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 > Introduction

Guest Author:
Fabio P. BarbieriHISTORY OF BRITAIN, 407-597, by Fabio P. Barbieri

Fabio P. Barbieri (1962) has a degree in Religious Studies and Social Anthropology from the School of Oriental and African Studies, University of London, but did always love mythology. His first book is on the subject of comparative mythology: Gods of the West I: Indiges, (Societe' Belge d'Etudes Celtiques - Series of Memoirs, 1998). This work on the Arthurian epic and history is his current project.

Faces of Arthur Index

Vortigen Studies Index

The Arthurian Collection is a part of Vortigern Studies



British History
click here

Introduction: Sources and their Treatment

Fabio P. Barbieri

This book is dedicated to Donatella Barbieri and her family for being good people, and to Debbie*Wallace.

Darkness lies over the earliest English and Welsh history.  One awesome cry of pain, the polemic of the British Saint Gildas, comes from a time shortly before the English conquest; one Welsh monastic writer, three centuries after him, makes a compilation of a few Latin sources known to him, themselves mostly legendary and, when they are not, Irish and English in origin, and does his best to make sense of them; his name might, or might not, have been Nennius[1].  And that is all. On the English side, little exists before the tremendous phenomenon of Bede: a great genius indeed, but more than two centuries removed from the birth of his people, and, like Nennius[2], either unable or unwilling to consider any source not in Latin for the earliest histories of Britons and English. A scanty few Greek and Continental sources supply a scanty few dates; the Anglo-Saxon Chronicle, even later than Bede, recovers a few tribal traditions he had not seen fit to take down; hagiographic legend mentions saints and kings, with what credibility it is often hard to assess. (One such hagiography, that of Saint Kentigern/Mungo, may have a few things to tell us.).

But from the moment the Roman Emperor Honorius, caught between several rocks and hard places, gave the local administration of Britain (ciuitates) unprecedented leave to organize their own defence without imperial authority - an act that was widely interpreted as a grant of independence - no coherent historical narrative exists. The curtain falls over Britain, still a Latin country and a Roman province, in 410AD; and it rises again with the arrival of the first Archbishop of Canterbury, to a country divided between Teutons and Celts, in 597. Even then, historical narrative only reaches the islands in stages, along with the return of the Roman church. Coherent accounts are scarce for Wales and the Welsh-speaking lands (Brittany, Cornwall, Cumbria, Strathclyde) before the ninth, the tenth, even the eleventh century.

What is left then? Archaeology, of course; but more importantly, a considerable literary heritage. Nennius, Bede, Gildas, Saints’ lives, chronicles, bardic poetry, bardic legend, and Geoffrey of Monmouth’s majestic reworking of mostly lost lore: these may, individually and collectively, be dubious, dangerous or surely false as direct historical narrative - but they are literature. Therefore the tools of literary scholarship may be used on them, and - so long as we treat them for what they are, and not allow ourselves to take for a direct historical source what is clearly not - do a lot towards discovering actual historical facts or developments. Sources may be elucidated, showing which parts of a narration are the author’s own work and which the work of others; discrepancies discovered, indicating that the author has misunderstood his source, therefore discovering important facts about both. Authors’ views of their past may be clarified; so may the views of their sources, showing how the former and the latter differed, and what the difference has to tell us about intervening events. This can show us the social and ideological assumptions held by the various authors, and discover their personal feelings about their time and people. They may, as a work of supererogation, suggest actual and definite events, or even give unexpected credibility to the odd fragment of legend. And once the results from these various individual analyses are brought together, along with the data from historical and archaeological sources, they may form a considerably more elaborate and detailed picture than we thought possible.

The approach I will follow is not exactly new in the field of history. It represents, in my view, the joining together of two well-known kinds of investigation: Quellenforschung, investigation of sources, which has been for generations the staple of Biblical, especially Old Testament, studies; and the literary and psychological investigation of text, which is coming more and more into its own as a major area of historical research. To quote a few examples I have read and appreciated, the eminent classicist G.W.Bowersock used it to great effect in his remarkable Fiction as History; Erich S.Gruen, another eminent classicist, made telling use of literary analysis in Heritage and Hellenism; and Paul Strohm, Tolkien Professor of English at Oxford, made a telling and almost revolutionary contribution to medieval studies - overturning, it seems to me, several accepted “facts” - in his England’s Empty Throne: Usurpation and the language of Legitimation, 1399-1422.

I mention these three books because I am familiar with them, but I have no doubt that they represent a more general progress in the study of history, in which text is approached primarily as text and not just for the direct actual data it can supply. What I think is peculiar to the present study, however, is the determined use of Quellenforschung to recreate a whole bookshelf of lost sources, and the cumulative use of conclusions already reached to establish new facts. This makes for a formidable bulk, for which I do not apologize. The study of no single text in isolation - not even Gildas or Nennius or Geoffrey of Monmouth - is capable of giving us a reliable picture of any part of this mysterious past of ours; as too many scholars have found out. It is necessary to use every available source, and to use each of them separately, at least at first, to gain all the information it is capable of giving us. Also, the process of investigation is part of the demonstration itself: a reader who started half-way through the book would simply not understand how and why I reach my conclusions and the weight of evidence for them.

