Geoffrey of Monmouth's Welsh Sources

A Companion to Geoffrey of Monmouth By Georgia Henley Joshua Byron Smith LEIDEN | BOSTON For use by the Author only | © 2020 Koninklijke Brill NV Contents Acknowledgements IX A Note on Translations x List of Figures xI Abbreviations xiI Notes on Contributors xIV Introduction and Biography Joshua Byron Smith 1 Part 1 Sources 1 Geoffrey of Monmouth’s Welsh Sources Ben Guy 31 2 Geoffrey of Monmouth’s Classical and Biblical Inheritance Paul Russell 3 Geoffrey of Monmouth and the English Past Rebecca Thomas 4 Riddling Words: the Prophetiae Merlini 129 Maud Burnett McInerney 67 105 Part 2 Contemporary Contexts 5 Early Manuscript Dissemination Jaakko Tahkokallio 155 6 Early Reactions to Geoffrey’s Work Simon Meecham-Jones 7 The Latin Reception of the De gestis Britonum 209 Siân Echard 181 For use by the Author only | © 2020 Koninklijke Brill NV vi Contents 8 Geoffrey of Monmouth’s De gestis Britonum and Twelfth-Century Romance 235 Françoise Le Saux 9 The Most Excellent Princes: Geoffrey of Monmouth and Medieval Welsh Historical Writing 257 Owain Wyn Jones 10 Geoffrey of Monmouth and the Conventions of History Writing in Early 12th-Century England 291 Georgia Henley Part 3 Approaches 11 Colonial Preoccupations in Geoffrey of Monmouth’s De gestis Britonum 317 Michael Faletra 12 Geoffrey and Gender: the Works of Geoffrey of Monmouth as Medieval “Feminism” 341 Fiona Tolhurst 13 Geoffrey of Monmouth and Race Coral Lumbley 14 Religion and the Church in Geoffrey of Monmouth Barry Lewis 369 397 Part 4 Reception Introduction to Part 4: The Medieval Reception of Geoffrey of Monmouth 426 Georgia Henley and Joshua Byron Smith 15 Geoffrey of Monmouth’s Byzantine Reception Thomas H. Crofts 427 For use by the Author only | © 2020 Koninklijke Brill NV vii Contents 16 The De gestis Britonum in Castile Paloma Gracia 17 The Reception of Geoffrey of Monmouth in the Crown of Aragon Nahir I. Otaño Gracia 18 The Middle Dutch Reception of Geoffrey of Monmouth David F. Johnson 19 The English Reception of Geoffrey of Monmouth Elizabeth Bryan 20 The Anglo-Norman and Continental French Reception of Geoffrey of Monmouth’s Corpus from the 12th to the 15th Centuries 454 Jean Blacker 21 The German Reception of Geoffrey of Monmouth Joshua Byron Smith 22 The Old Icelandic “Brut” Hélène Tétrel 23 The Reception of Geoffrey of Monmouth in Ireland Joshua Byron Smith 24 The Reception of Geoffrey of Monmouth’s Work in Italy Fabrizio De Falco 25 Geoffrey of Monmouth in Portugal and Galicia Santiago Gutiérrez García 26 The Scottish Reception of Geoffrey of Monmouth Victoria Shirley 487 27 The Reception of Geoffrey of Monmouth in Wales Ben Guy 494 432 437 442 449 467 469 475 477 482 Bibliography 499 Manuscripts 552 General Index 555 For use by the Author only | © 2020 Koninklijke Brill NV Chapter 1 Geoffrey of Monmouth’s Welsh Sources Ben Guy Introduction: Britons, Bretons, and the Unworthy Welsh It has long been recognized that Geoffrey of Monmouth drew on sources originating from the Brittonic-speaking world. This fact is frequently mentioned in scholarly literature, though it is rarely accompanied by detailed supporting evidence. It was, after all, with the Britons, both contemporary and ancient, that Geoffrey was primarily concerned, and it was to the Britons that he looked for source material concerning the history of Britannia.1 One might legitimately ask whether it is possible, or even necessary, to distinguish between sources that originated from different Brittonic-speaking regions. It would appear that the three surviving Brittonic languages had not yet become mutually unintelligible by the 12th century. Gerald of Wales commented on this matter in the first recension of his Description of Wales, completed around 1194, some 60 years after the propagation of the De gestis Britonum: Indeed, Cornwall and Brittany use almost the same language, which is, nevertheless, still intelligible to the Welsh in many and almost in all cases, on account of their original relationship. Inasmuch as it is less refined and rougher, it is closer to the ancient British language, or so I think myself.2 1 To avoid confusion, I shall continue to employ the adjective “Brittonic” rather than “British” when referring to the medieval and ancient Britons. I avoid the term “Celtic”, which is meaningless in this context. 2 Gerald of Wales, The Description of Wales i.6, ed. J.F. Dimock, Giraldi Cambrensis Opera, 8 vols., London, 1861–91, vol. 6, pp. 153–228, at p. 177: “Cornubia vero et Armorica Britannia lingua utuntur fere persimili, Kambris tamen, propter originalem convenientiam, in multis adhuc et fere cunctis intelligibili. Quae, quanto delicata minus et incomposita magis, tanto antiquo linguae Britannicae idiomati magis, ut arbitror, appropriata.” Translation adapted from Gerald of Wales, The Journey Through Wales and The Description of Wales, trans. L. Thorpe, Gerald of Wales: The Journey through Wales / The Description of Wales, Harmondsworth, 1978, p. 231. For the dates of the recensions of The Description of Wales, see R. Bartlett, Gerald of Wales, 1146–1223, Oxford, 1982, p. 216. © The Author, 2020 | doi:10.1163/9789004410398_003 This is an open access chapter distributed under the terms of the CC BY-NC 4.0 license. For use by the Author only | © 2020 Koninklijke Brill NV 32 Guy Modern linguists would agree with Gerald’s observation about the mutual intelligibility of Medieval Cornish, Breton, and Welsh, and his consideration of the relationship between the modern languages and their ancient Brittonic precursor furnishes an interesting and early example of philological speculation.3 But linguistic factors had not forestalled the advent of a divergence in perceived identity and history. While all concerned were aware of their supposed descent from the ancient Britons, centuries of geographical separation and divergent historical development had caused the Britons of Cornwall, Brittany, and Wales to view themselves as distinct groups within the loosening Brittonic family. This process seems not to have been especially advanced by the 9th century. The Welsh Latin Historia Brittonum, written in 829 or 830, refers to the Bretons simply as Brittones Armorici, “Armorican Britons”, and no particular word is used to differentiate the Britons of Wales from Britons elsewhere.4 The Welshman Asser, writing later in the 9th century, simply uses the word Britannia, without further specificity, to describe Wales, just as the same word was used at that time to describe Brittany.5 Each was unambiguously a “land of the Britons”. By the middle of the 12th century, however, circumstances had definitively changed, and Welsh writers were rapidly turning their Britannia in the west of Britain into Wallia, “Wales”, and their fellow Britons into Walenses, “Welsh”, responding in part to new terminological distinctions introduced by their Anglo-Norman neighbors.6 On the other hand, the Britons of Brittany, in contrast to the Welsh, were able to continue flourishing successfully within the Anglo-Norman realm as self-identifying Britones, preserving the earlier terminology, which remains in use today.7 3 L. Fleuriot, “Langue et société dans la Bretagne ancienne”, in J. Balcou and Y. Le Gallo (eds.), Histoire littéraire et culturelle de la Bretagne, 3 vols., Paris, 1987, vol. 1, pp. 7–28, at p. 9; id., Dictionnaire des gloses en vieux Breton, Paris, 1964, pp. 13–14; J.E.C. Williams, “Brittany and the Arthurian Legend”, in R. Bromwich, A.O.H. Jarman, and B.F. Roberts (eds.), The Arthur of the Welsh: The Arthurian Legend in Medieval Welsh Literature (Arthurian Literature in the Middle Ages, 1), Cardiff, 1991, pp. 249–72, at pp. 253–54; J. Loth, L’Émigration bretonne en Armorique du Ve au VIIe siècle de notre ère, Rennes, 1883, p. 92. 4 Historia Brittonum (Harley 3859) §27, ed. Faral, LLA, pp. 2–62, at p. 21 (hereafter referred to as HB (Harl. 3859)). 5 E.g. Asser, Life of King Alfred §79, ed. W.H. Stevenson, Asser’s Life of King Alfred. Together with the Annals of Saint Neots, Oxford, 1959, pp. 63 and 65. See too the translation in S. Keynes & M. Lapidge, Alfred the Great: Asser’s Life of King Alfred and Other Contemporary Sources, Harmondsworth, 1983, pp. 93–94. 6 H. Pryce, “British or Welsh? National Identity in Twelfth-Century Wales”, EHR 116 (2001), 775– 801, at pp. 792–96. 7 P. Galliou & M. Jones, The Bretons, Oxford, 1991, pp. 181–82. For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 33 During this time, nobody was more keenly aware of such developments than Geoffrey of Monmouth. Had he not possessed an intricate understanding of the cultural self-awareness of different groups of Britons in his own time, he would not have been so careful to distinguish between the origins of the Cornish, Bretons, and Welsh in his history. Most remarkable is the distinction made between the Cornish and the rest of the Britons. Geoffrey attributed to the Cornish an ethnic distinction that arose prior to the foundation of Britain by Brutus. While Brutus and his band of Trojan exiles were navigating the Tyrrhenian sea, they encountered another group of Trojan exiles, descended through four generations from those who had fled from Troy with Antenor: Their leader was called Corineus, a just man and a good advisor, of great character and boldness … When the Trojans realised their common ancestry, they took Corineus and his people with them. Later they were called Cornish after their chief and in every battle proved more helpful to Brutus than the rest.8 Geoffrey later explains how, following the establishment of Brutus in Britain and the naming of his people as “Britons” after him, Corineus founded Cornwall, which he called Corineia after himself.9 Although Cornwall thereafter remains part of Britain for the remainder of Geoffrey’s account, the Cornish never lose their unique proclivity for excellence, as has been discussed by Oliver Padel.10 The Bretons and the Welsh, on the other hand, are, in no uncertain terms, latter-day Britons. The Bretons are the descendants of those Britons settled in Armorica by Maximianus, then king of Britain, during his campaign of conquest in Gaul: He [Maximianus] issued an edict to the effect that a hundred thousand common people should be gathered to be sent to him, as well as thirty thousand knights to protect them from hostile attack in the country they were to inhabit. Once all this was organised, he spread them throughout 8 9 10 DGB, i.17.330–36: “Erat eorum dux Corineus dictus, uir modestus, consilii optimus, magnae uirtutis et audaciae … Agnita itaque ueteris originis prosapia, associauerunt illum sibi nec non et populum cui praesidebat. Hic, de nomine ducis postmodum Cornubiensis uocatus, Bruto in omni decertatione prae ceteris auxilium praestabat.” DGB, i.21.462–67. O.J. Padel, “Geoffrey of Monmouth and Cornwall”, CMCS 8 (1984), 1–28. For use by the Author only | © 2020 Koninklijke Brill NV 34 Guy all the regions of Armorica, making it a second Britain [altera Britannia], which he presented to Conanus Meriadocus.11 Thenceforth, the Britons of Armorica in Geoffrey’s narrative are called Armorici Britones, “Armorican Britons”, or more simply Armoricani, and Brittany is altera Britannia, minor Britannia, or Armoricana Britannia. The extent to which Geoffrey favored the Armorican Britons over the Insular Britons is well known, and has led to the plausible suggestion that Geoffrey was himself of Breton origin.12 The contrast between Geoffrey’s portrayal of the Insular Britons and the Armorican Britons following the establishment of Brittany is emphasized most starkly in the speech that Geoffrey puts into the mouth of Salomon, king of the Armorican Britons, in his address to Caduallo, the recently exiled king of the Insular Britons: When the people of this new Britain of mine lived with your subjects in your Britain, it was the mistress of all the neighbouring realms, and there was no one who could conquer it except the Romans. And although they subjugated it for a time, the Romans were driven out shamefully, their governors lost and slain. But after my subjects came here, led by Maximianus and Conanus, the remaining Britons never again enjoyed the privilege of maintaining uninterrupted control of their land. Many of their leaders upheld the ancient prowess of their fathers, but more proved to be weaker heirs, who forgot it completely when their enemies attacked. Thus I am grieved by the weakness of your people, since we share the same origins and you are called British, just as we are, we who bravely protect this land you see from the attacks of all its neighbours.13 11 12 13 DGB, v.86.350–55: “Fecit itaque edictum suum ut centum milia plebanorum in Britannia insula colligerentur qui ad eum uenirent, praeterea triginta milia militum qui ipsos infra patriam qua mansuri erant ab hostili irruptione tuerentur. Cumque omnia perpetrasset, distribuit eos per uniuersas Armorici regni nationes fecitque alteram Britanniam et eam Conano Meriadoco donauit.” J.E. Lloyd, A History of Wales from the Earliest Times to the Edwardian Conquest, 2 vols., 3rd ed., London, 1939, vol. 2, pp. 523–24; id., “Geoffrey of Monmouth”, EHR 57 (1942), 460–68, at pp. 466–68; Tatlock, LHB, p. 443; B.F. Roberts, “Sylwadau ar Sieffre o Fynwy a’r Historia Regum Britanniae” [Remarks on Geoffrey of Monmouth and the Historia Regum Britanniae], Llên Cymru 12 (1972–73), 127–45, at pp. 128–29. For further discussion of this matter, see the Introduction above, pp. 11–19. DGB, xi.194.332–44: “Cum gens huius meae Britanniae una cum uestratibus in uestra