The Heroic Age

A Journal of Early Medieval Northwestern Europe

Founded 1998   |   ISSN 1526-1867

Issue 16 (2015)

Traces of an Arthurian Source in Geoffrey of Monmouth's Historia? A Critical Analysis of Geoffrey Ashe's "Historical Abstract"

Howard M. WisemanMailto: Icon

Centre for Quantum Dynamics, Griffith University, Australia

School of Mathematics and Physics, University of Queensland, Australia

©2016 by Howard M. Wiseman. All intellectual property rights reserved. This edition copyright © 2016 by The Heroic Age. Permissions granted for educational and personal purposes only.

Abstract: Geoffrey Ashe proposed that Geoffrey of Monmouth's account of Arthur's Gallic warfare and subsequent downfall is best explained as drawing upon a hypothetical "historical abstract" of the recorded deeds of Riothamus, king of the Britons, in c. 470. Here I show that Ashe's argument fails even on its own terms, and suggest a more plausible hypothetical source that may have inspired certain aspects of Geoffrey of Monmouth's account.

1. Introduction

§1.  The career of King Arthur in Geoffrey of Monmouth's Historia Regum Britanniae (HRB) of c. AD 1137 falls naturally into three unequal parts. The first (Book 9, Chaps. 1–9) begins with his coronation and ends with his marriage. This part is dominated by Arthur's victories against the Saxons, Scots, and Picts—the genuine enemies of the Britons in the post-Roman period as authenticated by Gildas' testimony in the sixth-century De Excidio Britannia (DEB). This part of the "history" is easily accounted for as Geoffrey's imaginative elaboration upon the brief history in the DEB, the Arthurian battle-list of the eighth- or ninth-century Historia Brittonum (HB), and certain other Brittonic traditions.1 That is, the degree of embellishment in this part is no more than Geoffrey applies to earlier (Vortigern, Ambrosius) and later (Cadwallon) historical characters.

§2.  The second part of Arthur's career (9.10–20, 10) begins with his conquest of the "six islands"2 and ends with his abortive invasion of Italy. This part is dominated by Arthur's victories against the Roman Empire in Gaul. It is considerably longer than the first part, but has no counterpart in early (first millennium) recorded Britonnic traditions. For this reason it has been common to ask: did Geoffrey have some other source for this part of Arthur's career? Such a source could have been merely a record of the deeds of another individual (e.g. Magnus Maximus) that served as an inspiration to Geoffrey. More interestingly, it could have been a record of the deeds of someone whom Geoffrey (rightly or wrongly) identified with Arthur of the HB. Some notable historical characters suggested in the second instance are the second-century Lucius Artorius Castus (Malcor 1999), and the fifth-century Riothamus. Two centuries ago, Sharon Turner (1820, 249) wrote "Either this Riothamus was Arthur, or it was from his expedition that Jeffry [sic.], or the Breton bards, took the idea of Arthur's battles in Gaul." This same idea has most recently been championed by Geoffrey Ashe (1981).3

§3.  The third and briefest part of Arthur's career (Book 11, Chaps. 1–2) is the story of Modred's rebellion, ending in Arthur being carried from the field of the final battle, by the river Cambula in Cornwall, to Insula Avallonis, in the year 542. Once again this could easily be explained as embroidery upon Britonnic traditions of Arthur and Medraut, a brief notice of which is preserved in the tenth-century Annales Cambriae. In support of this hypothesis, among Modred's troops in the HRB are the old enemies of Saxons, Scots, and Picts. By this reading, the second part of Arthur's career in the HRB is a fantastical interruption between the first and third parts (Gidlow 2004, 273). On the other hand, the second and third parts dovetail diegetically4 and Ashe argues that they are inseparable.

§4.  Ashe has two motivations for making this argument: first, he wishes to link his suggested source for the second part with the treachery, the Insula Avallonis, and the 542 date in the third part (see Section 3 below). Second, Geoffrey begins the third part (11) by mentioning, for the first time since the dedication in Book 1, the notorious "very ancient book in the British tongue." Ashe would like to believe in the reality of this book, not—as Geoffrey claimed—as the source for the entire HRB, but at least as the putative missing source for the second part of Arthur's career. Ashe goes so far as to give an imagined "historical abstract" which he claims is "based fairly on recognized records" and could, in Geoffrey's hands, have given rise to this second part.

§5.  The purpose of the present paper is to examine Ashe's methodology in creating his "historical abstract," and whether it could have served even the limited purpose of inspiring the second part of Arthur's career in the HRB. As such, my paper is largely complementary to previous critiques of Ashe's hypothesis (e.g. Padel 1995; Hanning 1995), which have been more concerned with arguing that Riothamus could not be "the historical Arthur" or even the prototype for this Arthur, as Ashe would like to believe.