The texts available to us vary greatly, as we would expect, both in quality and in length, from mere scraps and folktales to powerful and refined works of literature; and each, therefore, demands to be treated in different ways. The literary approach suitable to great and accomplished writers such as Bede and Geoffrey of Monmouth, or to a soaring poet such as the historical Taliesin, or even to a self-taught but brilliant preacher such as Saint Patrick, would be wasted on the scraps of data and mere annalistic detritus that forms so much of Nennius; which is not to say that attention to the way that Nennius assembled his material, or to the differences of style in various chapters will not pay dividends. (As a matter of fact, one chapter in Book 6 will consist of nothing but a word-by-word analysis of several Nennian chapters.) Nennius tells a little about Gwynedd, a lot about Powys, and next to nothing about Dyvedd, south Wales and Cornwall. This must be borne in mind when finding, for instance, Geoffrey "of Monmouth" delivering a completely non-Nennian account of "Cunedagius" - the final G is an archaic element of the spelling - and the life of Cadoc shows a lively and individual Arthurian tradition apparently unknown to the author of the Historia Brittonum. They probably were picking up on South Welsh matters to which the earlier compiler had no access, either because of war or of poor communications with the South.

It is worth paying attention to the fact that Nennius was held to be very authoritative for three centuries, translated into Irish and widely attributed to the hallowed authority of Gildas the Wise. Jejune though his accumulation of documents may seem to us, the tenth century seems to have disagreed; which is a reason to treat him with some respect. Also, some writers are more relevant than others; I will only pay attention, for instance, to certain chapters of Geoffrey of Monmouth, without wanting to make an analysis of his whole work. Each work, in short, demands its own particular kind of attention and analysis.

The greatest writer of our corpus, with apologies to the genius and holiness of Bede, is the earliest: Saint Gildas, an almost unknown literary giant; and it is after him that I intend to style the heart of the period I am studying – thus far inelegantly and irrelevantly called “sub-Roman”: Britain before the English will, from now on be “Gildasian Britain” in this book. But I will begin not from him, but from the mysterious person we call “Nennius”[3], a Welsh cleric who, about 834AD, wrote is an oddly mixed little book in seventy-five extremely short chapters, covering all sorts of things from the origins of mankind to a small but colourful selection of travellers’ tales.

Nennius’ choice of materials is curious. That he says nothing at all about Gwynedd is explained by the fact that he was certainly writing for a Gwynedd audience, in the fourth year of King Mervyn (about 830AD) and expected them to know all about it anyway. He had more to say about Powys, and a lot about Builth and Gwrtheyrnion, two tiny twin kingdoms about the headwaters of the Severn and Wye. But he says almost nothing about South Wales and contemporary related countries, Cornwall, Brittany, Strathclyde, Cumberland. What is more, the tales of miracles and oddities with which he fills his last few chapters almost all come from South Wales and the Bristol Channel area, proving that he found them as alien and legendary as darkest Africa. Other areas are developed at surprising length for a document that is mostly more in the nature of a summary than of extended narrative; in particular, there is so much about the history of English Northumbria that the narrative is left quite unbalanced. A number of notices about the north of Britain (where the ebb and flow of English conquest from Northumbria left for centuries a number of Welsh-speaking[4] principalities, of which Strathclyde was the most enduring) have no chronological framework, but have, in recent year, drawn the interested glances of scholars.

Later manuscripts of Nennius bear a Preface describing his work and the sources he used. Professor David Dumville attacked its authenticity with powerful arguments: it is almost certainly a well-made piece of fiction in which an eleventh-century editor takes on the author’s voice to explain to his contemporaries what must have seemed, even then, an odd and unwieldy piece of historical writing. The range of materials available to the eleventh century was already much larger than that from which the author of the Historia Brittonum had had to work, and the editor took it upon himself to apologize for his author (as editors do), explaining in “Nennius”’ voice what problems he had met. The editor assumes (a mistake that has dogged every reading of Nennius since) that the reason for Nennius’ very peculiar selection of materials is that he did not have that much available: Coacervavi omnes quod inveni, he says: I, “Nennius”, gathered and piled up every record I could lay my hands on. This involves a value judgement on his predecessor: as these records seem to have mostly been English and Irish, “I” find “myself” forced to say that Doctores insulae Britanniae - the learned of the island of Britain, that is of the Welsh nation - nullam peritiam habuerunt, used to have no knowledge [of the past] whatever. Editorial cleverness stretches to the use of a typically Nennian word - peritiam - for “knowledge, item of information”; but all these are statements that anyone who had read the Historia Brittonum with any care could easily make. The predominance of English and Irish records, especially for the sixth and seventh centuries, is blatant, and what British records Nennius knew are fragmentary, flagrantly legendary, mutually contradictory, and have little and sometimes nothing to do with the Classical historical tradition that dominated mediaeval minds.

The writer of the Preface divided Nennius’ sources into five separate headings: 1) Annales Romanorum; 2) Chronicae Sanctorum Patrum; 3) scripti Scottorum; 4) scripti Saxonum; and, 5) traditio veterum nostrorum. It is important to understand the kind of documents Nennius worked from. If we took the terms of the Preface for what they mean to us in the twentieth century, with our libraries full of mediaeval texts from every part of the Christian West; then the five headings - Roman annals such as Orosius’; the vast medieval production of hagiography; the considerable annalistic and hagiographic tradition of Ireland and the kingdoms of England; and a less easily defined but not negligible body of inherited native Welsh legends - would make Nennius a formidably learned figure by any standard, and his book a terrible disappointment. Yet the Historia is in fact visibly assembled from many different sources, and the sources named in the Preface can be clearly recognized as separate strands in the collection; as I said, the Preface is a clever piece of fiction, written by someone who had read the text with care. Nennius really did read several separate books, assembling their accounts with as much order as he could achieve. It is just that the annals and traditions that he consulted had not reached anything like the elaboration and complexity of later annalistic tradition. They represented an earlier stage, coming in a few cases directly from a vaguely Christianized pagan British Celtic tradition; others may have been shreds of decayed and largely lost documents, or even - as I will argue in the case of ch.30 - confused memories of books read by the author considerably earlier, whose originals he no longer had available.