Britannia cohabitaret, dominabatur omnium prouincialium regnorum, nec fuit uspiam populus praeter Romanos qui eam subiugare quiuisset. Romani autem, licet eam ad tempus subditam habuissent, amissis rectoribus suis ac interfectis cum dedecore expulsi For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 35 Through their continued degeneracy, the Insular Britons did not retain their cherished Brittonic nomenclature for long: As their culture ebbed, they were no longer called Britons, but Welsh, a name which owes its origin to their leader Gualo, or to queen Galaes or to their decline … The Welsh, unworthy successors to the noble Britons, never again recovered mastery of the whole island, but, squabbling pettily amongst themselves and sometimes with the Saxons, kept constantly massacring the foreigners or each other.14 Geoffrey therefore emphasizes his claimed historical distinctions between the Cornish, Bretons, and Welsh with clarity and consistency. The Cornish are a special group among the Britons, but their separate origin deep in legendary history serves to underscore their distinction from the Britons proper, and as a result they are not made to bear any responsibility for the Britons of the present. The Bretons and the Welsh, on the other hand, are the direct products of the later stages of Geoffrey’s historical arc: the former, the descendants of the Armorican Britons, have courageously maintained the spirit and name of their ancient forebears, whereas the latter, the descendants of the Insular Britons, have grown feeble through civil war, and have lost their right to the Brittonic name, becoming, instead, Welsh. By casting the Cornish as the remote descendants of Corineus’s merry band, Geoffrey effectively exonerates them from the charges that he is leveling against the Welsh, making the latter the unique witnesses to the decline of ancient Britannia. It is essential to appreciate Geoffrey’s presentation of the various Brittonic peoples in order to interrogate his use of source material emanating from the Brittonic regions properly. For instance, when Geoffrey refers to his infamous “very old book in the British tongue”, which, as stated at the very end of the DGB, had been brought by Walter, archdeacon of Oxford, “ex Britannia”, it 14 abscesserunt. Sed postquam Maximiano et Conano ducibus ad hanc uenerunt prouinciam, residui qui remanserunt numquam eam deinceps habuerunt gratiam ut diadema regni continue haberent. Quamquam enim multi principes eorum antiquam patrum dignitatem seruarent, plures tamen debiliores heredes succedebant, qui eam penitus inuadentibus hostibus amittebant. Vnde debilitatem populi uestri doleo, cum ex eodem genere simus et sic Britones nominemini sicut et gens regni nostri, quae patriam quam uidetis omnibus uicinis aduersatam uiriliter tuetur.” DGB, xi.207.592–94 and 598–600: “Barbarie etiam irrepente, iam non uocabantur Britones sed Gualenses, uocabulum siue a Gualone duce eorum siue a Galaes regina siue a barbarie trahentes … Degenerati autem a Britannica nobilitate Gualenses numquam postea monarchiam insulae recuperauerunt; immo nunc sibi, interdum Saxonibus ingrati consurgentes externas ac domesticas clades incessanter agebant.” For use by the Author only | © 2020 Koninklijke Brill NV 36 Guy seems probable, given the overall thrust of Geoffrey’s narrative, that Britannia here is intended to refer to Brittany rather than Wales.15 Although Welshmen writing in Latin in the late 11th and early 12th centuries could still refer to Wales as Britannia,16 Geoffrey makes it quite clear that, within the terms of his history, Wales had long forfeited that hallowed name, and that the only remaining Brittonic Britannia was altera Britannia, or Brittany. It is partly for this reason that efforts to equate Geoffrey’s avowed source-book with a manuscript containing a historical compilation similar to the expanded version of the Historia Brittonum found in London, British Library, Harley 3859 must ultimately fail.17 Although Geoffrey certainly did use a compilation of exactly that type, as is discussed below, he is unlikely to have found it in a book brought out of Brittany. Geoffrey’s “very old book in the British tongue” is no more than a rhetorical device intended to lend his account credence and mystery, as has long been recognized by scholars.18 This is not to say that Geoffrey did not use Breton sources, nor even that Walter did not provide Geoffrey with some ancient book to translate,19 but, as with so many other aspects of Geoffrey’s work, one cannot assume that his description of that book was designed for anything other than rhetorical impact; it is not a statement of historical fact. Nevertheless, whether Geoffrey’s book existed in reality or merely in rhetoric, there seems little reason to doubt that Geoffrey intended his contemporaries to believe that he had translated a book written in the Brittonic language. William of Newburgh, for one, bemoaned that Geoffrey had sought to translate fictitious accounts of the Britons into Latin.20 15 16 17 18 19 20 DGB, Prologus 2.9–10 and xi.208.605. Alternatively, Joshua Byron Smith has suggested that the phrase “ex Britannia” refers to ancient Britain, rather than contemporary Brittany or Wales. See Introduction above, pp. 21–24. Pryce, “British or Welsh?” pp. 777–78. E.g. S. Piggott, “The Sources of Geoffrey of Monmouth: I. The ‘Pre-Roman’ King-List”, Antiquity 15 (1941), 269–86. See now S. Harris, The Linguistic Past in Twelfth-Century Britain, Cambridge, 2017, pp. 91– 99. In the Introduction above, pp. 21–24, Joshua Byron Smith argues that Geoffrey borrowed the “ancient British book” device from Dares Phrygius. See also the similar comments in Roberts, “Sylwadau”, pp. 134–35 and Harris, Linguistic Past, pp. 93–94. Further evidence for Walter’s book is provided by Geffrei Gaimar: see I. Short, “Gaimar’s Epilogue and Geoffrey of Monmouth’s Liber vetustissimus”, Speculum 69:2 (1994), 323–43. See also Le Saux’s chapter in this volume. William of Newburgh, The History of English Affairs i.3, ed. and trans. P.G. Walsh and M.J. Kennedy, William of Newburgh: The History of English Affairs, Book I (Edited with Translation and Commentary), Warminster, 1988, pp. 28–29; cf. B. Guy, “Gerald and Welsh Genealogical Learning”, in G. Henley and A.J. McMullen (eds.), Gerald of Wales: New Perspectives on a Medieval Writer and Critic, Cardiff, 2018, 47–61, at p. 52; Harris, Linguistic Past, p. 95. For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 37 Perhaps it was Geoffrey’s chosen presentation of the divergent historical development of the Insular and Armorican Britons that led him to promulgate the idea that his most lauded authority, Gildas, was formerly a resident of Brittany. Geoffrey used Gildas’s The Ruin of Britain extensively, as has been expertly demonstrated by Neil Wright.21 In the DGB, Geoffrey refers to Gildas by name no less than seven times, usually attributing to him some account or information that can be found nowhere in Gildas’s work (nor, for the most part, in the Historia Brittonum, which Geoffrey might also have known to be attributed to Gildas).22 His personal convictions aside, it probably suited the temper of Geoffrey’s work to locate such a venerable Briton as Gildas in Brittany rather than Britain; Wales certainly would not do. In the VM, Merlin explains to his sister that he wishes to speak with Telgesinus (Geoffrey’s version of the legendary Welsh poet Taliesin), who had recently returned “from Armorican parts”, where he had been receiving instruction from Gildas.23 The idea that Gildas resided in Brittany was not invented by Geoffrey. It is found in the earliest Life of St Gildas, written in St-Gildas-de-Ruys in Brittany, probably in the 11th century.24 Not everyone agreed; Gildas does not retire to Brittany in the Life of St Gildas composed by Caradog of Llancarfan for Glastonbury Abbey, sometime in the middle of the 12th century, presumably because this would have contradicted his claim that Gildas retired to Glastonbury.25 In this instance, it would no doubt have suited Geoffrey to follow a Breton view over a Welsh view. In Geoffrey’s terms, it seems, Gildas was no Welsh source. Geoffrey in fact makes no mention whatsoever of having drawn on any source material of Welsh provenance. And yet he most assuredly did so, and to a considerable 21 22 23 24 25 N. Wright, “Geoffrey of Monmouth and Gildas”, AL 2 (1982), 1–40; id., “Geoffrey of Monmouth and Gildas Revisited”, AL 5 (1985), 155–63. See also his “Geoffrey of Monmouth and Bede”, AL 6 (1986), 27–59. See below, pp. 49–50. VM, ll. 684–88: “de partibus Armoricanis”. Life of St Gildas §16, ed. and trans. H. Williams, Two Lives of Gildas: By a Monk of Ruys, and Caradoc of Llancarfan, Felinfach, 1990, pp. 36–37 (repr. from H. Williams, ed., Gildas, 2 parts (Cymmrodorion Record Series, 3), London, 1899–1901, vol. 2, pp. 315–420, at pp. 346–47). J.S.P. Tatlock, “Caradoc of Llancarfan”, Speculum 13:2 (1938), 139–52, at pp. 140–42; id., “The Dates of the Arthurian Saints’ Legends”, Speculum 14:3 (1939), 345–65, at pp. 352–53, n. 1; A. Gransden, “The Growth of the Glastonbury Traditions and Legends in the Twelfth Century”, The Journal of Ecclesiastical History 27 (1976), 337–58, at pp. 340 and 346 (repr. in J.P. Carley (ed.), Glastonbury Abbey and the Arthurian Tradition, Cambridge, 2001, 29–53); J. Scott, The Early History of Glastonbury: An Edition, Translation and Study of William of Malmesbury’s De Antiquitate Glastonie Ecclesie, Woodbridge, 1981, p. 3. For use by the Author only | © 2020 Koninklijke Brill NV 38 Guy degree. Might it be that Geoffrey deliberately suppressed any acknowledgement of his debt to Welsh materials in order to emphasize the role of his putative “Breton” book as the sole conduit of Brittonic historical authority? How could Geoffrey’s Gualenses have accurately preserved the ancient traditions of their Brittonic forebears when they had become so unworthy, having lost the very name of Briton? It may have been a problem for Geoffrey’s designs that, although he wished to emphasize the martial prowess and moral dignity of the Bretons over the Welsh, the overwhelming majority of written sources of Brittonic origin that he could discover seem to have been written in Wales, and concerned Wales to a far greater degree than the other Brittonic regions. This is suggested most persuasively by Geoffrey’s spellings of personal and placenames, which almost invariably display Old Welsh rather than Old Breton features.26 The bias toward Welsh displayed by Geoffrey’s spellings might be the product of a comparative dearth of relevant Breton sources available to him. Such a dearth is certainly apparent today, for there are few native compositions surviving from early medieval Brittany aside from saints’ lives and charters. Indeed, the seeming near-absence of extant written sources from early medieval Brittany relating to the activities of kings, perhaps as noticeable in Geoffrey’s day as it is now, may be more than an accident of textual survival. It has been argued that vertical power structures in Brittany were relatively weak during the early Middle Ages and were such as to obviate the need for the types of king-populated historical texts that provided intellectual legitimization for medieval states and their rulers. By contrast, texts of this kind were actively produced in early medieval Wales, where regal authority was much better established throughout the period.27 The legitimizing historical texts in question include origin legends, chronicles, and genealogies, genres of writing that gained wide currency during the Middle Ages because of their utility for conferring legitimacy upon contemporary political authority. It was no doubt the Brittonic purview and linguistic orientation of such texts that made them so 26 27 T.D. Crawford, “On the Linguistic Competence of Geoffrey of Monmouth”, Medium Ævum 51 (1982), 152–62, at p. 156. For example, one finds the diphthong /au/ < /ɔ:/ spelled au, a development peculiar to Welsh among the Brittonic languages, in Cledaucus, Ebraucus, Enniaunus, Gualauc, Kaerebrauc, Maglaunus, Mapcledauc, and Rudaucus. For this sound change, see P. Sims-Williams, “The Emergence of Old Welsh, Cornish and Breton Orthography, 600–800: the Evidence of Archaic Old Welsh”, BBCS 38 (1991), 20–86, at pp. 63–71. C. Brett, “Breton Latin Literature as Evidence for Literature in the Vernacular, A.D. 800– 1300”, CMCS 18 (1989), 1–25, at pp. 19–25; ead., “Soldiers, Saints, and States? The Breton Migrations Revisited”, CMCS 61 (2011), 1–56, at pp. 38–43. For a recent view of the origins of this situation in Brittany (with references to earlier literature), see B. Guy, “The Breton Migration: a New Synthesis”, Zeitschrift für celtische Philologie 61 (2014), 101–56. For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 39 useful to Geoffrey for his great literary venture. On the other hand, it is worth acknowledging that the bias toward Welsh sources may be illusory: if the lack of comparable texts from early medieval Brittany is indeed an accident of textual survival and does not reflect what would have been available to Geoffrey, then we might underestimate Geoffrey’s use of such sources, especially considering the dangers highlighted below surrounding arguments from silence. The remainder of this chapter examines the sources of Welsh origin that Geoffrey can be shown to have used in his work. The first section considers Geoffrey’s linguistic abilities and the extent to which his access to Welsh source material might have been hindered by a language barrier. The second section explores his use of the Historia Brittonum, and of the annals and genealogies that accompanied the version of the Historia Brittonum at his disposal. The third section briefly draws attention to Geoffrey’s access to ecclesiastical texts of Welsh provenance, such as Rhygyfarch’s Life of St David and De situ Brecheniauc (“Concerning the Establishment of Brycheiniog”). The fourth and final section turns to the contentious issue of the relationship between Geoffrey’s work and Welsh poetry. 