power was held to spring from Cunedda. It is an interesting coincidence that in the mid-610s one Ceredig, king of Elmet, is driven out of his kingdom, or possibly killed[11] completing the English conquest of Northumbria. The coincidence is by no means perfect: the date for Ceredig's expulsion or death is about a generation later than that suggested by Dumville - if only "with modest confidence" - for the death of Ceredig of Ceredigion. But Dumville relied for his dating on a generation-length of twenty-five years, by no means a safe measurement in that age of uncertainty and violence. On the other hand, while Dumville's chronological calculations do not exclude that his Ceredig should have lived on into the six hundreds, they certainly do exclude his traditional dating, which would demand a generation-length of 37.5 years - completely unimaginable in the circumstances.

What we have, therefore, is the end of a powerful northern dynasty (Taliesin, writing one or two generations before the end of the North, makes Gwallawg of Elmet an ordained gwledig - ordained, here, must mean anointed - whose field of action is the whole island) in a period compatible with the establishment of a new kingdom in Gwynedd, whose founding hero has the same name as the last king of Elmet. There is no record or suggestion that Ceredig was an interloper; but the sources seem to conspire to confer on him a Cuneddan legitimacy to which he obviously wasn't entitled. This would make it the third dynasty to come to Wales from outside, after Maelgwn's and Vortigern's. It does not seem casual; and it would be no more than sense to look for a remote, sheltered hiding-place when the English came a-calling.

We do have a genealogy of the lords of Elmet showing their descent from Coel and a relationship with Cunedda through his marriage with Coel's "daughter" Gwawl. It stops at Gwallawg and (implicitly) at his son Ceredig, last king of Elmet, and knows nothing of any descent in Wales; but it almost certainly comes from the north, in particular from Strathclyde, and has every reason to have lost contact with more southern realities when Elmet fell in 616, and even more when the Gododdin ceased to exist probably about 638. In other words, we probably have two separate documents about the house of Ceredig; one, of northern origin and some historical value, up to its expulsion from Elmet; the other, Welsh in origin and valuable only from Ceredig himself on, from their arrival in Ceredigion. It is even possible that Ceredig himself may never have seen the land named for him: if, as the Annales Kambriae seem to indicate, he died in the fall of Elmet, it may be that his descendants carried his memory with them to their land of exile as that of a patriarch, in which case his supposed reign and death in Ceredigion would be a notice of the same order as the supposed death of Gurthiern in Quimperlé.

The North, we must remember, fell to the English later than the South: the seismic collapse of central and southern England (as it is today) began, in my view, a few years after Gildas completed his masterpiece about 561, and was complete by Augustine's arrival in 597 - indeed, by the time Gregory the Great started planning the English venture, in the early 590s, showing that he had lost any confidence in the possibility of a British recovery. Gregory's own well-known passage from his commentary on Job[12], describing in allegorical terms the conversion of the English, is heavy with the recent terror of a sudden and apparently irresistible wave of conquest, which, he seems to be saying, had been going on right up to the minute when Augustine reached Kent - but which is now at rest. But Chester only fell in 606 or so, and Elmet as much as ten years later. It is even possible that the northern lords may have seen the disaster coming and prepared some bolt-holes in advance; this would explain why all the intrusive dynasties in question seem to be from the North - the South did not have the chance to escape its doom.

Two centuries later, the situation seems to have changed. There is a visible discrepancy between the *Gesta Germani and the Powys genealogy: in the former, Cadell and Vortigern are contemporaries, while in the latter Cadell is the grandson of Vortigern. And from what we have seen, the latter is decidedly the more credible dating: Cadell's Powys belongs in the Latin-speaking resurgent Celticism of the sixth century, not in the Roman Britanniae of Germanus, Vitalinus and Faustus. Whatever drop of historical data there may be in the bucket of legend of the *Gesta Germani, it does not include the presence of Cadell. In fact, I argued, it seems calculated to put the Catellids in their place, kings of Powys indeed, but descended from a servant. It also denies the authenticity of the Catellid pedigree - proudly raised by an eighth-century Catellid king in the Pillar of Eliseg, an unusual gesture for a Dark Age Welsh lord - and the common Vortigernid blood. Only the house of Pascent, ruling the land named for the ancestor, are truly descended from him. Was the Pillar of Eliseg something of a Catellid polemic in stone, against such insinuations?