But one thing cannot be emphasized enough: what we are talking about is a written tradition. Not a single Nennian passage can be shown not to derive from written documents. It is too easy to imagine an oral origin for obviously legendary narratives, but we should never forget that the ancient Welsh could read and write (or at least their leading classes could, which for my purposes comes to the same thing), and that they carried on - with their own modifications - the institutions of a religion, Christianity, based on books. The techniques of research we need are not those of Axel Olrik[5] so much as those of Erich Auerbach[6]. Even texts of decidedly oral origin, such as the early Welsh poetry of Taliesin and Aneirin, passed through a written phase long before the earliest manuscripts we have; they also must be regarded as having a close affinity to written text in the sense that their text is largely fixed, through demanding memory techniques, in much the same way as Sanskrit sacred poetry, which remained both oral and largely unchanged for centuries. Even more to the point, these texts existed as part of a book-writing culture, and cannot be treated as though - even at the time of their creation - writing were not a part of the highest-status activities of the culture (namely learning). The historical Taliesin[7], the earliest of all known Welsh poets, calls his patron Urien “in truth the king of the baptized world[8]”; baptism is part of Christianity, and Christianity creates a learned (priestly) class of book readers and book writers. In other words, the historical Taliesin may not have mentioned books and priests in his poems (they were not the sort of thing he made poetry about), but he damn well knew that they existed.

Of all Nennius’ material, the easiest parts to unravel are the Irish and English strands; we have a good deal of similar material from England and Ireland anyway. Nobody will reasonably contend that the legend of Patrick or of the four conquests of Ireland are not of Irish origin, or that (pace the late John Morris) the stories of Hengist and Horsa and the dynastic lists of various English kingdoms do not come straight from English documents, with some elaborations of Nennius’ own doing[9]. It is more important and more interesting to try to assess what other documents, otherwise unknown to us, were known to Nennius or in his possession when he wrote.

The first reference to a source not Irish or English or otherwise known to us, is that to “the annals of the Romans” in his chapter 10. For four centuries before Nennius, the study of time-reckoning was a major academic field throughout the Roman and post-Roman Christian world, stimulated by the need to achieve a uniform measurement of the time from the life of Christ, and by related problems such as settling the date of Easter. A by-product was the writing of “annals”, lists of years with their main events. There is nothing strange about the existence of a set of “Roman” annals; but the content Nennius ascribes to them is worth quoting it at some length, to show what kind of document these “annals” were, and what view we should take of them.

“... in the annals of the Romans, it is so written: after the war of Troy, Aeneas came to Italy with his son Ascanius, and, having overcome Turnus, took Lavinia, daughter of Latinus son of Faunus son of Saturn, for a wife, and, after the death of Latinus, took the kingship of the Romans or Latins. Aeneas moreover established Alba and then married his wife [there[10]], and she bore him a son called Silvius.”[11]

Thus far, this is a reasonably accurate account of the Roman national legend, apart from one understandable blunder, the disappearance of Picus from Latinus’ genealogy[12]. Now it gets interesting:

“Silvius also married a wife, and she became pregnant, and it was announced to Aeneas that his daughter-in-law was pregnant, and he sent word to Silvius his son that he should send word to his magus to examine the wife, and discover what she bore in her womb, male or female. And the magus examined the wife and returned. For that prophecy the magus was slain by Ascanius, for he told Ascanius that the wife bore in her womb a boy, and he would be a son of death, for he would slay his father and his mother and be hateful to all men. So it happened...[13]

We need go no further to prove that we are suddenly into a different world. From the point of view of Roman legend, the blunders are scarcely credible. Aeneas only reigned over the Latins three years before being taken up to the gods, or killed: the notion of his surviving to see his son’s wife pregnant is preposterous. What is more, Nennius knew this, and reported Aeneas’ three-year term, with no apparent sense of incongruity, in the very next chapter. Conversely, his son Ascanius reigned for fifty years - it must have taken him a long time to bear his “son of death”!

No less significant is the fact that the previous few sentences had clearly understood the difference between Ascanius, who was with Aeneas before he reached Latium, and Silvius, born in “Alba” (Silvius, not Ascanius, was the actual forefather of the Latin kings in Alba Longa, while Ascanius was the ancestor of a priestly dynasty in Lavinium); but suddenly the two are confused and - though the narrative is not clear - seem to become one and the same; unless we are to take it that Ascanius murdered the magus for bringing bad news about his half-brother.

I don’t need to belabour the point. This is a rather clumsy attempt to insert a native origin legend of Britain into the Classical framework, making the founding hero Britto, or Brutus, a previously unknown grandson of Aeneas, and indeed an heir of the Roman royal line (since he is the first-born son of the heir Silvius). The legend itself is highly Celtic in type, with the exotic word magus thinly concealing, as it does so often in Welsh and Irish Latin, the native Druid of the royal household, sent to predict the fortune of the heir. The murder of such a personage - druids were sacrosanct in pagan Celtic courts - opens the sequel of misfortunes that include the death of Silvius and his wife, and the exile and wanderings of their son, Brutus exosus, Brutus the universally-hated.