1 The British Tongue Brutus named the island of Britain after himself and called his followers Britons. He wanted to be remembered for ever for giving them his name. For this reason the language of his people, previously known as Trojan or “crooked Greek”, was henceforth called British.28 So Geoffrey describes the origins of the British tongue. It has been pointed out that his designation of British as curuum Graecum, “crooked Greek”, relies on an etymology of Cymraeg (the Welsh word for the Welsh language) that could only arise from direct knowledge of Welsh.29 Cymraeg has here been etymologized as Welsh cam Roeg, literally “crooked Greek”. The loss of the G in Roeg (from Welsh Groeg) is not a liberty on Geoffrey’s part; it is a grammatically regular change in the second element of a compound in Welsh, showing that the person responsible for the etymology had more than a superficial knowl28 29 DGB, i.21.459–62: “Denique Brutus de nomine suo insulam Britanniam appellat sociosque suos Britones. Volebat enim ex diriuatione nominis memoriam habere perpetuam. Vnde postmodum loquela gentis, quae prius Troiana siue curuum Graecum nuncupabatur, dicta fuit Britannica.” Harris, Linguistic Past, p. 93; Crawford, “On the Linguistic Competence of Geoffrey”, pp. 155–57; Roberts, “Sylwadau”, p. 137, n. 45; Tatlock, LHB, p. 445, n. 39. For use by the Author only | © 2020 Koninklijke Brill NV 40 Guy edge of the language. Unfortunately, one cannot now know whether Geoffrey invented the etymology or was informed about it by another Welsh speaker. Various attempts have been made to determine Geoffrey’s linguistic ability, and in particular his competence in one or more of the Brittonic languages.30 This has been deemed important for establishing the veracity of his claim that the DGB was translated from a “very old book in the British tongue”, and for judging the likelihood that he could have employed other vernacular sources with success. However, the value of framing the problem in this way is highly questionable. It has already been noted that Geoffrey’s alleged source-book is a rhetorical device, rendering somewhat futile the attempt to establish whether he could feasibly have translated it. Secondly, it is always hazardous to claim that Geoffrey “misunderstood native material” and extrapolate from that that his command of Welsh (or Breton) was less than firm.31 Geoffrey had no interest in reproducing his source material exactly. At every stage in his works, he crafted the accounts that he found in his sources so that they blended seamlessly with the majestic progression of his imagined history. One underestimates the intimacy between Geoffrey and his sources at great peril. Geoffrey’s self-proclaimed epithet, Monemutensis, “of Monmouth”, seems to imply that he was brought up in or around Monmouth on the southern border between Wales and England, presumably in the late 11th or early 12th century.32 This is significant because, by 1075, the lord of Monmouth was Wihenoc of La Boussac, one of the many Breton followers of William the Conqueror.33 Such a state of affairs might provide a plausible context for Geoffrey’s positive portrayal of the Bretons. It is indeed quite possible that Geoffrey’s own family arrived in Wales in the wake of the establishment of Wihenoc as lord of Monmouth. As Sir Rees Davies astutely observed, “Geoffrey’s father may well have been a 30 31 32 33 Tatlock, LHB, p. 445; Crawford, “On the Linguistic Competence of Geoffrey”. B.F. Roberts, “Geoffrey of Monmouth and Welsh Historical Tradition”, Nottingham Medieval Studies 20 (1976), 29–40, at p. 36; cf. id., “Geoffrey of Monmouth, Historia regum Britanniae, and Brut y Brenhinedd”, in Bromwich et al. (eds.), The Arthur of the Welsh, pp. 97–116, at pp. 109–10; Piggott, “Sources”, p. 282. For a good overview of Geoffrey’s life, see J.C. Crick, “Monmouth, Geoffrey of (d. 1154/5)”, Oxford Dictionary of National Biography, Oxford University Press, 2004, <http:// www.oxforddnb.com/view/article/10530> (accessed 27 June 2018). H. Guillotel, “Une famille bretonne au service du Conquérant: Les Baderon”, in Droit privé et institutions régionals: Etudes historiques offertes à Jean Yver, Paris, 1976, pp. 361–66; K.S.B. Keats-Rohan, “The Bretons and Normans of England 1066–1154: the Family, the Fief and the Feudal Monarchy”, Nottingham Medieval Studies 36 (1992), 42–78, at p. 49; ead., Domesday People: A Prosopography of Persons Occurring in English Documents, 1066–1166. I. Domesday Book, Woodbridge, 1999, pp. 54–55. For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 41 first- or second-generation Breton settler in Monmouth, an area rich in opportunities, formal and informal, for contacts between settlers and natives.”34 What would such a scenario imply about Geoffrey’s linguistic abilities? It might be instructive to indulge in a little speculation, if only to realize the plurality and complexity of the possibilities. If Geoffrey’s father were indeed a first- or second-generation Breton settler in Monmouth, it is very likely that T.D. Crawford was correct to assert that Geoffrey’s first language would probably have been Anglo-Norman French.35 Fluency in French would have been an essential tool for enabling Geoffrey to interact with friends and patrons in Monmouth and in his later home in Oxford. It is indeed entirely possible that Geoffrey’s hypothetical “Breton” ancestors were French- rather than Breton-speaking before they came to Britain.36 On the other hand, it is equally possible that Geoffrey’s family was Breton-speaking, and that Breton remained the private language of the family for a few generations after they had settled in Monmouth, even though French would have dominated their interactions in the public sphere. One suspects that Geoffrey’s perceived competence in Breton is implied in his claim to have translated the “very old book in the British tongue”. Although, as discussed above, the claim is unlikely to have been literally true, its rhetorical impact was presumably predicated on its assumed plausibility to contemporaries. The claim was read by those who knew Geoffrey and whom Geoffrey wanted to judge him favorably. Whatever he claimed about the contents of Walter’s alleged Breton book, it is difficult to believe that Geoffrey would have professed himself to his associates as the translator of a long Breton narrative had he no observable familiarity with the language. Whatever his family’s origins, it cannot be doubted that Geoffrey, growing up in Monmouth, would have had a long acquaintance with Welsh. If Geoffrey were a Breton-speaker of any competence, one would imagine that Welsh would not have been unduly challenging for him, and that he could have rapidly become comfortable reading the written language, especially since the spelling systems of Old Breton and Old Welsh (and indeed Old Cornish) were so similar. Even if Geoffrey knew nothing of Breton, Welsh would not have been inaccessible to him. No more evidence of Geoffrey’s linguistic adeptness is required than the substantial Latin compositions that flowed from his 34 35 36 R.R. Davies, The Age of Conquest: Wales, 1063–1415, Oxford, 2000, p. 106. Crawford, “On the Linguistic Competence of Geoffrey”, pp. 152–53. Tatlock similarly commented that “no doubt one of his vernaculars was Norman-French”: Tatlock, LHB, p. 445. Cf. Roberts, “Sylwadau”, p. 128, n. 9; Crawford, “On the Linguistic Competence of Geoffrey”, p. 157. For use by the Author only | © 2020 Koninklijke Brill NV 42 Guy pen, which afford ample testimony to his confident Latinity. Had he applied the same ability to Welsh in support of his academic interests, he might have acquired considerable facility with at least the written language, if not also the spoken. It is likely that Geoffrey received his early education in one of the churches of south-east Wales, and we have other evidence (such as the vernacular description of Llandaff’s privileges, known as Braint Teilo) for the cultivation of written Welsh in scholarly circles in the south-eastern churches of Geoffrey’s day.37 Though most of the comments above are ultimately speculative, they should hopefully make the point that Geoffrey was, at the least, multilingual and proficient at linguistic study. Modern scholars will never be in a position to judge Geoffrey’s exact knowledge of Welsh or Breton. The only safe assumption is that language would not have been an insurmountable barrier between Geoffrey and the sources that he wished to access. With this in mind, we may venture forth, with Geoffrey, into Gualia. 2 The History of the Britons He was grieved, however, that his brother Nennius lay between life and death, seriously injured; for the wound Caesar had inflicted in their duel had proved incurable.38 It has always been clear to students of Geoffrey that the Historia Brittonum was one of the primary sources of inspiration for the DGB. The Historia Brittonum is an account of the Britons written in Latin and produced in Gwynedd, in North Wales, in 829 or 830. One of the three branches of the Historia Brittonum’s textual tradition contains a prologue in which the author of the text identifies himself as one Nennius, but the authenticity of this prologue has been disputed by modern critics.39 Geoffrey’s Nennius, brother of Lud and Cassibellaunus, 37 38 39 For Braint Teilo, see now P. Russell, “Priuilegium Sancti Teliaui and Breint Teilo”, Studia Celtica 50 (2016), 41–68. DGB, iiii.57.78–81: “Angebatur tamen ex alia parte dolore, quia frater suus Nennius, letaliter uulneratus, in dubio uitae iacebat; uulnerauerat enim illum Iulius in supradicto congressu et plagam inmedicabilem intulerat.” David Dumville argued that the prologue was a later concoction in which the work was attributed to Nennius because of his fame as a scholar of the Britons: D.N. Dumville, “ ‘Nennius’ and the Historia Brittonum”, Studia Celtica 10/11 (1975–76), 78–95. Others have argued that the prologue is more likely to have been a part of the original Historia Brittonum: P.J.C. Field, “Nennius and his History”, Studia Celtica 30 (1996), 159–65; B. Guy, For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 43 is unlikely to bear any relation to the author of the Historia Brittonum. Then again, it is something of a pleasing irony to read of Geoffrey’s Nennius fighting so valiantly against Caesar. The author of the Historia Brittonum was, after all, the first known writer to portray Caesar’s assault on Britain from a point of view sympathetic to the Britons, following almost nine centuries of historiographical defamation that began with Caesar’s own account in The Gallic Wars. As Geoffrey remarks, “Nennius congratulated himself on being able to exchange even a single blow with so famous a man.”40 The Historia Brittonum is a synthetic account of the Britons from their origins to their wars with the English kings in the 7th century, assembled from a variety of sources, including origin legends, saints’ lives, and genealogies, as well as popular Latin texts such as Jerome’s translation of Eusebius’s universal chronicle, Gildas’s The Ruin of Britain, and Bede’s Ecclesiastical History.41 Many of the most famous incidents in Geoffrey’s history appear in their earliest recorded forms in the Historia Brittonum. These include the settlement of Britain by Britto/Brutus; the foundation of Brittany by Maximianus; the tale of Vortigern, Hengist, and the Treachery of the Long Knives; the account of the two embattled dragons of Snowdonia; and, of course, the catalogue of Arthur’s victories against the Saxons. In the Historia Brittonum, these events are only loosely connected, and do not act as components of an integrated political narrative. Geoffrey, however, wove the Historia Brittonum’s disjointed episodes into a coherent story with uncanny sleight of hand. It is argued below that Geoffrey did not draw on the Historia Brittonum indiscriminately. Instead, he carefully incorporated some episodes, altered others, and left some out altogether. He was nevertheless acutely conscious of the original meanings of the episodes and indicated as much in his renditions of them. There is evidence that Geoffrey was familiar with the “Harleian” recension of the Historia Brittonum, as well as with the Welsh annals and genealogies that are interpolated into the copy of the Historia Brittonum in the Harley manuscript from which the recension is named. For instance, the annals probably enabled Geoffrey to deduce his famous date for the battle of Camlan, while the genealogies offered Brittonic name forms that were used in many parts of 40 41 “The Origins of the Compilation of Welsh Historical Texts in Harley 3859”, Studia Celtica 49 (2015), 21–56, at pp. 45–54. DGB, iiii.56.57–58: “Nennius ultra modum laetatur se posse uel solum ictum tanto uiro ingerere.” For general accounts of the text, see D.N. Dumville, “Historia Brittonum: An Insular History from the Carolingian Age”, in A. Scharer and G. Scheibelreiter (eds.), Historiographie im frühen Mittelalter, Wien, 1994, pp. 