Now the Catellid claim to Vortigernid and Maximid ancestry is probably quite early, since N - as I pointed out again and again - seems bent on clearing both Maximus and Vitalinus/Vortigern of blame for the misfortunes of Britain, and N is most probably to be dated to the time of Cadwallon of Gwynedd, that is the second quarter of the seventh century; this argues that there was already then a royal house - allied with Cadwallon - who claimed descent from both Maximus and Vortigern, and the only such house known is that of Powys. (Being squeezed between Gwynedd and Mercia - ruled by Cadwallon's great ally Penda - Powys probably could not opt out of an alliance with these two powerful "friends" even if it wanted to; and to judge by N's apparent enthusiasm for Cadwallon's exterminating policy - an enthusiasm apparently shared by all those Britons who formed Cadwallon's "huge" army - they may not even have wanted to.) In other words, the Powysian claim to Vortigernid descent was already in place by about 630, a long time before the author of the *Gesta Germani tried to demolish it. It seems that by his time the two royal houses were not on the best of terms. I already suggested that the relationship of Nennius with king Mervyn's Gwynedd suggested an alliance against Powys, and this consideration strengthens it.

I have done enough to explain the effect of Vortigernid-Pascentiad ideas, though Nennius, on my material; I have nothing more to say. But I do have one parallel in mind for the status to which they were reduced by the age of Charlemagne. Their pathetic position in the rocks of Builth and the headwaters of the Severn, and their desperate clinging to ghostly ancient claims, has a rather touching echo in Ireland. In the words of Professor F.J.Byrne, "time and again it was the fate of dynasties who lost their hold on these plains [the valleys of the Liffey, Barrow and Slaney, centres of ancient Leinster] to retreat eastwards across the mountains into political impotence: the Ui Garrchon and the Ui Enechglass in the sixth century, and the Ui Mail in the eighth, were to be followed after the Anglo-Norman invasion by their supplanters, the O'Tooles and O'Byrnes of Ui Duinlange... Even in the seventeenth century a sept of the Kavenaghs, claiming to be the heirs of the twelfth-century Mac Murchada kings of Ui Cheinnsselaig and Leinster, and living at Garryhill in county Carlow, 'betaking themselves to that place for securitie it being a place of fastness and compassed with very large woods and bogs', maintained a forlorn relic of their royal dignity. 'This House of Garrchoyle for a testimony that they were the eldest of the Kavenaghs and descended from the stock of the Kings of Leinster had a great seate and a vessell or cup to drinke out of called Corne-cam-more.'[13]"

These Kavanaghs had even more in common with the house of Vortigern than other dispossessed east coast families, since Diarmait Mac Murchada, calling in Strongbow and precipitating Henry II's intervention, had a thoroughly Vortigern-like role in the history of Ireland[14]; and their survival in the desolate corners of Carlow is exactly parallel to the survival of the house of Vortigern in the mountains of Radnor, at the extreme limits of the Severn area on which their power had originally been based. Builth is an isolated area, walled in by high mountains and unable to affect the rest of Wales or England; Gwrtheyrnion, though less completely fenced-in, represents the extreme or fag end of the Severn lands, barely worth forming a lordship out of, and shut off from the greater Severn valley by the sizeable obstacle of the kingdom of Powys. Maps of prehistoric Welsh settlements show few or no fortifications in this area, and, for what it's worth, it was this one sliver of worthless land that had received the name of the "usurper", as if in a further act of damnatio memoriae. It is possible that, hemmed in everywhere else, they may have expanded into Builth as the only direction in which they could exert any influence. If I have read the origin of Ceredigion correctly, it is worth comparing the allotment of the valuable area of Cardiganshire with the unpleasing land allotment and evident isolation of the Vortigernids of central Wales. The Catellids had Cadell himself - praised, to the exclusion of Vortigern, as the very example of a mighty warrior, in Taliesin's ode to Cynan Garwyn - to resort to as a great ancestor; the Pascentiads had only Vortigern; by the time of Geoffrey, the poisonous nature of his image had blackened even the name of Pascent.


[1]SNYDER op.cit., pp.72-3.

[2] These two names, Eldad and Edoc, are suspiciously similar to the two collaborators of Ambrosius in Geoffrey, Eldol and Eldadus. The relationship between these two probably historical Vortigernids and the two heroic figures of Geoffrey is very difficult to assess. In my view, Geoffrey’s Eldadus and Eldol represent an unhistorical reduplication of the two aspects of the historical St.Illtyd, first a warrior, then a monastic teacher; who lived in the second half of the sixth century (that is, two or three generations after Ambrosius). The account of Illtyd was preserved chiefly in Brittany (where he was remembered as the great teacher) and in his own monastery of Llanilltyd Fawr; it seems possible that a legend had formed elsewhere in Wales, displacing the Saint in time and attributing him to Emrys/Ambrosius. Geoffrey’s fervid systematizing imagination would then have done the rest. Such reduplications of Saints are not unknown in Ireland, and there is no reason not to suspect them in Wales; and it means that we should be careful (as if we weren’t already) of Welsh accounts of Saints.