The chapter that follows (11) is an unobtrusive and intelligent effort to harmonize Roman and British traditions further, for it names the successor of Silvius and brother of Britto as Posthumus. According to Dionysios of Halikarnassos (1.71.1) and Livy (1.3.6-7), Silvius’ successor was Aeneas Silvius; according to Virgil (6.767), Procas Silvius; according to Ovid, Latinus Silvius. The name Posthumus does not appear in any Roman dynastic list known to me, except that the first Silvius was sometimes known by the name. Posthumus means “born after his father’s death”, there can be no doubt it was meant to designate Britto/ Brutus’ younger brother, born after Brutus had already committed his unintentional parricide.

The author of this chapter had access to a good annalistic tradition: he correctly gives Aeneas’ time as ruler of Latium as three years, and is aware that all the kings of Alba after Aeneas’ son were named Silvius. It was placed next to the Annales Romanorum entry by someone with access to more sources and a better understanding of them - probably Nennius himself.

This means that the legend in Chapter 10 is certainly not by Nennius. And this agrees with his own witness: he claims to be quoting an earlier document called Annales Romanorum (in annalibus autem Romanorum sic scriptus est) and to be quoting it word for word - sic scriptus est: “so [and in no other way] it is written”. Now, quite apart from the sudden fall into un-Classical legend, here the chapter suddenly acquires a style quite unlike most of the rest of Nennius: a style, quite frankly, of ghastly ugliness, but hideous in very idiosyncratic ways. It has a dreadful habit of repeating words and expressions in successive sentences: Aeneas autem Albam condidit et uxorem duxit - Silvius autem uxorem duxit; ut mitteret magum suum ad uxorem considerandam - et magus consideravit uxorem, and so on. Practically all the entry is built out of these jejune repetitions. Repetitions may be found in other Nennian passages, but to nothing like this extent[14] (except, as we will see, in a passage that probably came from the same source); and their consistency indicates the hand of a conscious, if perverse, literary artist who was not Nennius.

The repetitions suggest to me the rhythms of song or chanting, possibly a heroic lay, in which the repetitions would be balanced in lines rather than merely strung out in continuous prose; this immediately suggests a literal translation of a lost Welsh-language poem about the woes of Britto/Brutus. We notice that personal names grow much rarer when the native material is introduced; Britto’s unfortunate mother is never named, nor is the magus, and Silvius and Aeneas are often described only as pater or filius; as if the author had other names in mind, and in order not to slip - and indeed, not to confuse his public - preferred to refer to the characters only by their relationships. That is to say, he identified legendary Roman figures with native ones. But the bad style is equally present in the first few sentences, the ones which come from genuine Roman tradition. Therefore this is not a translation, but rather evidence of a manner. This is the way in which the author of the Annales Romanorum would tell a heroic story, whether in Welsh or in Latin.

This conclusion is reinforced when we realize that the unknown author, in spite of his grotesque Latin style, is a very good storyteller, with a fine understanding of writing for effect. Withholding the prophecy of the magus until after Ascanius has murdered him increases its force. We know that it something horrible before we ever hear what it is; nor does the revelation let us down, for surely no worse horror can be imagined than a child fated to murder both his parents. There is also the power of extreme economy in the two words that follow, sic evenit, so it happened. Though they open another period, they do so in an admirably dry and firm manner, like the stroke of a dead bell. This man could tell a story; he just could not tell it in Latin[15].

These, then, were the so-called “Annals of the Romans”. They did not belong to a primary Roman or Continental tradition, but represent rather the attempt of some unknown Welshman or group of Welshmen, at some point between the age of Gildas and that of Nennius, to harmonize native traditions such as the legend of Britto with some sort of Roman history textbook. The text does suggest that the unknown author had read Roman annals of some sort: the information about Aeneas in the first few sentences is correct enough, but only as much as might be expected from an annalistic entry, nor does the writer show more understanding than such a basic source might have given him. He had not read Virgil (he knows nothing of Virgil’s invention of the town Laurentum) or anything of like complexity. And he was almost certainly the first to attempt to marry British and Classical tradition - at least, the first he himself knew of. He was drawing on no other man’s work. The rough edges and uncorrected blunders, as well as the Welsh style clearly perceptible under the Latin, clearly suggest that he had no previous example to correct, that he was the first. But he felt the work had to be done: though quite unaware of Classical elegance and variation, he clearly saw the Roman tradition as primary, and sought, not to graft any Roman branch on a British tree, but, to the contrary, to insert a British twig - if a noble and impressive one - in the great oak of Roman traditions.

Were the Annales Romanorum actual annals, written in chronological form? Heroic verse had little to do with plain annalistic style; but, except for the style, there is no reason to doubt, and that is easily explained. Our author wished not only to join his national traditions to the great learning of the Mediterranean, but to do so in the high literary manner; and this suggests that he was disappointed by the bald and unadorned form in which his source had cast the high history of Rome. But however he may have felt about his source, he prized at least its testimony of a great past. He may have taken it upon himself to rewrite every entry; he certainly did so with the important episode of Aeneas. Stylistically, he failed, since the manner of his time - whatever its origin - was pretty nearly everything good Latin prose ought not to be, but the attempt is visible. (Metrical chronicle, after all, was to become a major literary form in the Middle Ages.) He would also have been unhappy about the lack of information about Britain to be expected in a late-Roman chronicle[16]; and as he went to native Welsh accounts for information, he evidently regarded them as reputable sources, comparable in truth if not in dignity with the great Mediterranean frame of history.