406–34; id., “The Historical Value of the Historia Brittonum”, AL 6 (1986), 1–26. For use by the Author only | © 2020 Koninklijke Brill NV 44 Guy the history. Moreover, since Geoffrey’s copy of the genealogies corresponded in certain respects to the version used in places and in texts connected to Caradog of Llancarfan, it is suggested that the latter might have provided Geoffrey with his copy of the Harleian recension of the Historia Brittonum, interpolated with the relevant annals and genealogies. A good example of how Geoffrey borrowed episodes from the Historia Brittonum, but recrafted them to suit his own designs, is provided by the legend of Vortigern and the two dragons.42 Many aspects of the story are shared by the versions in the Historia Brittonum and the DGB. In both, the Saxons rebel against Vortigern, who flees westward to Snowdonia. There he orders a fortress to be built, but on each day the previous day’s construction work has mysteriously disappeared. To remedy the situation, his magi advise that the foundations of the fortress be sprinkled with the blood of a boy without a father. Such a boy is duly located, but, once he is brought into Vortigern’s presence, the boy questions the advice of the magi and instructs the king to dig underneath the foundations to discover the real explanation for the problem. The boy had rightly perceived that the foundations of the fortress are unstable because they had been built over a pool of water. Within the pool, moreover, are two dragons, who begin to fight once they are revealed. As the boy explains, one dragon is red, representing the Britons, while the other is white, representing the Saxons. The combat between the two signifies the struggle for supremacy in Britain. The Historia Brittonum briefly explains that the red dragon will ultimately be victorious, but in the DGB matters are made rather more complex by the introduction of Merlin’s long prophecy. Although the versions told in the Historia Brittonum and the DGB run in parallel insofar as the elements described above are concerned, Geoffrey’s subtle changes of emphasis impart significant new shades of meaning to the tale. In the Historia Brittonum, the basis of the story is onomastic. It is obvious that the fortress in question is Dinas Emrys in Snowdonia, since at the end of the story the fatherless boy reveals his name to be Ambrosius (the Latin name from which Welsh Emrys derives), and consequently, as the narrator explains, “he was seen to be Emrys Wledig”, who was presumably a well-known figure of legend in North Wales in the early 9th century.43 The story thus “explains” how the fortress acquired its name. Furthermore, because it was evidently understood that the Welsh name Emrys was equivalent to Latin Ambrosius, the name allowed the author of the Historia Brittonum to fashion an additional link between the story of the fortress and the period of Vortigern’s kingship. The boy 42 43 HB (Harl. 3859) §§40–42, ed. Faral, vol. 3, pp. 30–33. HB (Harl. 3859) §42, ed. Faral, vol. 3, p. 32: “Embreis Guletic ipse videbatur.” For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 45 reveals that his father was actually a consul of the Roman people, implying that this Ambrosius was the Ambrosius Aurelianus of Gildas, whose parents are specified to have been Roman nobles.44 We are then told that Vortigern gave Ambrosius “the fortress with all the kingdoms of the western region of Britain, and he himself with his magi went to the northern region”.45 It thus appears that the author of the Historia Brittonum used the story to explain the transfer of power in western Britain from Vortigern to Ambrosius. Geoffrey was certainly aware of the political implication of the Historia Brittonum’s version of the story, but he put his own spin on the tale by refocusing it on the prophet Merlin. In Geoffrey’s version, Merlin takes the place of Ambrosius as the fatherless boy summoned to the fortress in Snowdonia. At one point, Geoffrey alludes to the Historia Brittonum’s portrayal of events by ambiguously referring to the boy as Ambrosius Merlinus; one suspects that he understood the onomastic implication of the Historia Brittonum’s story and wished to preserve that feature in his account, even if it no longer provided a central element.46 But the aspect of the Historia Brittonum’s story that most enthralled Geoffrey was the boy’s ability to explain the meaning of the warring dragons, for it was this that prompted the introduction of Merlin’s prophecy in Geoffrey’s version. Geoffrey’s Merlin is based on Myrddin, the prophet of Welsh legend, who is discussed in more detail below. It has been suggested that Geoffrey changed the name to “Merlin” in order to evade the unfortunate coincidence in spelling between Myrddin and the French word merde, meaning “excrement”.47 Geoffrey inserts a subtle indication of his awareness of the Welsh name by having Vortigern’s envoys find Merlin not in campus Elleti in Glywysing, as in the Historia Brittonum, but in Kaermerdin, “Carmarthen”, the second element of which in the Welsh version of the name (modern Welsh “Caerfyrddin”) is indeed Myrddin. Geoffrey again demonstrates his appreciation of the Historia Brittonum’s version of the story in the way that he ends his account. Due to the change in the identity of the boy, the story can no longer end with Vortigern’s granting power in western Britain to Ambrosius. However, Geoffrey shapes his narrative so as to preserve the same chronological sequence, and in the first sentence following Merlin’s final prophecy he immediately states that “As soon as the 44 45 46 47 Gildas, The Ruin of Britain §25.3, ed. and trans. M. Winterbottom, Gildas: The Ruin of Britain and Other Works (Arthurian Period Sources, 7), Chichester, 1978, pp. 28 and 98. HB (Harl. 3859) §42, ed. Faral, vol. 3, p. 32: “arcem … cum omnibus regnis occidentalis plagae Brittanniae, et ipse cum magis suis ad sinistralem plagam pervenit.” DGB, Prophetiae 111.31. Cf. Tatlock, LHB, p. 175. For an alternative suggestion, see P. Russell, Vita Griffini filii Conani. The Medieval Latin Life of Gruffudd ap Cynan, Cardiff, 2005, pp. 125–26. For use by the Author only | © 2020 Koninklijke Brill NV 46 Guy next day dawned, Aurelius Ambrosius and his brother landed, accompanied by ten thousand knights.”48 Ambrosius had set off, it should be noted, from Brittany; Geoffrey here introduces the Bretons into the story even when they were lacking entirely from his source. Just as interesting as the episodes of the Historia Brittonum that Geoffrey incorporated into his history are the episodes that he silently discarded. These include the Historia Brittonum’s account of St Germanus and Cadell Dyrnllug, which in the Historia Brittonum was designed to provide an explanation for the origins of the kings of Powys.49 Germanus is given only very summary treatment in the DGB, presumably because Geoffrey did not wish to dwell upon the Pelagian heresy, which the historical Germanus was sent to Britain to eradicate.50 It is an interesting feature of Geoffrey’s history that he omits all mention of Powys, despite his evident enthusiasm for employing authenticlooking names for the various ancient kingdoms, lordships, and peoples in his narrative. Perhaps Powys could not be integrated neatly into the DGB’s geopolitical scheme; Geoffrey is quite explicit at one point that the Venedoti are the Norgualenses, “North Welsh”, and the Demetae are the Suthgualenses, “South Welsh”.51 The VM is equally clear about the division of Wales between the Venedoti and the Demetae, leaving no room for the Historia Brittonum’s Povisi.52 Geoffrey’s reluctance to grant Powys a place in his history presumably reflects the kingdom’s relative lack of importance in the centuries prior to Geoffrey’s lifetime; it was probably only during the early decades of the 12th century that Powys re-emerged as a significant Welsh kingdom.53 Geoffrey’s account of Arthur’s early battles against the Saxons owes much to the Historia Brittonum, but, again, he has not followed his source slavishly. While the Historia Brittonum names nine sites at which twelve battles were fought by Arthur, Geoffrey selected only four: the river Duglas, the province of Lindsey, the forest of Colidon, and the hillside in the region of Bath (pagus Badonis).54 More significantly, Geoffrey added a crucial element to Arthur’s campaigns 48 49 50 51 52 53 54 DGB, viii.118.22–23: “Nec mora, cum crastina dies illuxit, applicuit Aurelius Ambrosius cum germano suo, decem milibus militum comitatus.” HB (Harl. 3859) §§32–35, ed. Faral, vol. 3, pp. 23, 25, and 27. DGB, vi.101.369–76. For a recent treatment of St Germanus and Britain, see A.A. Barrett, “Saint Germanus and the British Missions”, Britannia 40 (2009), 197–217. DGB, ix.156.329–30. VM, ll. 21 and 26. D. Stephenson, Medieval Powys: Kingdom, Principality and Lordships, 1132–1293, Woodbridge, 2016, ch. 1. The idea of a tripartite division of Wales between Gwynedd, Powys, and Deheubarth only emerged from the second half of the 12th century, as witnessed by the writings of Gerald of Wales and the Welsh lawbooks. HB (Harl. 3859) §56, ed. Faral, vol. 3, pp. 38–39; DGB, ix.143–47. For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 47 that was entirely absent from the Historia Brittonum: the Bretons. Here we are confronted with an indication of the potential difficulty that Geoffrey may have encountered while writing a history of the Britons which was favorable to the Bretons but which used primarily Welsh source material. Following the establishment of the Armorican Britons by Maximianus, the Historia Brittonum makes no further mention of Armorica or its Brittonic-speaking inhabitants. For Geoffrey, however, the Armorican Britons become a constant source of strength and support for the Insular Britons. Arthur is no exception. Having no choice but to lift the siege of York due to the overwhelming numbers of the enemy, Arthur and his counselors determine to seek the assistance of Arthur’s nephew Hoel, king of the Armorican Britons, who dutifully comes to support his uncle with 15,000 men. Only then is Arthur able to continue his campaigns and, together with Hoel, defeat the Saxons in the province of Lindsey. A more surprising source for the DGB is the Historia Brittonum’s collection of mirabilia, “wonders” or “marvels”. Shortly after Arthur’s final victory over the Scots and Picts, Hoel finds himself amazed by the 60 rivers, islands, crags, and eagles’ nests of Loch Lomond, where Arthur had recently blockaded his enemies for a fortnight.55 The same features are attributed to Loch Lomond in the Historia Brittonum.56 In a curious aside, Arthur then tells Hoel about two other wonders, which also derive from the Historia Brittonum.57 It is not at all clear why these descriptions have been included in Geoffrey’s narrative. A debt to the Historia Brittonum more profound than the sum of the individual episodes transferred into the DGB is implicit in the overall scope and conception of Geoffrey’s historical project. Geoffrey’s account ranges from the fall of Troy to the death of Cadualadrus in 689. Throughout this entire period, Geoffrey’s Britons enjoy almost unbridled sovereignty over the island of Britain. Geoffrey’s decision to extend the supremacy of the Britons as far as the late 7th century had profound consequences for the ways in which later writers conceived the advent of English rule in Britain.58 Yet it was a decision that accorded with his Welsh sources. The Historia Brittonum, though written in the 55 56 57 58 DGB, ix.149–50. HB (Harl. 3859) §67, ed. Faral, vol. 3, p. 58. See A. Woolf, “Geoffrey of Monmouth and the Picts”, in W. McLeod (ed.), Bile ós Chrannaibh: A Festschrift for William Gillies, Ceann Drochaid, 2010, pp. 269–80, at pp. 273–76. Note that John Morris, in his translation of the Historia Brittonum, incorrectly translates stagnum Lumonoy as “Loch Leven” rather than “Loch Lomond”: Historia Brittonum, ed. and trans. J. Morris, Nennius: British History and the Welsh Annals (Arthurian Period Sources, 8), London, 1980, p. 40. For a possible source of Morris’s confusion, see Woolf, “Geoffrey of Monmouth”, p. 275. HB (Harl. 3859) §§69–70, ed. Faral, vol. 3, p. 59. R.W. Leckie, Jr., The Passage of Dominion: Geoffrey of Monmouth and the Periodization of Insular History in the Twelfth Century, Toronto, 1981. For use by the Author only | © 2020 Koninklijke Brill NV 48 Guy 9th century, does not mention any events later than the battle of Nechtansmere in 685,59 and the latest king of the Britons mentioned is Cadwaladr son of Cadwallon, who was “reigning among the Britons after his father” during the reign of Oswiu, king of Northumbria (642–70).60 This is the Cadwaladr who appears in the early medieval pedigree of the kings of Gwynedd in North Wales, and indeed the Historia Brittonum designates his father Cadwallon as rex Guenedotae regionis, “king of the kingdom of Gwynedd”, on two separate occasions.61 The significance of Cadwaladr’s terminal position within the context of the Historia Brittonum is very difficult to judge, because the part of the text dealing with the 7th century is structured around a collection of early English genealogies and a Northumbrian king-list, and the fragments of narration interpolated therein lack continuity and integration.62 However, the significance of the Historia Brittonum’s reluctance to peer beyond the reign of Cadwaladr should not be overlooked. An important point of comparison is the 10th-century Welsh prophetic poem Armes Prydein Vawr (“The Great Prophecy of Britain”).63 This poem is the earliest surviving text in which a certain Cadwaladr appears as one of the two deliverers of the Britons, who are prophe59 60 61 62 63 HB (Harl. 3859) §57, ed. Faral, vol. 3, p. 39. HB (Harl. 3859) §64, ed. Faral, vol. 3, p. 43: “regnante apud Brittones post patrem suum”. 