[3]Whether the Catuvellauni or any of their enemies were properly Belgic does not concern us here; however, if they were, it would make the similarity of name and role between the Continental and British Atrebates more significant. By the same token, I don’t believe that the Continental Atrebates and the British ones were ethnically related; it was their political position that gave rise to their name. If Commius Atrebates, first Caesar’s ally and later his foe, became lord of the British Atrebates after his flight from Gaul - as coins struck in his name seem to show - it was not because of ethnic kinship, but because he had fulfilled a similar role in the Continental kingdom. After all, the fact that he seems to have been intruded on the British Atrebates even from outside the island, does seem to show that the Atrebates were the sort of tribe who could have a foreign king imposed on them.

[4] BYRNE, op.cit.31.

[5] MEIRION PENNAR, Taliesin Poems, 45.

[6]The pillar of Eliseg, integrated by one of the many fantastic genealogies of Morgan ab Owain in the Jesus College 20 manuscript, give this sequence: P.C.BARTRUM, Early Welsh genealogical tracts, cit.

[7]DUMVILLE, Historical value op.cit., 11.

[8]MOLLY MILLER, Pedigrees op.cit., 259-271.

[9]Throughout this study, I have not dared to identify the "storytellers" I frequently mention, with the well-known historical class of Welsh bards/poets. That is because I have no direct evidence that the tradition of extended storytelling which I have identified in several places can be directly attributed to them. I have, however, no doubt whatsoever that this is the case.

[10]It is perhaps significant that neither of them seems to know anything of the pseudo-history of A, with its northern origin.

[11]Possibly by Edwin of Northumbria; but the Annales Kambriae clearly separate the two events, dating Ceredig’s death one year before Edwin's conquest of the throne of Northumbria. Since Edwin was an exile in East Anglia until his host Redwald went to war on his behalf and put him on the throne over his enemies' dead bodies, what the Annales seem to be saying is that Edwin cannot possibly have been the conqueror of Elmet: his enemy Aethelfrid should be ascribed it.

[12]GREGORY THE GREAT, Moralia in Iob, 27.11.20-21, quoted by BEDE, HIstory 2.1: Ecce quondam tumidus, iam substratus pedibus sanctorum seruit Oceanus; eiusque barbaros motus, quos terreni principes edomare ferro nequiuerant, hos formidine Dei sacerdotum ora simplicibus uerbis ligant; et qui cateruas pugnantium infidelis nequaquam metuerat, iam nunc fidelis humilium linguam timet. "Indeed, look how that once enraged ocean now lies down and serves the feet of the Saints; and those very barbarous energies which earthly lords could not beat down with iron, now by the fear of God are bound by the simple words of priests; and the infidel who never feared any horde of fighting men, is now a believer and fears the tongue of humble men". One must admit that there was a certain amount of wishful thinking in this.

[13]F.J.BYRNE, Irish Kings and High-Kings, 1973, pp 130, 153.

[14]The coincidence that the Kavenaghs and the lords of Builth and Gwerthrynion were both descended from a notorious figure of usurper who let in aliens to support him is probably less than casual: Celtic legend knew such figures, and therefore the cultural categories to fit Vortigern and Diarmait were ready to hand. To one such traitor, Geoffrey's Arviragus, Welsh tradition attributed the coming of Caesar to Britain; and it is a curious but certain fact that the historical Caesar was almost certainly called to Britain by a British prince called Mandubracius, pursuing a personal feud against the greatest king of the period, Cassiuellaunus. Had the great conqueror killed Cassiuellaunus and placed Mandubracius in his place, instead of forcing on them a precarious peace, the correspondence between reality and myth would have been perfect. Rachel Bromwich believes Arviragus to be derived, at several removes, from Mandubracius; but as there certainly was a Welsh legend of Caesar (Ulkessar) from which Geoffrey drew with both hands, and as she cannot explain the Welsh "translation" of Arviragus as Afarwy, I would rather believe that Geoffrey has, more suo, Classicized a native tradition.

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

Comments to: Fabio P. Barbieri

VortigernStudies is copyright © Robert Vermaat 1999-2007. All rights reserved