As for Nennius himself: if one thing is clear, it is that he treats the Annales Romanorum exactly as he treats Irish and English annalistic. He is unaware of their Welsh and comparatively recent origin, betraying no sign of doubt, reading them as an old book of unquestioned authority. Indeed, he seems to regard them as more authoritative than the alternative source quoted in chapter 17[17], which he places as it were in appendix. For this there are two possible explanations: either that the Annales had been written in “a far away country of which Nennius knew nothing”, like those parts of South Wales of which he told such curious tales; or that they had been written before the horizon of living memory.

Now the author of the Annales evidently had no direct contact with any living Classical culture. He almost certainly got his learning from one single source, annalistic in nature, which he valued enough (as a source of information) to want to rewrite it in what he considered the high style; in other words, when the Annales Romanorum were written, a single annalistic text was a precious treasure. The age of Nennius was different: Nennius sat down to write with at least three different sets of annals (the Annales Romanorum and at least one item each from Ireland and England) and sundry other documents in front of him. It follows that the author of the Annales Romanorum must have lived at the height of Welsh isolation, in the long obscure period in which Wales was separate from the Roman church, and therefore from international European culture, and had even drawn the irritation and contempt of Irish ecclesiastics[18]; in which they were talking with nobody. Now the later seven hundreds were an age of rapprochement, completed in 768 on the initiative of bishop Elvodug of Gwynedd (claimed by the Preface to have taught Nennius himself). The late six or early seven hundreds seem a good guess for its date[19].

Nennius had an orderly mind, though his methods need some explaining. He opens chapter 10 with a clear statement that he knew of two separate traditions as to the first men to reach Britain. Chapter 10 was one: Nennius took it, because of its “Roman” colouring, as the more trustworthy account, and accordingly placed it as the first item in what he regarded as a canonical list of British peoples and their origins. Nennius’ first nine chapters had been dedicated to an account of the general chronology of creation and mankind; chapter 10 was the first of what we might call an ethnography of Britain and surrounding islands. From 10 to 15, each chapter details the origin of one of their nations - Picts; Scots of Ireland; Scots of Britain; even a brief mention of the Saxons, who could not be left out of any list of peoples of Britain, though their origin would have to be treated later. Chapter 16 wraps it up with a chronological calculation of the date of King Mervyn of Gwynedd.

Having completed its overview of contemporary British peoples, Nennius then turns back and warns the reader that, however much more credible the “Roman” account may be, there is another peritia of British ethnogenesis in existence. Chapter 17 is surely this other peritia, tradition. It is not an oral account; it is written, ancient, and in Latin, ex veteribus libris veterum nostrorum, out of the ancient books of our ancients; and therefore not to be disregarded, however unlikely it may sound to educated contemporary ninth-century ears. In effect, chapter 17 is in the nature of a footnote. In other words, Nennius aligned together all the traditions he could make to fit each other, not necessarily in a chronological order - the ethnography of Britain in chs. 10-16 includes accounts from different supposed dates - but in a clear one at least. Where he had one which absolutely did not fit, he placed it at the end of the group of chapters to which it was an alternative. Not having such devices as footnotes, it was his best way to signal incompatible material.

His historian’s desire to make as much of his material as possible agree, however, led him to a considerable amount of harmonization. Britain’s name was said to have come from the first ancient hero to have settled the island, known, from Nennius until the “enlightened” eighteenth century, as Brutus, a prince from Aeneas’ family. He was the hero of a long and complex legend told first by Nennius himself and then, at far greater length, by the immensely influential Geoffrey of Monmouth. Now, his original name was not Brutus at all; Nennius sometimes calls him that, but he uses at least as frequently the name Britto. Britto is the standard late-Roman noun and adjective (plural Brittones) for anyone or anything from Britain. The poet Ausonius, a Gaul from Bordeaux - then as now a main sailing port for Britain - uses it as standard for a Briton; more interestingly, two centuries later a Frankish embassy in Constantinople hoaxed[20] the historian Procopius, who knew of “the greatest of islands” only from literary sources, into believing that his literary Brettania was different from an island called Brittia, close to Gaul’s shores and inhabited by Brittones. The word Brittia is clearly a back-formation from britto; therefore the Franks of 530-550 or so knew Britto, Brittones as the word for people from Britain, and specifically for Romano-Celts (the passage distinguishes the Brittones from the Angiloi, English, and from the Frissones, Frisians).

The name Britto, clearly formed on the ordinary national name or adjective, is a typical founding-hero name, like the Ion of Euripides’ tragedy and the Israel awarded to Jacob after he wrestled with the angel; whereas Brutus is clearly a secondary form. By the time Nennius wrote, this change of name was already on the way; in fact, the evidence shows that he must have contributed to or even decided upon it himself. He first mentions the hero, calling him Brutus, in chapter 7; now ch.7 is surely Nennius’ own work, his systematic medieval Christian account of human origins. In the non-interpolated, original part of ch.10 the hero is definitely Britto and no other name is used; but this is the part that Nennius, as we have seen, copied word for word from the Annales Romanorum. The chapter, in various manuscripts, is followed by a number of interpolations, which all ignore the spelling Britto and go for Brutus; which shows that by the tenth and eleventh century, when they were added, Brutus had become the standard spelling. In chapter 11, Nennius uses Britto, probably to agree with the previous chapter, which ch.11 was meant to correct; he is Britto again in the vast and wholly un-Classical genealogy of the European nations in chapter 17; but in the following chapter (probably interpolated), he is Brutus - Brittones a Bruto, “the British [are named] for Brutus”.