§6.  Ashe claims that evidence for his "historical abstract" exists in the Preface to the Legenda Sancti Goeznovii, which has a dubious date of 1019. Thus I begin in Section 2 by showing that this source, even if it is pre-Galfridian, does not live up to Ashe's claims. Then in Section 3 I analyze Ashe's hypothetical text, and show that it too fails, even on Ashe's own terms of explaining the supposedly anomalous prominence of Arthur's Gallic conquests in the HRB. I conclude in Section 4 with a summary and my own suggestion for a hypothetical source of the most puzzling details in Arthur's military career in the HRB.

2. The Legenda Sancti Goeznovii

§7.  The Legend of St. Goeznovius, a Breton hagiography of the Cornish-born saint Goueznou of Léon, was putatively authored by William of Léon in the year 1019. Only the historical background in the Preface is relevant to the present topic, for which the author (William?) claims the written authority of an Ystoria Brittanica. The part of the Preface relating to the fifth and sixth centuries is as follows (Ashe 1985, 103):

In the course of time, the usurping king Vortigern, to buttress the defence of the kingdom of Great Britain, which he unrighteously held, summoned warlike men from the land of Saxony and made them his allies in the kingdom. Since they were pagans and of devilish character, lusting by their nature to shed human blood, they drew many evils upon the Britons. Presently their pride was checked for a while through the great Arthur, king of the Britons. They were largely cleared from the island and reduced to subjection. But when this same Arthur, after many victories which he won gloriously in Britain and in Gaul, was summoned at last from human activity, the way was open for the Saxons to go again into the island, and there was great oppression of the Britons, destruction of churches, and persecution of saints. This persecution went on through the times of many kings, Saxons and Britons, striving back and forth.

Tatlock (1939, 361–365) argued that the date of 1019 was a spurious late addition, since, in his view, the preface clearly epitomized the HRB (presumably the Ystoria Brittanica referred to) and so could not date earlier than c. 1140. Ashe disputes this argument, pointing to a number of aspects of the account in the HRB that are omitted from that of the Legenda. Although one could easily explain these omissions by the brevity of the summary, for the purpose of this paper, I will not dispute Ashe's position. Ashe does not propose that the Legenda is the missing source for the greatest (second) part of Arthur's career in the HRB—the single phrase "and in Gaul" is too meager a basis for that. Rather, Ashe sees the Legenda as independently drawing upon Geoffrey's source.

§8.  Even if we follow Ashe in accepting the Ystoria Brittanica of the Legenda as Geoffrey's "certain very ancient book," we may still question his conclusions about the nature of that source. Ashe's first major conclusion is: "Behind this naive text [the Legenda] we can infer a [lost] source with some relation to fact." To address this, we should ask what is contained in the above excerpt5 from the Legenda that could not have been taken from extant Brittonic records from the first millennium, in particular the HB. There are a number of items: (i) The unlawfulness of Vortigern's rule; (ii) Arthur's title, "King of the Britons;" (iii) his expulsion or subjection of the Saxons; (iv) his victories in Gaul; (v) the destruction of churches and persecution of saints. With regard to (i), this is a small leap to make from the account in the HB where Vortigern is condemned by a council of the British clergy, and hated for his sin by all men of his own nation. Similarly, (ii) and (iii) are natural extrapolations from the Arthur of the HB who was the leader in battle of the British kings and who fought against the Saxons and was victorious in all his campaigns. Item (v) simply reflects the different interest of the author of the Legenda from that of the HB. This leaves item (iv). It would hardly be surprising if an eleventh century Breton author had invented such a detail, to connect his prefatory history more directly with the following Saint's life. Ashe, by contrast, is convinced that these claimed victories in Gaul "reflect … a real historical situation" (Ashe 1981, 306).

§9.  Ashe's second major conclusion, which presupposes the reality of Arthur's victories in Gaul, is that they must have been in the 460s. His argument relies on the following facts: (i) Arthur's victories immediately follow the Saxon devastation "which certainly belongs to the 440s and 450s" (Ashe 1981, 306); and (ii) Arthur's title "king of the Britons" could only have applied "from some time in the 440s to some time in the 470s" (Ashe 1981, 308). With regard to fact (i), Ashe's confidence in the dating of the Saxon devastation is misplaced; Dumville (1984, 83) dates it as late as c. 490. Also, the Legenda does not specify for how long the Saxons "drew many evils upon the Britons;" if it corresponded to the period in the DEB from the Saxon "storm" to their "last defeat" then it could have been more than 44 years (O'Sullivan 1978, chapter 7; Wiseman 2000). Thus there is no reason to restrict Arthur's victories in the Legenda to the fifth century, let alone to the 460s. With regard to fact (ii) this claim is contradicted by the common attribution of the title "king of the Britons" to individuals in the sixth and later centuries, in several histories predating the Legenda and the HRB6 . The title "king of the Britons" was often claimed by (or for) the most powerful of the Brittonic kings, and carries no implication of his being the direct ruler of all or even most of the Britons, as Ashe assumes.