8th-century figures do occur in the Historia Brittonum among its genealogies of English kings, but they are accorded no attention beyond the simple mention of their names. See D.N. Dumville, “The Anglian Collection of Royal Genealogies and Regnal Lists”, Anglo-Saxon England 5 (1976), 23–50, at p. 45; K.H. Jackson, “On the Northern British Section in Nennius”, in N.K. Chadwick (ed.), Celt and Saxon: Studies in the Early British Border, Cambridge, 1963, rev. ed. 1964, pp. 20–62, at pp. 22 and 60–61. HB (Harl. 3859) §61 and §64, ed. Faral, vol. 3, pp. 41 and 43. For the genealogy, see Early Welsh Genealogical Tracts, ed. P.C. Bartrum, Cardiff, 1966, p. 9. D.N. Dumville, “On the North British Section of the Historia Brittonum”, WHR 8 (1977), 345–54, at pp. 349–54; K.H. Jackson, “On the Northern British Section”, pp. 25–27; H.M. Chadwick & N.K. Chadwick, The Growth of Literature, 3 vols., Cambridge, 1932–40, vol. 1, p. 155. For this poem, see Armes Prydein Vawr, ed. and trans. I. Williams and R. Bromwich, Armes Prydein: The Prophecy of Britain from the Book of Taliesin, Dublin, 1972; D.N. Dumville, “Brittany and Armes Prydein Vawr”, Études celtiques 20 (1983), 145–59; A. Breeze, “Armes Prydein, Hywel Dda, and the Reign of Edmund of Wessex”, Études celtiques 33 (1997), 209– 22; H. Fulton, “Tenth-Century Wales and Armes Prydein”, Transactions of the Honourable Society of Cymmrodorion, new series, 7 (2001), 5–18; C. Etchingham, “Viking-Age Gwynedd and Ireland: Political Relations”, in K. Jankulak and J. Wooding (eds.), Ireland and Wales in the Middle Ages, Dublin, 2007, pp. 149–67; G. Isaac, “Armes Prydain Fawr and St David”, in J.W. Evans and J.M. Wooding (eds.), St David of Wales: Cult, Church and Nation, Woodbridge, 2007, pp. 161–81; N. Tolstoy, “When and Where was Armes Prydein Composed?” Studia Celtica 42 (2008), 145–49; T.M. Charles-Edwards, Wales and the Britons, 350–1064, Oxford, 2013, pp. 519–35. For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 49 sied to return to lead the Britons to victory over the English. The other deliverer is a certain Cynan. The identities of these two characters are nowhere made explicit in Armes Prydein Vawr, but Geoffrey, who probably knew the poem, or one very like it (as discussed below), offered a solution in the VM: Cadwaladr is Cambrorum dux, “leader of the Welsh”, and Cynan is from Armorica.64 In other words, he seems to identify the two deliverers of Welsh prophecy with his own Cadualadrus and Conanus Meriadocus. It is impossible to know if Geoffrey’s assumptions or stipulations matched the ideas of Welsh composers of prophecy, but the position of Cadwaladr son of Cadwallon, upon whom Geoffrey’s Cadualadrus is partially based, as the latest king of the Britons in the Historia Brittonum may well indicate that Geoffrey’s identification of the Cadwaladr of prophecy is correct. If Cadwaladr son of Cadwallon had acquired the role of prophetic deliverer in Wales no later than the 9th or 10th centuries, one wonders what the perceived historical significance of his reign to the Welsh during the same early period was. Whatever it was, it seems likely that Geoffrey was privy to it, and seized upon it as the basis for the final act of his history. Geoffrey certainly made good use of the Historia Brittonum; but which version of the text did he use? There are five primary Latin recensions of the Historia Brittonum, each of which had a different pattern of circulation during the Middle Ages. The five recensions are as follows: – The Harleian recension: probably best represents the original 9th-century text, and circulated in manuscripts particularly in south-eastern England in the late 11th and 12th centuries.65 – The Gildasian recension: the vulgate text from the 12th century to the end of the Middle Ages, similar to the Harleian recension but truncated and attributed to Gildas.66 – The Vatican recension: created in England in 943 or 944, during the reign of King Edmund; the text was abbreviated and reworded from an English point of view, and appears in manuscripts from the 11th century onwards.67 64 65 66 67 VM, ll. 967–68. See Guy, “Origins”; D.N. Dumville, “The Liber Floridus of Lambert of Saint-Omer and the Historia Brittonum”, BBCS 26 (1975), 103–22. No critical text of the Harleian recension has been published, but for the text of the fullest manuscript witness, see HB (Harl. 3859). See D.N. Dumville, “Celtic-Latin Texts in Northern England, c. 1150–c. 1250”, Celtica 12 (1977), 19–49, at p. 19. For descriptions of the manuscripts of the recension, see D.N. Dumville, “The Textual History of the Welsh-Latin Historia Brittonum”, 3 vols., unpublished PhD thesis, University of Edinburgh, 1975, vol. 2, ch. 6. The latter is now available online: <https:// www.era.lib.ed.ac.uk/handle/1842/8972> (accessed 22 June 2019). See D.N. Dumville, Historia Brittonum 3: The “Vatican” Recension, Cambridge, 1985. For use by the Author only | © 2020 Koninklijke Brill NV 50 Guy – The Chartres recension: a fragmentary text related to the Vatican recension, preserved only on flyleaves taken from a Breton manuscript of the first half of the 11th century.68 The Chartres manuscript, along with the flyleaves, was unfortunately destroyed in 1944. – The Nennian recension: redacted in its extant form in the second half of the 11th century, possibly in Abernethy in Scotland, and preserved only in extracts added to the margins of Cambridge, Corpus Christi College, 139 between 1164 and 1166; closely related to Lebor Bretnach, the Irish version of the Historia Brittonum.69 It was suggested by Theodor Mommsen that Geoffrey used a copy of the Gildasian recension, because at one point in the DGB Geoffrey states that the miracles of St Germanus were described by Gildas in his book.70 Gildas’s The Ruin of Britain does not mention St Germanus, but the Historia Brittonum does: it describes a number of miracles performed by Germanus during his sojourn in Britain. The implication might be that Geoffrey used a version of the Historia Brittonum ascribed to Gildas. However, as Alex Woolf has pointed out, Geoffrey’s account of Germanus actually derives from Bede rather than the Historia Brittonum.71 More significantly, Michael Reeve has adduced textual evidence which shows that Geoffrey cannot have relied solely on a manuscript of the Gildasian recension, because he accurately quotes the Historia Brittonum at a point when the extant witnesses to the Gildasian recension are faulty.72 Therefore, while it is possible that Geoffrey was aware of the attribution of the Historia Brittonum to Gildas in some manuscripts, we should not read too much into Geoffrey’s direct references to Gildas, especially since, as Neil Wright has cautioned, most such references are spurious and have no basis in any text attributed to Gildas.73 68 69 70 71 72 73 See D.N. Dumville, “An Irish Idiom Latinised”, Éigse 16 (1975/76), 183–86. For the text, see Faral, LLA, vol. 3, pp. 4–28; F. Lot, Nennius et l’Historia Brittonum, Paris, 1934, pp. 227–31. See T.O. Clancy, “Scotland, the ‘Nennian’ Recension of the Historia Brittonum, and the Lebor Bretnach”, in S. Taylor (ed.), Kings, Clerics and Chronicles in Scotland 500–1297: Essays in Honour of Marjorie Ogilvie Anderson on the Occasion of her Ninetieth Birthday, Dublin, 2000, 87–107; Dumville, “Nennius”. For the Lebor Bretnach, see the edition Lebor Bretnach: The Irish Version of the Historia Brittonum Ascribed to Nennius, ed. A.G. Van Hamel, Dublin, 1932, and the textual discussion in D.N. Dumville, “The Textual History of the Lebor Bretnach: a Preliminary Study”, Éigse 16 (1976), 255–73. DGB, vi.101.375–76; T. Mommsen, Chronica Minora saec. IV. V. VI. VII. Vol. 3 [Minor Chronicles of the 4th, 5th, 6th, 7th centuries, Vol. 3] (Monumenta Germaniae Historica, Auctores Antiquissimi, 13), Berlin, 1898, p. 133; Piggott, “Sources”, p. 272. Woolf, “Geoffrey of Monmouth”, p. 274. DGB, p. lviii (esp. n. 62). Wright, “Geoffrey of Monmouth and Gildas”, pp. 22–24. For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 51 One distinguishing feature of the fullest manuscript of the Harleian recension is the appearance of a set of annals and a collection of genealogies embedded within the text, between the chronological calculations in chapter 66 and the list of the cities of Britain in chapter 66a. The annals are known as either the “Harleian chronicle” or the “A-text of Annales Cambriae”, and the genealogies as the “Harleian genealogies”.74 In their extant forms, both the chronicle and the genealogies belong to the middle of the 10th century. It has been argued that the annals and genealogies were a feature of the archetype of the Harleian recension, but that for various reasons they were not included in the few other surviving manuscript witnesses to the recension.75 Both the annals and genealogies were used by Geoffrey, making it likely that he had access to a version of the Harleian recension of the Historia Brittonum. Geoffrey’s use of the annals is less obvious than his use of the genealogies. An event noted in the early section of the DGB, during the reign of Riuallo, may contain a textual echo: it is said that “While he was king, it rained blood [cecidit pluuia sanguinea] for three days and people died from a plague of flies.”76 This may be compared with the annal for 689 in the Harleian chronicle, which reads pluuia sanguinea facta est in Brittannia, “it rained blood in Britain.”77 Another verbal borrowing may be seen in Geoffrey’s reference to Margadud rex Demetarum, “Margadud king of the Demetae”, at the battle of Chester (which Geoffrey places in Leicester); this probably emulates the obituary of Morgetiud rex Demetorum, “Maredudd king of the Demeti”, in the annal for 796 in the Harleian chronicle.78 A chronicle like the Harleian chronicle was almost certainly the source for Geoffrey’s famous date for the battle of Camlan. The DGB contains only three precise dates: the date of Lucius’s death in 156, the date of Camlan in 74 75 76 77 78 Both are edited in E. Phillimore, “The Annales Cambriæ and the Old-Welsh Genealogies from Harleian MS. 3859”, Y Cymmrodor 9 (1888), 141–83 (repr. in J. Morris (ed.), Genealogies and Texts (Arthurian Period Sources, 5), Chichester, 1995, pp. 13–55). Guy, “Origins”, pp. 53–54. DGB, ii.33.287–89: “In tempore eius tribus diebus cecidit pluuia sanguinea et muscarum affluentia homines moriebantur.” In one particular respect, Geoffrey’s copy of this chronicle might have preserved a reading that was closer to the “Breviate chronicle” or “B-text of Annales Cambriae”, which derives from the same common source as the Harleian chronicle: the Breviate chronicle, like Geoffrey, uses the verb cecidit rather than facta est in this annal. However, overall it is likely that Geoffrey’s copy of the chronicle was closer to the Harleian version than the Breviate version, as argued below. For the three surviving Latin versions of this annal in parallel, see Annales Cambriae, AD 682–954: Texts A–C in Parallel, ed. and trans. D.N. Dumville, Cambridge, 2002, pp. 2–3. DGB, xi.189.213; cf. xi.200.480; Annales Cambriae, ed. and trans. Dumville, pp. 8–9. For use by the Author only | © 2020 Koninklijke Brill NV 52 Guy 542, and the date of Cadualadrus’s death in 689.79 Although Geoffrey’s 542 date has attracted a certain amount of rather credulous speculation, such as is inevitable in an “Arthurian” context, no consensus has developed regarding its origin.80 Fortunately, the two other dates are easier to explain. The date of Lucius’s death has been borrowed from Bede, who states that Lucius sent his letter to Pope Eleutherius during the joint empire of Marcus Antoninus Verus (i.e. Marcus Aurelius) and Aurelius Commodus (i.e. Lucius Verus), which Bede says began in 156 (actually 161).81 Bede was likewise the source for the date of Cadualadrus’s death. Geoffrey’s Cadualadrus, king of the Britons, is a merger of two historical kings of the second half of the 7th century: Cadwaladr, king of Gwynedd, and Cædwalla, king of the West Saxons. It was the latter who provided Geoffrey with the most convenient way to date the death of the final king in his epic narrative; Bede dated Cædwalla’s death to 20 April 689, and so Geoffrey duly transferred this date to his Cadualadrus.82 However, Geoffrey also had access to a source containing a date for the death of Cadwaladr of Gwynedd: the Harleian chronicle.83 Although modern scholars have deduced that the Harleian chronicle places the death of Cadwaladr of Gwynedd in the year 682,84 the chronicle itself does not contain any absolute dates; instead, it simply numbers its annals in groups of ten. This feature, coupled with the relative proximity of the two dates 689 and 682, would have made it easy for Geoffrey to equate the obituary of Cædwalla of the West Saxons in Bede (689) with the obituary of Cadwaladr of Gwynedd in the Harleian chronicle (usually deduced as 682). The Harleian chronicle was probably the only source accessible to Geoffrey that offered a date for another key moment in his history: the battle of Camlan. Again, although scholars have deduced that the Harleian chronicle places 79 80 81 82 83 84 DGB, v.73.8, xi.178.83–84, and xi.206.585–86. For example, see G. Ashe, “ ‘A certain very ancient book’: Traces of an Arthurian Source in