The move from the natural and native name Britto to Brutus, Roman and unnatural (you cannot derivate the sounds britto or Britannia or even Prydein from Brutus - the two Us and the N are in the way), is parallel to the insertion of Britto/Brutus’ legend, which we have agreed was native, into a framework of Roman historical legend. Nennius stopped for us, as it were in a photograph, the process of normalization away from the archaic and native spelling into the regular Latin name, by a process of guesswork that argued that, if the ancestor of the Britons came from the Troyans of Latium, then his name must have been recognizably Latin. Just as such savants as those who wrote the Annales Romanorum and the Historia Brittonum had made up their minds that their national founder had to be inserted into Roman historical traditions, so too one or more of them decided that his name had to be a normal Roman name: not the eccentric and probably barbarous Britto, unknown as a personal name to the Roman annals they consulted, but the widespread Brutus, which shared at least its consonants, and which belonged to not one but several celebrated figures of Classical history. The fact that the most famous Brutus was a murderer, and that his victim Caesar was rumoured to be his physical father, may also have influenced the choice of name for a hero who killed his own father and was exosus, universally hated; certainly Caesar, the supposed first Roman conqueror of Britain, loomed large in Nennius’ mind.

The consistent keynote of Dark and Middle Ages thought was the effort to fit the various traditions inherited from half a dozen barbarian cultures, from the Church, and from mere folklore, into the Classical frame of learning which, however deformed and only half-understood, dominated all minds. In this respect, both Nennius and the author of the Annales Romanorum are in the mainstream of Mediaeval thought, however strange and alien their products may seem to us. The Historia Bittonum is quite clearly based upon a selection of previously existing Latin written evidence. However much respect he may show towards five early Welsh poets he obviously regarded as classics[21], nothing suggests that he would use anything except written Latin texts as historical sources. A well-known hypothesis is that the list of Arthur’s twelve battles in ch.56 is originally from a Welsh-language heroic poem; that is certainly possible, but it is at odds with the rest of the book, whose sources are all ecclesiastical, written and, so far as anyone can tell, Latin (some at least of his Irish sources may originally have been in Old Irish). If such a versified battle list existed, my guess is that it may have gone through a written stage and a Latin translation before Nennius picked it up, as did, for instance, the legend of Britto/Brutus[22].

Nennius says almost nothing else about Arthur, and the little he says comes from traveller’s tales; this helps to prove my point, since nothing is more certain than that Welshmen of his time knew far more of Arthur than the little he uses. His own traveller’s tales include a mention, as of something everyone would know, of the hunt of Twrch Trwyth (ch.73), later written down in Cullhwch and Olwen[23]. The only reason for him not to relate this or similar legends of the great warrior king is that they were not cast in a form suitable for Latin-language historical writing. Had he been willing to use Welsh-language and oral sources, the indications are that he could have written a book several times as long. The brevity of the Historia Brittonum is a function of his rigorous attitude to sources: they must be written and Latin.

This attitude is common to virtually every historian between the Roman Empire and the Carolingian renaissance. Only Latin sources, preferably Classical, are to be used. Jordanes, the Roman historian of the Gothic peoples, refuses to consider the fabulae of his own subjects about their own origins[24]. Bede says nothing of English and Welsh origins except what could be found in the Latin writings of Gildas and the Life of St.Germanus. Gregory of Tours fills page after page with Roman witnesses to Frankish antiquities, however transient, but says hardly a word of the Frankish account of themselves; we find out more about their legends from that notorious fable-believer Procopius, who had been clearly taken for several rides by one or more imaginative Frankish ambassadors, than from him. As for Procopius’ untypical willingness to use barbarians as sources - which his Frankish contacts seem to have thoroughly abused - the fact is that he was not a trained historian, but a Byzantine high bureaucrat who turned to history to record the great (and evil) events he had witnessed. The craft and near-priestly rules of late-classical historical writing had far less hold on him than on cloistered and actually ecclesiastical figures such as Jordanes, Gregory, Bede and Nennius.

An unspeakable amount of nonsense has been written because of disregard of this obvious point. Historians suspicious of any Christian writer of history and everlastingly on the look-out for everyone's "bias" but their own have suspected Bede of every kind of misrepresentation under the sun, in particular with regard to the poor dear Welsh - to whom he is held to have been unfair - and to Germanic paganism, on which he is held to be deliberately silent for nefarious purposes of his own (in particular, not to remind the kings that in the pagan world they were priests too). The point is quite simply that Bede had no written or eyewitness sources for any of those things, and, unlike the authors of the Anglo-Saxon Chronicle, was not disposed to risk his neck on unreliable oral accounts of things beyond the horizon of living witness. His introduction clearly shows that he restricted himself to two kinds of source: either written and Latin ones, or oral accounts from people close to the facts, no more than one generation or one witness removed from events. He says next to nothing about the Welsh church because, as he tells us, the Welsh did not speak to the English or even consider them proper Christians (his account of Augustine's unfortunate negotiations with them must come from a Canterbury source), not because he was hell-bent on defaming them; by the same token, he speaks of St.Columba in tones of solemn doubt. What would be his reason to doubt the holiness of a man whose influence over the North - not excluding Bede's own home country, through his beloved Aidan - he regarded as an almost unmitigated Good Thing? For one reason, and one reason alone: that he had no written Latin account of him.