§10.  In summary, even if the Arthurian section of the Legenda is an independent trace of Geoffrey's hypothetical missing source, it differs significantly from Geoffrey's probable insular sources only in its claim that Arthur had victories in Gaul. This could easily be an invention of its author, or a piece of Breton folklore. Moreover, even if it were a genuine record, it can date Arthur's victories in Gaul no more precisely than can the floruit of Arthur of the HB be dated, to say c. 460–c. 540 (Wiseman 2000).

3. Ashe's hypothetical "historical abstract"

§11.  Here is the proposed "historical abstract" by Ashe (1981), upon which Geoffrey supposedly built the second part of Arthur's career, with the addition of sentence numbering to aid in the discussion below:

(i) When Leo reigned at Constantinople, the Romans still laid claim to Gaul. (ii) Parts of it were held by the Britons, and by nations which had come out of Germany. (iii) At that time a fleet came over with the army of the king of the Britons, Riothamus. (iv) These were warriors against the Saxons. (v) In the island of Britain the Saxons had drawn back after great devastation. (vi) The king and his men passed through the northern part of Gaul and advanced into country neighbouring the Burgundians, who were allied to the Romans. (vii) The strength of the Britons in Gaul was brought to nothing by the treason of the deputy of the Imperator. (viii) Because of it a hostile army was raised and a battle was fought in which many of them perished. (ix) It was the year 442. (x) Their leader departed into the region where the place called Avallon is. (xi) After this time the Saxons grew strong upon the sea and in Britain. (xii) They gained ground when the Britons became divided.

Ashe claims that: "Everything here is more or less historical, and based fairly on recognized records." This claim is probably equivocal enough to be irrefutable, but let us examine the historical bases of the passage in detail.

§12.  Sentence (i) is historical, and supported by many records. The same is true of (ii) as regards the Germans, and as far as the Britons are concerned, there is certainly indirect evidence for their settlement in Armorica by this time (Wiseman 2011, 11–13). The expedition of Riothamus in c. 469, as per (iii), is reported by Jordanes, and corroborated by Sidonius and Gregory of Tours. Sentences (iv) and (v) represent a reasonable supposition. While there are no reliable records that necessitate conflicts between Britons (in Britain or on the continent) and Saxons in the years prior to 469, many modern historians have inferred from the DEB that the Saxon revolt predated Riothamus's expedition. Note that sentence (v) is deliberately ambiguous—in the DEB it is the Britons who are devastated, but here the sentence is intended to be one that could be misread (by Geoffrey, hypothetically) as involving a devastation of the Saxons, suggesting an association with Arthur the Saxon slayer of the HB.

§13.  Jordanes and Gregory say that the Britons occupied Bourges, west of Burgundia, which supports sentence (vi). However the vagueness of the description is again to be deliberately misleading—Arthur's major Gallic battles in the HRB are fought just north of Burgundia, not west of it. The next two sentences (vii, viii) are positively Delphic in their ambiguity. The Praetorian Prefect of Gaul, Arvandus, was guilty of encouraging the Visigoths to attack "the Britons beyond the Loire" (Sidonius). But it is unknown whether this treason led to the battle between Riothamus and Euric at Déols near Bourges (south of the Loire). Ashe's artifice would, he supposes, have allowed Geoffrey to incorrectly imagine that Riothamus was the Imperator, that the treasonous "deputy" was in Britain, that the reason the Britons' strength in Gaul was brought to naught was that they had withdrawn to face a hostile army raised by the traitor in Britain, and that this was where the final battle was fought.