Geoffrey of Monmouth’s History”, Speculum 56:2 (1981), 301–23, at p. 317. For an incisive critique of Ashe’s methodology, see R.W. Hanning, “Inventio Arthuri: a Comment on the Essays of Geoffrey Ashe and D.R. Howlett”, Arthuriana 5:3 (1995), 96–99, at pp. 96–98. Bede, Ecclesiastical History i.4, ed. and trans. B. Colgrave and R.A.B. Mynors, Bede’s Ecclesiastical History of the English People, Oxford, 1969, pp. 24–25. Bede, Ecclesiastical History v.7, ed. and trans. Colgrave and Mynors, pp. 470–71. Additionally, Geoffrey could have worked out a date for Cadwaladr of Gwynedd’s death using chapter 64 of the Historia Brittonum, which appears to claim that Cadwaladr died in the famous plague during Oswiu’s reign (i.e. in 664) (HB (Harl. 3859) §64, ed. Faral, vol. 3, p. 43). However, Geoffrey seems to have ignored this claim, which in any case is probably incorrect (cf. Charles-Edwards, Wales, pp. 355–56; K.H. Jackson, “On the Northern British Section”, p. 35). Phillimore, “Annales Cambriæ”, p. 159; Annales Cambriae, ed. and trans. Dumville, p. 2. For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 53 Camlan in 537, the original text does not offer an absolute date.85 Geoffrey’s only option was to count back the years from an event with a known date to an event with an unknown date. Counting back from the obituary of Cadwaladr of Gwynedd in the Harleian chronicle would have revealed to Geoffrey a gap of 147 marked years between that event and the battle of Camlan. All Geoffrey needed to do was subtract 147 from his absolute date for the death of Bede’s Cædwalla, in 689, and he had deduced a date for Camlan: 542. The result is all the more striking because it implies that Geoffrey used a text of the Welsh annals that contained the same errors as the Harleian chronicle. All copies of the Welsh annals inevitably contain copying errors, especially because it was so easy to omit or insert year markings in sections of the annals in which no actual events were recorded. This is why there is a discrepancy between the 147 marked years separating Camlan from Cadwaladr’s death in the extant text of the Harleian chronicle and the 145 years separating the two dates which scholars have attributed to the chronicle’s events, 537 and 682. Only one other copy of the Welsh annals survives in which the number of years between Camlan and Cadwaladr’s death can be counted: the late-13th-century “Breviate chronicle”, or “B-text of the Annales Cambriae”, which derives from the same common source as the Harleian chronicle. By comparing the Harleian chronicle and the Breviate chronicle with one another and with external sources, it is possible to infer that, between their records for Camlan and Cadwaladr’s death, the Harleian chronicle, by comparison with the Breviate, is missing four annals and has three additional annals, whereas the Breviate chronicle, by comparison with the Harleian, is missing four annals and has no additional annals.86 The discrepancy means that the Breviate chronicle contains only 144 marked years between Camlan and Cadwaladr’s death, and could not have been used by Geoffrey to deduce the date 542 for Camlan. This strongly suggests that Geoffrey used a version of the Welsh Latin annals that was closer to the Harleian chronicle embedded in the Historia Brittonum, confirming in turn that he probably had access to a version of the Harleian recension of the Historia Brittonum. Geoffrey’s use of a text like the Harleian genealogies has been better documented, since the relationship between Geoffrey’s work and the genealogies has been studied by Edmond Faral, Arthur E. Hutson, and Stuart Piggott.87 One 85 86 87 Phillimore, “Annales Cambriæ”, p. 154. These calculations rely on the excellent work of H. Gough-Cooper in Annales Cambriae: A, B and C in Parallel, from St Patrick to AD 954, 2016, <http://croniclau.bangor.ac.uk/ documents/AC_ABC_to_954_first_edition.pdf> (accessed 30 April 2017), pp. 7–16. Faral, LLA, vol. 2, pp. 117–18, 137–39, and 276; A.E. Hutson, British Personal Names in the Historia regum Britanniae, Berkeley, 1940; id., “Geoffrey of Monmouth”, Transactions of the Honourable Society of Cymmrodorion (1937), 361–73, at pp. 368–73; Piggott, “Sources”. For use by the Author only | © 2020 Koninklijke Brill NV 54 Guy of the clearest examples of Geoffrey’s deployment of these genealogies comes in his list of the attendees at Arthur’s Whitsun court at Caerleon, among whom are the following ragtag bunch: Donaut Mappapo, Cheneus Mapcoil, Peredur Maberidur, Grifud Mapnogoid, Regin Mapclaut, Eddelein Mapcledauc, Kincar Mabbangan, Kinmarc, Gorbonian Masgoit, Clofaut, Run Mapneton, Kinbelin Maptrunat, Cathleus Mapcatel, Kinlith Mapneton88 Most of these names have been lifted wholesale from a few adjacent sections of a text very like the Harleian genealogies. Compare the names in bold, italicized, or set in smallcaps with the following extracts from the genealogies:89 [U]rbgen map Cinmarc map Merchianum map Gurgust map Coil Hen. [G]uallauc map Laenaec map Masguic Clop map Ceneu map Coyl Hen. [M]orcant map Coledauc map Morcant Bulc map Cincar braut map Bran Hen map Dumngual Moilmut map Garbaniaun … [D]unaut map Pappo map Ceneu map Coyl Hen. [G]urci ha Peretur mepion Eleuther Cascord maur … [R]un map Neithon map Caten map Caurtam map Serguan map Letan map Catleu map Catel map Decion map Cinis Scaplaut map Louhen map Guidgen map Caratauc map Cinbelin map Teuhant … This is the only section in Geoffrey’s history where he retains the Old Welsh map (“son (of)”) formula found in the genealogies; elsewhere he picks out the names and epithets but does not explicitly use the patronymics. This is not to say that he was unaware of them. In the first extract from the genealogies just quoted may be found the name Dumngual Moilmut; this was Geoffrey’s source for the name of his great lawgiver, Dunuallo Molmutius, whose relationship with his son, Brennius, the conqueror of Rome, was determined by the relationship between Dumngual Moilmut and his son Bran Hen, “Bran the Old”, in the genealogies.90 A high proportion of the Brittonic name-forms in the DGB can be found distributed across almost every section of the Harleian genealogies, making it very 88 89 90 DGB, ix.156.340–43. Phillimore, “Annales Cambriæ”, pp. 173–76; cf. Tracts, ed. Bartrum, pp. 10–11. Piggott, “Sources”, p. 279. Geoffrey may have had another Welsh source for his Dunuallo Molmutius: see Roberts, “Sylwadau”, pp. 136–37; M.E. Owen, “Royal Propaganda: Stories from the Law-Texts”, in T.M. Charles-Edwards, M.E. Owen, and P. Russell (eds.), The Welsh King and his Court, Cardiff, 2000, pp. 224–54, at pp. 229–30. For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 55 likely that Geoffrey used a version of the text similar to that which survives embedded in the Historia Brittonum in the Harley manuscript. He seems to have favored some sections of the genealogies over others. He made good use of the sections concerning the legendary heroes of the Brittonic north, the subjects of the first two extracts quoted above. He also made frequent use of the pedigrees of the kings of Gwynedd and Dyfed (his two principal “Cambrian” kingdoms), which are the first two pedigrees in the Harleian genealogies. As many as nine of the names of Ebraucus’s sons and daughters may have been taken from these two pedigrees: Iagon, Chein, and Aballac from the Gwynedd pedigree (compare Iacob, Cein, and Aballac) and Margadud, Regin, Kincar, Gloigin, Tangustel, and perhaps Ragan from the Dyfed pedigree (compare Margetiut, Regin, Cincar, Gloitguin, and Tancoystl). One might question the extent to which Geoffrey understood the genealogies that he quarried for name forms. He knew that the genealogies were lists of names, but did he know the proper historical contexts to which those names pertained? Despite the Harleian genealogies containing no dates and few place-names, Geoffrey does indicate that he could contextualize some of them. A particularly striking example concerns Geoffrey’s King Tenuantius, successor of Cassibellaunus and father of Kimbelinus. Tenuantius is Geoffrey’s version of the Tasciovanos of history, the father of Cunobelinos and grandfather of Caratacos. But while the latter two are known to us through Roman writers, such as Suetonius, Dio Cassius, and Tacitus, Tasciovanos is known solely through his coins. The only written source that mentions the father of Cunobelinos prior to the DGB is the Harleian genealogies, in the pedigree forming the third extract quoted above, which incorporates the three generations Caratauc map Cinbelin map Teuhant, “Caratacos son of Cunobelinos son of Teuhant”.91 According to John Koch, Teuhant would be the regular Old Welsh derivative of Tasciovanos, suggesting that at this point the Harleian genealogies have incorporated accurate information about the family that had been preserved in oral tradition.92 Since there is no reason that Geoffrey would have known the name of Cunobelinos’s father from independent sources, he must have realized that the pedigree’s Caratauc and Cinbelin corresponded to the pre-Roman kings Caratacos and Cunobelinos mentioned in his other 91 92 Piggott, “Sources”, p. 280; J.T. Koch, “A Welsh Window on the Iron Age: Manawydan, Mandubracios”, CMCS 14 (1987), 17–52, at p. 17. J.T. Koch, “Llawr en asseð (CA 932) ‘The laureate hero in the war-chariot’: Some Recollections of the Iron Age in the Gododdin”, Études celtiques 24 (1987), 253–78, at pp. 266–70. For use by the Author only | © 2020 Koninklijke Brill NV 56 Guy sources, and then correctly deduced from this that Teuhant was Cunobelinos’s predecessor. Arthur’s Whitsun court at Caerleon provides other examples of Geoffrey’s comprehension of the genealogies. Among the attendees may be found Caduallo Lauihr rex Venedotorum, “Caduallo Lauihr, king of the Venedoti”, and Stater rex Demetarum, “Stater, king of the Demetae”.93 The two names have been taken respectively from the Gwynedd pedigree (Catgolaun Iauhir) and the Dyfed pedigree (Stater) in the Harleian genealogies, showing that Geoffrey understood to which kingdoms those pedigrees pertained. In the case of Caduallo Lauihr, he is even roughly correct about the implied date; the historical Cadwallon Lawhir of Gwynedd was the father of Maelgwn Gwynedd, who, as we know from Gildas, flourished in the 6th century.94 Geoffrey demonstrates his thorough understanding of the Gwynedd pedigree later in his history in the conversation between Caduallo and Salomon of Armorica, in which Caduallo, who is himself based on the historical Cadwallon son of Cadfan of Gwynedd (d. 634), explains his descent from Malgo, Geoffrey’s version of Maelgwn Gwynedd.95 Throughout the post-Arthurian section of his history, Geoffrey’s successful coordination between the Gwynedd pedigree and other information derived from Gildas, Bede, and elsewhere creates an important element of continuity in the narrative. It does not matter that the pedigree offered by Geoffrey’s Caduallo contains a discrepancy when compared with the Harleian genealogies, in listing Ennianus, rather than Run, as Caduallo’s ancestor; it would not have satisfied Geoffrey to reproduce his source exactly. One further example of borrowing from the genealogies might suggest the origin of Geoffrey’s copy of the text. At the beginning of his reign, Dunuallo Molmutius, a typically strenuous scion of the house of Cornwall, defeats three kings in order to become king of Britain: Pinner, king of Loegria, Rudaucus, king of Wales, and Staterius, king of Scotland.96 The names Pinner and Staterius can only be based on the Pincr and Stater of the Dyfed pedigree in the Harleian genealogies; they are not, in fact, real names, but rather Latinate titles (pincerna, “cup-bearer”, and stator, “magistrate’s marshal”) artificially introduced into the pedigree in order to extend it further back in time.97 The name Rudaucus, on the other hand, has been taken from a version of the Gwynedd pedigree. 