The fact that Nennius shares this attitude to the full tells us two things: first that he belonged basically in a post-Classical culture; and second, that his intended public was one that shared such attitudes. Not long before, Alcuin had blasted the predilection of some English monks for their native English heroic traditions, making it perfectly clear that as far as he was concerned such things had nothing in common with the world their minds were supposed to inhabit; but Roman legend, transmitted through so many church historians and annalists, did. The history of the Roman empire, however full of paganism and violence, touched the sacred history of the Christian faith, in particular in that God-given period of Augustan peace which allowed Christ and His church to first preach the Gospel; the very tale told in the Gospels and Acts, let alone the subsequent history of the Church, from the age of the martyrs to the imperially sponsored great councils, made Roman history a necessary part of sacred learning, excluding alternative sources. Therefore anything in that period that took the form of historical/annalistic writing, had to ignore non-Roman, non-Latin sources.

Different attitudes must have prevailed in Wales in the previous century. The author of the Annales Romanorum valued Roman annalistic, but his use of Welsh traditions tells us that he was not dominated by the same exclusive mentality. Other possible translations from Welsh may include the native legend of Caesar, which Nennius edited into a framework derived from Bede; some of the traditions connected with the “Seven Emperors” who were supposed to have ruled Britain in the days of the Romans; the catalogue of Arthur’s battles; and the various legends of Vortigern which will concern us further on. All of these must, in my view, have been turned into Latin before Nennius[25], since they are far too few to represent a fair selection from what we cannot doubt was an ample and vigorous native storytelling tradition. Geoffrey of Monmouth did something truly revolutionary when he wrote, in Latin, a traditional history of Britain from native legends; but that was three centuries after Nennius.


[1]In spite of Professor Dumville's authoritative objections, I will call the author of the Historia Brittonum by his traditional name Nennius. That the eleventh-century Preface confidently named him shows that a tradition that identified him as the author was by then in existence, and there was a scholar by that name at large in North Wales at the right time. The notice that the famous Bishop Elvodug was his teacher is more dubious still - but hardly impossible; the Bishop may well have taught pupils at some point, and a young Nennius may have been one of them. After all, in a country as small as Gwynedd, every learned man must have been a valuable resource.  And given all the different anonymous sources we will encounter, I am grateful for any name, however spurious.

[2]I will show in the appendix on the legend of Gwyrangcon that Nennius certainly knew more than he set down, and ignored a strand of legend with which he must have been perfectly familiar.

[3]Throughout this work I will refer to John Morris' version of Nennius, number 8 in his Arthurian Period Sources, Chichester 1980. It is hardly ideal: Morris died half-way through the work, and nobody seems to have cleaned up what he left behind. As a result there are no notes, no distinction is made between presumably original passages and obvious interpolations (some not even in Latin but in Irish!), and at least one whole paragraph is missing in the translation. But it is in print and accessible.

[4]Properly speaking, Cumbrian, a separate dialect; but both the Welsh and the Cumbrians regarded themselves as one and the same nation, and many Welsh dynasties traced their descent from the North.

[5]A justly legendary Danish scholar, author among other masterpieces of Techniques of oral narrative research (English translation by K.Wolf and J.Jensen, Bloomington 1992).

[6]Probably the greatest literary critic who ever lived, whose Mimesis (New York 1954) contains almost everything one needs to learn to read a text with sensitivity and insight.

[7]As opposed to the better-known Taliesin of the legends, who is not part of the argument of this book at all.

8]The battle of Wensleydale, in MEIRION PENNAR (ed. & tr.), Taliesin poems, Llanerch 1988, p.51.

[9]DAVID N.DUMVILLE, The historical value of the Historia Brittonum, in Arthurian Literature VI (1986), 1-26.

[10]The Latin may be read to say that he took another wife after Lavinia; but there is no need, in my view, to introduce this unnecessary complication. The sequence seems to me quite clear: first, Aeneas built his town (Albam condidit), and then married his wife - his only wife, Lavinia; and in the new town she conceived and bore him a son.

[11]In annalibus autem Romanorum sic scriptum est: Aeneas post Troianum bellum cum Ascanio filio suo venit ad Italiam et, superato Turno, accepit Laviniam filiam Latini filii Fauni filii Saturni in coniugium, et post mortem Latini regnum obtinuit Romanorum vel Latinorum. Aeneas autem Albam condidit et postea uxorem duxit, et peperit ei filium nomine Silvium.

[12] Another blunder, the confusion of Alba and Lavinium, has a more complex genesis. It probably depends on a Roman source which ignores Laurentum (which, as a separate town, is a Virgilian invention known to no other source) and places the Latin royal seat in Alba from the beginning. It is, however, a blunder, since in all Roman sources that mention Lavinia, the city which Aeneas establishes when he marries Lavinia is Lavinium.

[13]Silvius autem duxit uxorem, et gravida fuit, et nuntiatus fuit Aenea quod nurus sua gravida esset; et misit ad Ascanium filium suum ut mitteret magum suum ad considerandam uxorem, ut exploraret quid haberet in utero, si masculum vel feminam. Et magus consideravit uxorem et reversus est. Propter hanc vaticinationem magus occisus est ab Ascanio, quia dixit Ascanio, quod masculum haberet in utero mulier, et filius mortis erit, quia occidet patrem suum et matrem suam, et erit exosus omnibus hominibus.  Sic evenit...

[14] Except, as I intend to show, in ch.36, which I am going to argue comes from the same source - see below, bk.6, ch.9.