§14.  There is no reference to an exact year in any of the primary sources for Riothamus. The year in sentence (ix) is given by the system of Victorius, corresponding to the year 470 by the Dionysian or Anno Domini system. Ashe requires Geoffrey, or a predecessor, to confuse the two dating systems—there are certainly many precedents for this, e.g. in the HB—and then to "correct" AD 442 to AD 542. Jordanes says that the defeated Riothamus fled to the neighboring Burgundians. Such a retreat might have passed near the small town of Avallon in Burgundy, but there is no reason to identify the region by this obscure place, and it is far more likely that Geoffrey of Monmouth's Insula Avallonis had a mythological origin. In sentence (xi), the growth of Saxon power on the sea is supported by Sidonius's Letter to Namatius. Finally, the situation in Britain here and in sentence (xii) is a reasonable inference from Gildas and Bede.

§15.  Ashe's abstract makes sense as a source for Geoffrey, and for the Legenda, only if Riothamus is identified with Arthur. Ashe states that this identification is the only one possible for an historical Arthur, and that it may well be true. For instance, Ashe suggests that "Riothamus" may have been a title, or an alternate (Celtic) name for the Roman Artorius. Elsewhere (Ashe 1982), he notes that Riotamus R. is an anagram of Artorius M., with Riot(h)amus indeed being described as Rex by Jordanes, and Arthur, in our earliest extant source (the HB), being described as Miles (a soldier). Ashe thus posits that the abstract may have ended thus:

His end is mystery. He is also called Arthur, being known by that name in the island of Britain.

At this point, it is crucial to take stock of the fact that Ashe's original motivation for reconsidering Arthur of the HRB is to "take seriously" the fact that Geoffrey's Arthur "is more a Gallic conqueror than anything else" (Ashe 1981, 304). To what extent does Ashe's abstract succeed by his own terms of being a source for this part of Arthur's career? To judge this we need to revisit the abstract, as Ashe would have Geoffrey (mis)understand it:

When Leo reigned at Constantinople, the Romans still laid claim to Gaul. But large parts of it were held by the Britons. At that time a fleet came over with the army of the king of the Britons, Arthur. These warriors had devastated the Saxons in Britain, forcing them to withdraw. Arthur and his men passed through Gaul and advanced into the country north of the Burgundians, who were allied to the Romans. But Arthur had to abandon Gaul because of the treason of his deputy in Britain who had raised an army there. They fought a battle there in which many Britons died, in the year of our Lord 442. Arthur departed to the place called Avallon. After this time the Saxons grew strong in Britain and gained ground when the Britons became divided.

As is now apparent, very little of this interpreted abstract concerns Arthur's "Gallic conquests." There is in fact no mention of fighting or annexation in Gaul at all, only the passage of Arthur's army from the sea to somewhere north of Burgundia, and back again.

4. Conclusion

§16.  I have criticized the "Riothamus theory" of Geoffrey Ashe on several grounds. First, the Legenda Sancti Goeznovii, even if it does predate the HRB, provides only weak evidence for an independent tradition of Arthurian conquests in Gaul. Second, the Legenda certainly does not pin down the time for such conquests to the 460s. Third, the hypothetical "historical abstract," which Ashe postulates as the source for both the Legenda and the HRB, is quite implausible in one place and requires, in several other places, very particular wording to allow for the hypothetical misunderstandings made by Geoffrey of Monmouth. Fourth, when those misunderstandings are written into the text, one finds that Ashe's abstract fails on his own terms. That is, it does not make Arthur "more a Gallic conqueror than anything else," and in fact makes no mention of victories in Gaul at all (unlike the Legenda).

§17.  To explain the location of Arthur's Gallic war in the HRB, Ashe hypothesizes the following crucial misunderstanding by Geoffrey of Monmouth of his (Ashe's) hypothetical source: the location of Arthur's army north of Burgundia. Now Gidlow (2004, 304–305) identifies a handful of elements in HRB suggesting that Geoffrey did have a lost Breton source. Chief among them are the locations mentioned with regard to the maneuvering of Arthur and the Romans on the northern border of Burgundy: the river Aube, Langres, Saussy, and Autun. Unfortunately for Ashe, his "historical abstract" does not mention these places, and could not, since the localization to the northern border of Burgundy is, according to Ashe, a misunderstanding.

§18.  In this context, there has been an interesting recent development. There is an extant, not hypothetical, source—the Vita Sancti Dalmatii of c. 800—which could be read as implying the presence of a Brittonic army in the region of northern Burgundia, in the period c. 534–c. 541 (Wiseman 2011, 24–29). The terminus ante quem here is precisely when Geoffrey of Monmouth places Arthur's battles in this region. Moreover the terminus post quem was the year of the siege of Autun by the Franks (Wiseman 2011, n. 65). I am not at all suggesting that the Vita Sancti Dalmatii is Geoffrey's "certain very ancient book." However, if that Vita does preserve a genuine record of Brittonic military activity in that part of Gaul—and there are many scenarios allowing for such activity (Wiseman 2011, 30–31)—then it is possible that another record of that campaign, with the locations named above and some indication of the date, survived in Brittany.