93 94 95 96 97 DGB, ix.156.329–30. Gildas, The Ruin of Britain §§33–36, ed. and trans. Winterbottom, pp. 32–36 and 102–05. DGB, xi.195.376–83. DGB, ii.34. E.W.B. Nicholson, “The Dynasty of Cunedag and the ‘Harleian Genealogies’ ”, Y Cymmrodor 21 (1908), 63–104, at p. 81; B. Guy, “The Earliest Welsh Genealogies: Textual Layering and the Phenomenon of ‘Pedigree Growth’ ”, Early Medieval Europe 26 (2018), 462–85, at p. 484. For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 57 In the Harleian genealogies, one of the ancestors of the kings of Gwynedd is called Patern Pesrut. However, versions of the same pedigree also appear in the Welsh Latin Lives of saints Cadog and Carannog. The Life of St Cadog was written by Lifris, archdeacon of Glamorgan, at the end of the 11th century, but the genealogy might have been added during the 12th century.98 Similarly, the genealogy in the probably 12th-century Life of St Carannog may have been inserted at a slightly later point, since it now separates two parts of what may once have been a unitary composition.99 In both these versions of the genealogy, the same ancestor is called Patern Peis Rudauc rather than Patern Pesrut. Rudauc, which in modern Welsh would be spelt rhuddog, is an adjective meaning “red, reddish-brown”, but it is not attested independently in any written text until 1707 (unlike the much commoner adjective rhudd, on which rhuddog is based).100 This renders it very likely that Geoffrey took the name Rudaucus from a version of the Gwynedd pedigree, a version which, moreover, was slightly closer to the version in the Lives of Cadog and Carannog than to the one in the extant Harleian genealogies. This is significant because the Lives of Cadog and Carannog themselves seem to have taken the genealogy from a text very similar to the Harleian genealogies that was circulating in places connected to Llancarfan, where St Cadog was the patron saint, during Geoffrey’s lifetime.101 There is further evidence for this. For example, Glastonbury Abbey, which at some point in the 12th century commissioned a Life of St Gildas from none other than Caradog of Llancarfan,102 was the place where additional material was added to William of Malmesbury’s The Early History of Glastonbury from the Harleian recension of Historia Brittonum and from genealogies like 98 99 100 101 102 For the date of the Life, see C.N.L. Brooke, The Church and the Welsh Border in the Central Middle Ages, ed. D.N. Dumville (Studies in Celtic History, 8), Woodbridge, 1986, pp. 72–73 and 89. For the suggestion that the genealogy is a later insertion, see H.D. Emanuel, “An Analysis of the Composition of the ‘Vita Cadoci’ ”, National Library of Wales Journal 7 (1952), 217–27, at p. 220. For the Life (or Lives) of St Carannog, see K. Jankulak, “Carantoc alias Cairnech? British Saints, Irish Saints, and the Irish in Wales”, in K. Jankulak and J.M. Wooding (eds.), Ireland and Wales in the Middle Ages, Dublin, 2007, pp. 116–48. GPC Online, University of Wales Centre for Advanced Welsh and Celtic Studies, Aberystwyth, 2014, <http://www.geiriadur.ac.uk/> (accessed 30 April 2017), s.v. rhuddog. It does not seem that the addition of the suffix -og to rhudd altered the word’s meaning. Cf. P. Russell, Celtic Word Formation: The Velar Suffixes, Dublin, 1990, p. 38. The evidence is set out more fully in B. Guy, Medieval Welsh Genealogy: An Introduction and Textual Study, Woodbridge, 2020, pp. 79–100. See above, n. 25. For use by the Author only | © 2020 Koninklijke Brill NV 58 Guy the Harleian genealogies, possibly at the end of the 12th century.103 It is quite possible that Caradog of Llancarfan himself, whom Geoffrey describes as “my contemporary”, provided Geoffrey with his copy of the Harleian recension of the Historia Brittonum, containing versions of the same interpolated annals and genealogies as are found in the extant Harley manuscript.104 3 The True Faith Religion will be destroyed again and archbishoprics will be displaced. London’s honour will adorn Canterbury and the seventh pastor of York will dwell in the kingdom of Armorica. St Davids will wear the pallium of Caerleon, and the preacher of Ireland will fall silent because of a baby growing in the womb.105 The quotation above is spoken as part of Merlin’s prophecies to Vortigern. The passage appears near the beginning of the prophecies and concerns events due to happen not long after the reign of Arthur. Its subject matter is readily identifiable, within the terms of Galfridian history. According to Geoffrey, when the Britons were converted to Christianity during the reign of King Lucius, three metropolitan dioceses were established, based in York, London, and Caerleon.106 This prophecy foretells certain events that will befall each one. London’s honor will pass to Canterbury during the time of St Augustine, even though Geoffrey does not explicitly mention Augustine’s foundation of the church of Canterbury; St Samson, whom Geoffrey has flee from York during Arthur’s campaigns against the Saxons, becomes archbishop of Dol by the time of Arthur’s Whitsun court at Caerleon;107 and St David’s wearing of the pallium of Caerleon is a reference both to Geoffrey’s St David, “archbishop of Caerleon”, dying in St Davids during the reign of Constantinus, and to the real 12th-century campaign of Bernard, bishop of St Davids, for the elevation of St Davids to the 103 104 105 106 107 Scott, Early History, pp. 187–88, nn. 22 and 24; D.E. Thornton, “Glastonbury and the Glastening”, in L. Abrams and J.P. Carley (eds.), The Archaeology and History of Glastonbury Abbey: Essays in Honour of the Ninetieth Birthday of C.A. Ralegh Radford, Woodbridge, 1991, pp. 191–203, at pp. 195–96 and 200–01. DGB, xi.208.602: “contemporaneo meo”. DGB, Prophetiae 112.46–50: “Delebitur iterum religio, et transmutacio primarum sedium fiet. Dignitas Lundoniae adornabit Doroberniam, et pastor Eboracensis septimus in Armorica regno frequentabitur. Meneuia pallio Vrbis Legionum induetur, et praedicator Hiberniae propter infantem in utero crescentem obmutescet.” DGB, iiii.72.418–26. DGB, ix.151.194–96 and ix.158.406–09. For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 59 status of an archbishopric.108 But it is the last part of the passage that concerns us most here. This appears to be a reference to two events in Rhygyfarch’s Life of St David, written late in the 11th century: St Patrick’s visit to Dyfed prior to David’s birth, and Gildas’s being struck dumb by the unborn David, still in his mother’s womb.109 In Geoffrey’s typical fashion, he has combined aspects of these two events together so as not to replicate either one too closely. Another reference to Rhygyfarch’s portrayal of Patrick’s visit to Dyfed occurs later in the history, where Geoffrey explains that St Patrick had founded St Davids and had foretold David’s birth.110 Again, Geoffrey has altered Rhygyfarch’s account; in the latter, David’s birth is foretold to Patrick by an angel, not by Patrick himself. Still, it is probably fair to deduce that Geoffrey was familiar with Rhygyfarch’s Life of St David. It is very likely that Geoffrey knew some of the hagiographical literature generated by the ecclesiastical controversies of South Wales in the first half of the 12th century.111 The controversies centered on Bishop Bernard of St Davids’ (unsuccessful) campaign to establish St Davids as the seat of an independent archbishopric, and Bishop Urban of Llandaff’s (successful) campaign to assert the independence of Llandaff as the center of a bishopric subordinate to Canterbury. Each of these campaigns produced saints’ lives and accounts of ecclesiastical history to be used as propaganda, culminating most famously in the Book of Llandaff.112 Some of Geoffrey’s passing references to events of ecclesiastical history bear witness to his familiarity with the claims that these dioceses were propagating through their texts. For instance, his reference in the VM to St Davids, where “the pall lost for many years will be recovered”, shows his cognizance of the claim of the church of St Davids to have been the seat of an archbishop earlier in its history.113 The claim is found in Rhygyfarch’s Life of 108 109 110 111 112 113 DGB, xi.179.89–91. For Bernard’s campaign, see M. Richter, Giraldus Cambrensis: The Growth of the Welsh Nation, Aberystwyth, 1972, at pp. 40–61; Episcopal Acts and Cognate Documents relating to Welsh Dioceses 1066–1272, ed. J.C. Davies, 2 vols., Cardiff, 1946–48, vol. 1, pp. 190–208. Rhygyfarch ap Sulien, Life of St David §3 and §5, ed. and trans. R. Sharpe and J.R. Davies, “Rhygyfarch’s Life of St David”, in J.W. Evans and J.M. Wooding (eds.), St David of Wales: Cult, Church and Nation, Woodbridge, 2007, pp. 107–55, at pp. 110–15; cf. Wright, “Geoffrey of Monmouth and Gildas Revisited”, pp. 156–57. DGB, xi.179.92–93; cf. Tatlock, LHB, p. 246. For more detailed discussion, see Barry Lewis’s chapter in the present volume. For the relationship between the Book of Llandaff and 12th-century ecclesiastical politics, see J.R. Davies, The Book of Llandaf and the Norman Church in Wales, Woodbridge, 2003. For a diplomatic edition of the whole manuscript, see The Text of the Book of Llan Dâv: Reproduced from the Gwysaney Manuscript, ed. J.G. Evans and J. Rhŷs, Oxford, 1893. VM, l. 623: “palla sibi reddetur dempta per annos.” For use by the Author only | © 2020 Koninklijke Brill NV 60 Guy St David, and was developed and elaborated as the 12th century progressed.114 Geoffrey’s reference to St Teilo, “a distinguished priest of Llandaff”, replacing St Samson as archbishop of Dol probably betrays his familiarity with the version of the Life of St Teilo preserved in the Book of Llandaff. Only this version of the Life, unlike the other, probably earlier, version preserved in London, British Library, Cotton Vespasian A. xiv, mentions Teilo as bishop of Llandaff and then later as bishop of Dol following St Samson.115 De situ Brecheniauc (“Concerning the Establishment of Brycheiniog”) is another Latin ecclesiastical text probably produced in South Wales in the first half of the 12th century that may have been used by Geoffrey. This text narrates the conception and birth of Brychan, the eponymous founder of Brycheiniog in south-central Wales, and then lists Brychan’s many sons and daughters, most of whom can be identified as saints associated with churches in Brycheiniog and other regions of South Wales. Arthur Hutson suggested that Brychan was the inspiration for Geoffrey’s Ebraucus, whose 20 sons and 30 daughters are enumerated in the DGB.116 As Hutson pointed out, some of the more unusual names among Ebraucus’s daughters are paralleled only among the names of Brychan’s daughters. These include Gorgon (compare Gurygon/ Grucon), Kambreda (compare Kein/Kein breit), and Claudus (compare Gladus/ Gluadus). In each of these three cases, the former of the two bracketed italicized forms has been taken from the version of De situ Brecheniauc in Cotton Vespasian A. xiv, while the latter has been taken from the related text known as Cognacio Brychan, found in London, British Library, Cotton Domitian A. i.117 The closer correspondence between the DGB and the forms found in Cognacio Brychan may suggest that Geoffrey drew on a version of the Brychan tract resembling the latter. 114 115 116 117 Rhygyfarch, Life of St David §§49–53, ed. and trans. Sharpe and Davies, pp. 142–47. DGB, ix.158.406–09: “Teliaus illustris presbiter Landauiae”. For the text of the Book of Llandaff’s version of the Life of St Teilo, see Life of St Teilo, ed. J.G. Evans and J. Rhŷs, The Text of the Book of Llan Dâv: Reproduced from the Gwysaney Manuscript, Oxford, 1893, pp. 97–117; for a summary of the differences between the two versions of the Life, see P.C. Bartrum, A Welsh Classical Dictionary: People in History and Legend up to about A.D. 1000, Aberystwyth, 1993, pp. 605–06. It has been argued that Teilo’s visit to Dol in the Book of Llandaff is modeled on the Breton Life of St Turiau; see G.H. Doble, Lives of the Welsh Saints, ed. D.S. Evans, Cardiff, 1971, pp. 183–86; J.R. Davies, Book of Llandaf, p. 117. Hutson, British Personal Names, pp. 16–22; id., “Geoffrey”, pp. 361–68. For Ebraucus’s daughters, see DGB, ii.27.99–104. Both versions are edited and translated in A.W. Wade-Evans, “The Brychan Documents”, Y Cymmrodor 19 (1906), 18–48. Both versions were edited again, without translations, in Vitae Sanctorum Britanniae et Genealogiae: The Lives and Genealogies of the Welsh Saints, ed. A.W. Wade-Evans, Cardiff, 1944, pp. 313–18. For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 4 61 Dark Sayings from a Dark Heart It is the will of the most-high Judge that the British shall be without their kingdom for many years because of their weakness, until Conanus shall arrive in his ship from Armorica, and that revered leader of the Welsh, Cadwaladrus. They will join together with the Scots, the Welsh, the Cornish, and the Armoricans in a firm league. Then they will restore to their own people the crown that had been lost. The enemy will be driven out and the time of Brutus will be back once more.118 This section of Merlin’s prophecy to Telgesinus in the VM is the closest that Geoffrey comes to paraphrasing a 10th-century Welsh prophetic poem that he almost certainly knew, known as Armes Prydein Vawr (“The Great Prophecy of Britain”).119 Armes Prydein Vawr foretells of an alliance of Welsh, Irish, Cornish, Bretons, and others who will rise up to defeat the English with the help of the returning leaders Cadwaladr and Cynan, just as in the VM.120 The poem was probably composed in the first half of the 10th century, while either Æthelstan (924–39) or his half-brother Edmund (939–46) were supreme in Britain, and it may have been inspired by the alliance between the Hiberno-Scandinavians of Dublin, the Scots of Alba, and the Britons of Strathclyde at the battle of Brunanburh in 937. The poet specifically recounts how the victory of the Welsh had been prophesied by no less a figure than Myrddin, the Welsh precursor of Geoffrey’s Merlin, whose appearance in this context may have been one of the inspirations for Geoffrey’s portrayal of Merlin as the chief political prophet of his legendary world.121 Prophecy, as a method of political commentary on past events and an expression of desires and anxieties about the future, was a popular literary genre 118 119 120 121 VM, ll. 964–72: “sententia summi / judicis existit, Britones ut nobile regnum / temporibus multis amittant debilitate, / donec ab Armorica veniet temone Conanus / et Cadualadrus Cambrorum dux venerandus, / qui pariter Scotos Cambros et Cornubienses / Armoricosque viros sociabunt federe firmo / amissumque suis reddent diadema colonis, / hostibus expulsis renovato tempore Bruti.” I have altered Clarke’s translation following advice from an anonymous reviewer. Cf. D. Edel, “Geoffrey’s So-Called Animal Symbolism and Insular Celtic Tradition”, Studia Celtica 18/19 (1983/84), 96–109, at p. 97; A.O.H. Jarman, “The Merlin Legend and the Welsh Tradition of Prophecy”, in Bromwich et al. (eds.), The Arthur of the Welsh, pp. 117–45, at p. 137. See above, pp. 48–49. Armes Prydein Vawr l. 17, ed. and trans. Williams