[15]I have come across a very similar, but far less tight, style - something like an inferior imitation of this - in the first part of the Breton Life of St.Gurthiern, a brief document which reached its current form at some point in the eleventh century. It is the fusion of two separate documents by two very different writers, the second with better Latin but even less to say; the first part does not make up for its poor grammar and repetitious style with the narrative ability of the Annales Romanorum, indeed it leaves a number of dangling loose ends, moves the characters around to little purpose, and leaves the impression that the author had no interest whatever in his material. It is however interesting that his manner and style should seem so close to those of the Annales. L.MAITRE & P.DE BERTHOU (eds.), Le cartulaire de Quimperlé, in Bibliotheque Bretonne Armoricaine IV (1902), 42-46 (the first part ends at page 45). A similar style also seems to have been employed in some narratives in the ancient charters of the Book of Llandaff.

[16]In the seventh century, following the end of direct seaborne trade, the Mediterranean world all but lost direct knowledge of Britain. Writing in the middle of the century, the Cosmographer of Ravenna mentions it immediately after China and Scytia, among the remote and fabulous regions, and evidently only knows it from books.

[17]In ch.27, he seems to contrast the Annales Romanorum with veteri traditione seniorum nostrorum, the ancient tradition of our elders; the inference being that the former were not Welsh, where in fact both documents he is discussing in that chapter clearly repeat, though with variations, a common Welsh legendary version of Roman history.

[18]An Irish canon about "people and places to be avoided", dated by Dumville to the early eighth century, treats the British (Welsh) as on the same level as heretics and Jews, because "they are turned against everyone else, and cut themselves off from Roman ways and from the unity of the Church"; that is, they are not as alien as the pagans, but they belong to a penumbra of dissident, only half-Christian bodies - Jews. heretical Christians (possibly Pelagians) and schismatics - whom it is well to leave strictly alone, because in each case they are inferior, on different grounds, to the real thing. Jews "are enslaved to shadows rather than [serving] the truth"; heretics can be "learned and studious in all things to do with the Church" but should be given a wide berth anyway. In this company, the Britons represent the third great category separated from the Catholics - nationalist schismatics moved more by national pride than basic doctrinal difference. A century later, an Irish writer, in a private letter to another, simply treats Welsh ecclesiastics as noodles. DUMVILLE ET AL., Saint Patrick: A.D.493-1993, Woodbridge 1993, 144, note 69 and texts therein quoted.

[19]the same token, the first part Life of St.Gurthiern, mentioned in note [15] By above, could well date to the same period, given the stylistic parallels. The point will become important later on. If I am correct, then it is worth noting that it could predate its being written down in the Quimperle' by as much as four centuries.Cartulary of

[20] That the Franks, at some point, stuffed Procopius full of lies, is important to the hypotheses of this study. I make the case in the enclosed Appendix V, below: Procopius and Britain.

[21]Neirin, Taliesin, Talhaern Tat Agwen, Bluchbardd, Cian Gweinth Gwawt; chapter 62. Significantly, they are mentioned almost in the same breath as Mailcun - Maelgwn, the Maglocunus of Gildas.

[22]Dumville: " was taken by our author" [i.e. the man I call Nennius] "from a written source, as is shown by his misinterpretation of Old Welsh scuit" - The historical value of the Historia Brittonum, in Arthurian Literature 6 (1986), p.13, n.42, quoting Bromwich - Concepts of Arthur, in Studia Celtica 10/11 (1975-6), 163-81. I would add that a Welsh word alone in a Latin passage would be likelier to be misunderstood - by a man who was after all a Welshman himself - than one in a Welsh poem, whether oral or written.

[23]Despite its incredibly baroque and often heavily humorous manner, the twelfth-century Welsh prose tale Cullhwch and Olwen is partly based on dynastic traditions. Cullhwch, Arthur's cousin, is the grandson of Kelyddon - Caledonia, Scotland; and his marriage to Olwen, daughter of the usurping Ysbadadden Head Giant, will at the same time mean the death of her villainous father and restore to his throne the legitimate king Constantine (a common dynastic name in Welsh Britain, for obvious reasons) who is Cullhwch's uncle by marriage; the whole story begins with the contested succession to the throne of Cullhwch's own father. Although the fabulous and grotesque elements might have been of little interest to Nennius, the dynastic elements would indubitably have been fodder for his work - if he had regarded it as a legitimate source.

[24]JORDANES 5.38.4-9: Nec eorum fabulas alicubi repperimus scriptas...nos enim potius lectioni credimus quam fabulis anilis consentimus. "Nor have we found their oral accounts (fabulas, from fari, to speak) written down anywhere... for we far rather believe in what we read than accept oral accounts from old women." This is a clear statement of principle: for a historian, oral and barbarous sources don't count.

[25]Curiously enough, historical writing had a rather similar history in Italy. For a century or more the Longobards were culturally and politically in a peculiar position: directly at war with the Emperor, threatening the Pope, the last Germanic kingdom to cling to Arianism, they rejected, and were rejected by, the centres of Roman civilization to an extent Franks and Englishmen like Bede, let alone full-blooded Romans like Jordanes and Gregory, could hardly share. Now Italy spoke Latin, and all the Longobard documents we have are in Latin; but items such as the law-code of Rothari and the anonymous Origo gentis Langobardorum, with little contact with the old culture, tell Longobard dynastic and pagan legends cheerfully. Eventually the kingdom drifted back into a religiously and culturally orthodox position, and produced one fine historian, Paul the Deacon, trained in all the orthodox disciplines. As a Longobard patriot, he set himself the task of writing his people's history; and finding some such legends in his sources, he admitted them to his pages, but very much with a wry mouth.  Refert hoc loco antiquitas ridiculam fabulam - "antiquity tells a ridiculous fable at this point"...

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