§19.  It is most likely that there was no kernel of truth behind Arthur's Gallic campaigns in the HRB. But if there was, I submit that the hypothetical record just suggested is a more plausible source than Ashe's far more ambitious, hypothetically misunderstood, hypothetical abstract.


1. Apart from the mirabilia of the HB, the most easily identified of these other sources is a list of Arthur's named possessions (including his wife) as preserved independently in Culhwch ac Olwen. [Back]

2. That is, Ireland, Iceland, 'Gotland', the Orkneys, Norway, and Denmark. The fact that these conquests are later repeated in the HRB by Geoffrey's Malgo perhaps argues that they should be included as part of Arthur's insular career. It makes no substantial difference to my case whether they are or not. [Back]

3. The case is argued at greater length, but without much greater effect, in his 1982 and 1985 books. [Back]

4. Arthur hears of Modred's rebellion in 10.20, but the consequence, his immediate cessation of his war with Rome is narrated in 11.1. [Back]

5. It should be noted that the earlier part of the preface contains material with no "relation to fact" and it is in this that the most compelling argument for its drawing upon the HRB is to be made (Tatlock 1939, 361–363). In particular, the presence of Corineus alongside Brutus as the conqueror of Britain is featured in no other extant text predating the HRB. [Back]

6. Some examples are the following: "the [nation of the] Britons … [has] one king over it" (Procopius, History of the Wars 8.20) referring to the mid-sixth century; "Solon son of Conaen, king of the Britons" (Annals of Ulster s.a. A.D. 613); "the king of the Britons, Cadwalla" (Bede, Historia Ecclesiastica 3.1) referring to the year 633; "Rhodri, king of the Britons" (Annales Cambriae, s.a. A.D. 754); "in the year 858 [correctly 848], in the 24th year of Mervin, king of the Britons" (HB 1.1). [Back]

Works Cited

Ashe, Geoffrey. 1981. "A certain very ancient book"—traces of an Arthurian source in Geoffrey of Monmouth's History. Speculum 56:301–323.  [Back]

———. 1982. The Kings and Queens of early Britain. London: Methuen.  [Back]

———. 1985. The Discovery of King Arthur. London: Guild Publishing.  [Back]

Dumville, David N. 1984. "The Chronology of De Excidio, Book I." In Gildas: New Approaches. Suffolk: Boydell.  [Back]

Geoffrey of Monmouth. 1966. The History of the Kings of Britain. Trans. Lewis Thorpe. Middlesex: Penguin.  [Back]

Gidlow, Chris. 2004. The Reign of Arthur: From History to Legend. Gloucestershire: Sutton.  [Back]

Gregory of Tours. 1974. The History of the Franks. Trans. Lewis Thorpe. London: Penguin.  [Back]

Hanning, R. W. 1995. "Inventio Arthuri: A Comment on the Essays of Geoffrey Ashe and D. R. Howlett." Arthuriana 5:96–100.  [Back]

Jordanes. 1915. The Gothic history of Jordanes. Ed. and trans. Charles C. Mierow. Princeton: Princeton University Press.  [Back]

Malcor, Linda A. 1999. "Lucius Artorius Castus Part 1: An Officer and an Equestrian." The Heroic Age: A Journal of Early Medieval Northwestern Europe 1.  [Back]

Morris, John. ed. and trans. 1980. British History and the Welsh Annals. London: Phillimore.  [Back]

O'Sullivan, T. D. 1978. The De Excidio of Gildas: Its Authenticity and Date. Leiden: Brill.  [Back]

Padel, O. J. 1995. "Recent Work on the Origins of the Arthurian Legend: A Comment." Arthuriana 5:103–114.  [Back]

Sidonius Apollinaris. 1936. Poems and Letters. Ed. and trans. W. B. Anderson. Cambridge, MA: Harvard University Press.  [Back]

Turner, Sharon. 1820. A History of the Anglo-Saxons. London: Longman and Co.  [Back]

Tatlock, J. S. P. 1939. "The Dates of the Arthurian Saints' Legends." Speculum 14:345–365.  [Back]

Wiseman, Howard. 2000. "The derivation of the date of the Badon entry in the Annales Cambriae from Bede and Gildas", Parergon 17, 1-10.  [Back]

———. 2011. "A British Legion Stationed near Orléans c. 530? Evidence for Brittonic Military Activity in Late Antique Gaul." Journal of the Australian Early Medieval Association 7:9–31.  [Back]