and Bromwich, pp. 2–3. For use by the Author only | © 2020 Koninklijke Brill NV 62 Guy during the Middle Ages.122 It was a literary form that was thoroughly exploited by Geoffrey, whose PM achieved fame and popularity as a work in its own right, in addition to forming the central linchpin of the DGB.123 But to what extent did Welsh examples of the genre influence Geoffrey’s particular brand of Merlinian prophecy? It is relatively uncontroversial to claim that Geoffrey may have known Armes Prydein Vawr, since the dating of that poem to the first half of the 10th century is fairly secure. But in this respect Armes Prydein Vawr stands almost alone, because the dating of the majority of early Welsh prophetic poems is contested and uncertain.124 Included in the latter category are the early Myrddin poems, the dating of which is inextricably bound up with the intractable question of their relationship with the VM.125 It has been persistently claimed that Geoffrey discovered the Welsh legend of Myrddin between the completion of the DGB around 1138 and the writing of the VM around 1150.126 This is because the account of Merlin’s life in the VM mirrors various aspects of Myrddin’s story in Welsh poetry, whereas 122 123 124 125 126 For an excellent summary focused on the 12th century, see R.W. Southern, “Aspects of the European Tradition of Historical Writing, 3: History as Prophecy”, Transactions of the Royal Historical Society, fifth series, 22 (1972), 159–80 (repr. in R.J. Bartlett (ed.), History and Historians: Selected Papers of R.W. Southern, Oxford, 2004, 48–65). For the later Middle Ages, see the collection of essays in M. Reeves, The Prophetic Sense of History in Medieval and Renaissance Europe, Aldershot, 1999. See J. Crick, “Geoffrey of Monmouth, Prophecy and History”, Journal of Medieval History 18:4 (1992), 357–71; ead., “Geoffrey and the Prophetic Tradition”, in S. Echard (ed.), The Arthur of Medieval Latin Literature: The Development and Dissemination of the Arthurian Legend in Medieval Latin (Arthurian Literature of the Middle Ages, 6), Cardiff, 2011, pp. 67–82; C. Daniel, Les prophéties de Merlin et la culture politique (XIIe–XVIe siècles), Turnhout, 2006; and Maud McInerney’s contribution to the present volume. Compare the lack of secure dates for the poems edited in M. Haycock, Prophecies from the Book of Taliesin, Aberystwyth, 2013. The dominant view of their relationship during much of the latter half of the 20th century was that of A.O.H. Jarman: see his “The Welsh Myrddin Poems”, in R.S. Loomis (ed.), Arthurian Literature in the Middle Ages, Oxford, 1959, pp. 20–30; id., The Legend of Merlin, Cardiff, 1960; id., “Early Stages in the Development of the Myrddin Legend”, in R. Bromwich and R.B. Jones (eds.), Astudiaethau ar yr Hengerdd / Studies in Old Welsh Poetry: Cyflwynedig i Syr Idris Foster [Studies in Old Welsh poetry presented to Sir Idris Foster], Cardiff, 1978, pp. 326–49; “Merlin legend”. Aspects of this view have recently been challenged: O.J. Padel, “Geoffrey of Monmouth and the Development of the Merlin Legend”, CMCS 51 (2006), 37–65; N. Tolstoy, “Geoffrey of Monmouth and the Merlin Legend”, AL 25 (2008), 1–42. J.J. Parry, The Vita Merlini (University of Illinois Studies in Language and Literature, 10.3), Urbana, IL, 1925, pp. 13 and 16; M.E. Griffiths, Early Vaticination in Welsh with English Parallels, Cardiff, 1937, p. 78; Jarman, Legend of Merlin, pp. 24–25; id., “Early Stages”, p. 349; id., “Merlin Legend”, p. 135; Roberts, “Sylwadau”, p. 139; VM, p. 29; Tolstoy, “Geoffrey of Monmouth”, pp. 11 and 13. For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 63 the account of Merlin in the DGB does not. As A.O.H. Jarman put it, “at some time subsequent to 1138, however, Geoffrey must have learnt more about the Myrddin legend and realised that the account given of him in the Historia was contrary to popular tradition.”127 But we have already noted how perilous it is to assume Geoffrey’s ignorance or miscomprehension on the basis on his failure to reproduce a source at his disposal exactly. Alignment with popular tradition was not one of Geoffrey’s primary concerns. Geoffrey’s creation of a new “Merlin” character through the merger of the fatherless boy of the Historia Brittonum and the Welsh prophet Myrddin was deliberate and considered, and provides no evidence at all for the extent of Geoffrey’s acquaintance with Welsh Myrddin poetry by 1138. This can be judged only through positive evidence, rather than evidence of absence. It is likely that the VM reflects Geoffrey’s familiarity with versions of some surviving Welsh poems.128 The parallels between the VM and the Welsh poems are all the more striking in view of the apparent obscurity of the VM during the Middle Ages, making it less likely that the Welsh poems have been influenced by the VM.129 One such poem is Yr Afallennau (“The Apple Trees”), the earliest extant copy of which is found in the mid-13th-century Black Book of Carmarthen. In this poem, the narrator prophesies political events, including great victories for the Welsh over the English, from underneath an apple tree. Although the narrator remains nameless, references to incidents from his past, including the battle of Arfderydd, his madness, and his sleeping in the forest of Celyddon, align him with Geoffrey’s Merlin in the VM. Geoffrey may allude to this poem or a poem with a similar theme in his repeated references to Merlin’s encounters with apples and apple trees.130 Another poem that seems to be reflected in the VM is Ymddiddan Myrddin a Thaliesin (“The Conversation of Myrddin and Taliesin”), also preserved in the Black Book of Carmarthen, which may have provided a model for the long conversation between Merlin and Telgesinus (Geoffrey’s Taliesin) in the VM.131 One of the topics discussed in 127 128 129 130 131 Jarman, “Merlin Legend”, p. 135. English translations of the Welsh Myrddin poems discussed below may be found in J.K. Bollard, “Myrddin in Early Welsh Tradition”, in P. Goodrich (ed.), The Romance of Merlin: An Anthology, New York, 1990, pp. 13–54. Tolstoy, “Geoffrey of Monmouth”, pp. 25–27 and 34–36. VM, ll. 90–95, 567, and 1408–16; cf. Jarman, Legend of Merlin, p. 25; id., “Merlin Legend”, p. 134; Padel, “Geoffrey of Monmouth and the Development of the Merlin Legend”, pp. 57–58; Tolstoy, “Geoffrey of Monmouth”, p. 38. Ymddiddan Myrddin a Thaliesin (o Lyfr Du Caerfyrddin) [The Conversation of Myrddin and Taliesin (from the Black Book of Carmarthen)], ed. A.O.H. Jarman, Cardiff, 1951, at p. 44; id., Legend of Merlin, p. 25; id., “Early Stages”, p. 332; Padel, “Geoffrey of Monmouth and the Development of the Merlin Legend”, pp. 45–46. For use by the Author only | © 2020 Koninklijke Brill NV 64 Guy the Ymddiddan is gueith Arywderit, “the battle of Arfderydd”, which in the VM turns Merlin mad and drives him into the forest of Calidon. The VM’s story may be compared with the final stanza of the Ymddiddan, where Myrddin states that, in the battle, “seven score generous men went mad, they perished in the forest of Celyddon.”132 The VM’s conversation between Merlin and Telgesinus may also have been inspired by Welsh poems linked with the legendary Welsh poet Taliesin. Telgesinus’s role in the VM is primarily that of a cosmological commentator, who divulges information to Merlin about the world’s waters, islands, and, curiously, fish. A similar range of cosmological expertise, including knowledge of fish, is attributed to the legendary persona of Taliesin in some of the poems preserved in the 14th-century Book of Taliesin.133 A final poem that Geoffrey may have drawn upon is Cyfoesi Myrddin a Gwenddydd ei Chwaer (“The Prophecy of Myrddin and Gwenddydd his Sister”), which is preserved in manuscripts from the end of the 13th century onwards. This is a long poem in which Gwenddydd questions her brother Myrddin in alternating stanzas about the future rulers of the Welsh. The poem is cast as prophecy, but begins by listing quasi-historical rulers of the Welsh, following first the Historia Brittonum’s account of the northern kings who opposed the English and latterly the Gwynedd pedigree up to the reign of Hywel Dda (d. 950). Thereafter the prophetic references become much vaguer, crystallizing only later in the poem in allusions to the 12th-century rulers Gruffudd ap Cynan, Owain Gwynedd, and King Henry.134 It has been suggested, quite plausibly, that the arrangement of the extant text is due to its being composite: namely, that an earlier prophetic poem, composed perhaps in the 10th century during the reign of Hywel Dda, was later augmented with stanzas referring to the 12th century.135 Many aspects of the poem, including the prophecy, the references to Arfderydd, Rhydderch, and Gwenddolau, and the role of Myrddin’s sister Gwenddydd (called Ganieda by Geoffrey), who in the VM 132 133 134 135 Ymddiddan Myrddin a Thaliesin ll. 35–36, ed. Jarman, p. 58: “Seith ugein haelon a aethan ygwllon, / Yg coed keliton y daruuan.” Translation is my own. Cf. M. Haycock, Legendary Poems from the Book of Taliesin, Aberystwyth, 2007, pp. 13, 156–57, 443, 515, 521, and 523. M.B. Jenkins, “Aspects of the Welsh Prophetic Verse Tradition: Incorporating Textual Studies of the Poetry from ‘Llyfr Coch Hergest’ (Oxford, Jesus College, MS cxi) and ‘Y Cwta Cyfarwydd’ (Aberystwyth, National Library of Wales, MS Peniarth 50)”, unpublished PhD thesis, University of Cambridge, 1990, pp. 80–83. It is not clear which son of which Henry is implicated in the phrases keneu Henri, “Henry’s cub” (l. 209) and mab Henri, “Henry’s son” (l. 213) (ibid., pp. 53 and 64). Jenkins, “Aspects of the Welsh Prophetic Verse Tradition”, pp. 40–41; J. Rowland, Early Welsh Saga Poetry: A Study and Edition of the Englynion, Cambridge, 1990, pp. 291–93; Tolstoy, “Geoffrey of Monmouth”, pp. 20–25; Charles-Edwards, Wales, pp. 337–39. For use by the Author only | © 2020 Koninklijke Brill NV Geoffrey of Monmouth ’ s Welsh Sources 65 finally joins Merlin and prophesies with him, imply that Geoffrey was familiar with the Cyfoesi or with something like it at the time that he composed the VM.136 Might he have known a version of the poem at an earlier stage, when he was composing the DGB? There may be a hint that he did in his treatment of Caduan, Caduallo’s father and predecessor. It has already been noted that Geoffrey was familiar with the pedigree of the kings of Gwynedd. It is possible that this pedigree was Geoffrey’s only source for Caduan, father of Caduallo, who is based on the historical 7th-century Cadfan of Gwynedd, father of Cadwallon; in this case, Geoffrey’s attribution of the kingship of the Venedoti and then of all the Britons to Caduan was solely a deduction from the pedigree, in light of the more famous position of the historical Cadwallon. However, if Geoffrey already knew the Cyfoesi, which lists Cadfan as king of the Welsh prior to Cadwallon, his decision would have had a surer foundation, and his ability to manipulate the pedigree of the kings of Gwynedd would be more readily explained. Conclusion: the Laurel Wreath We have brought the song to an end. So, Britons, give a laurel wreath to Geoffrey of Monmouth. He is indeed your Geoffrey, for he once sang of your battles and those of your princes, and he wrote a book which is now known as the “Deeds of the Britons” – and they are celebrated throughout the world.137 Who are these “Britons”, so beholden to Geoffrey of Monmouth? The Welsh, whom Geoffrey perniciously castigates in his history? The Bretons, who barely rate a mention in the poem for which this conclusion was written? The Britons of yore, who could look upon Geoffrey only as some distant, unknowable Homer? Or some combination of them all, the subject of an ironic paean for a people who only truly exist in Geoffrey’s pages? If there is any single conclusion to be drawn from this chapter, it is that Geoffrey of Monmouth was the master of his source material. He may have known the limitations of Breton source material, and he certainly knew the 136 137 Tolstoy, “Geoffrey of Monmouth”, p. 38. VM, ll. 1525–29: “Duximus ad metam carmen. Vos ergo, Britanni, / laurea serta date Gaufrido de Monemuta. / Est etenim vester, nam quondam prelia vestra / vestrorumque ducum cecinit scripsitque libellum / quem nunc Gesta vocant Britonum celebrata per orbem.” For use by the Author only | © 2020 Koninklijke Brill NV 66 Guy challenges presented by the relatively abundant Welsh source material. He understood how to use less tractable sources like bare genealogies and exiguous annals, and he understood how to weld them seamlessly to well-known narratives like Bede’s Ecclesiastical History. He consulted all the sources from Wales that he could find, in Latin and Welsh, but felt no compulsion to incorporate everything so discovered into his compositions. However, nothing absorbed into his work is left bare. Just as with the classical and biblical sources examined in the next chapter, Geoffrey deliberately sought to exercise the few readers who would have been conversant with the Welsh sources by masking his intertextual debts at every turn. But there was also an essential difference. Within the intertextual discourses of classical and biblical literature, Geoffrey was merely a passing participant; within the intertextual discourse of Brittonic history, Geoffrey was the enduring master architect.138 138 I would like to thank Paul Russell, Barry Lewis, and Rebecca Thomas for kindly suggesting improvements to various drafts of this chapter. For use by the Author only | © 2020 Koninklijke Brill NV