SCOTLANE ARTHURIATH D DAN N LEGEND.

BY R. LOOMISS . , A.M., D.LiTT., Professor Emeritu Englishf so , Columbia University. In approaching the vast body of legends which grew up about the name ohera f o called Arthur faces i e d on , with innumerable questions t nonbu , e promptlloomo s obviouslo s p e u squestiod th an ys ya f histori no c fact: Did the Arthur of legend live, or was he a mythical figure? Not so long thero ag e were scholar eminencf so seriouslo ewh y believen i s d thawa e h t origin a bear-god or an agricultural divinity, arguing from the linguistic almosn therw a s no i e t t d ar.unanimouBu an root1 t ar s s vot f boto e h philologue flesf bloodd o historian d han n san .ma a Professo s s thawa e h t r Kenneth Jackson unsurpasses i o wh , commans hi n dvarioue i th f do s kinds of evidence, has expressed himself in a chapter due to be published in 1958 followss a : "Did King Arthur ever really existonle Th y? hones t answer is 'We don't know, but he may very well have existed.' The nature of the evidence is such that proof is impossible." In spite of this noncommittal answer Jackson does offer a good deal of testimony—some of it set down within a hundred years of the period when the hero of the Britons lived if he lived at all—and I believe that the lates mosd an t t authoritative writer Britain Dare so th kn ni Ages—Oman , Collingwood, Hodgkin and Stenton—display no scepticism on the subject.2 Though conclusive proof is lacking, it is hard to believe that the passionate devotio e Welshth f o n , Cornis d Breton e memorhan th o t s f Arthuyo r was evoke medievaa y db malHarrisd s an l silence Mr spitn I th .f ef o e o Gildas and others writing in the 6th century, it is far more likely than not that a British commander named Arthur flourished about the year 500, perhaps a little earlier or a little later. The next question which shoulI d lik historiraise o et th ed iscDi : Arthu r havparticulay an e r connection wita Scotlandhe w h wha no s s i t Wa ? Briton from the north like Cunedda? Or did he conduct campaigns like those of William Wallace (who, to judge by his name, must have been of Welsh ancestry) north of the Tweed and the Solway Firth? Skene first

1 J. Bhys, Studies in the Arthurian Legend (Oxford, 1891), pp. 25-40. E. K. Chambers, Arthur of Britain (London, 1927), pp. 206-11. . OmanC 2 , England before Normane th Conquest, . (London ed . M P h . 8t . Stenton f ,1 1938)21 . , pp , Anglo-Saxon England (Oxford, 1943), pp. 3 f. R. H. Hodgkin, History of the Anglo-Saxons, 3rd ed. (Oxford, 1952), pp. 122, 182. VOL. LXXXIX. 1 2 PROCEEDINGS OF THE SOCIETY, 1955-56.

elaborate e thesith d s that Arthur's military caree s confinee wa th r o t d North, Felloa Societe d th f an w o Antiquariesf yo , Joh . StuarnS t Glennie,

publishe 1 n 186i d n impressiva 9 e essay, Arthurian Localities, supporting Skene. Both relied largely on the identification of the sites of Arthur's

battles2 , as given by Nennius, with places in Scotland. Nennius, be it remembered Welsa s wa h, priest whocenturyh 9t , e earlth ,n yi collecte da miscellaneous bod factsf yo whad an , t purporte factse b o dt , about British history oft-quoten a n I . d passage,3 dealing wit perioe hth d right aftee th r deat Hengistf ho , Nennius says that Arthur fought against them—apparently e Jutee Saxonth th f Kent—and o s an s d defeated the mtwelvn i e battles. givee placH name e sth th f eeo where each victory occurred, and, since four battle listee ar s s occurrinda placee on ,n gi ther e ninear e name alln i s . It is, of course, easy to sit down with a large-scale map of Northern England Southerd an n Scotlan matcd dan hnine eacth ef ho name Nenniun si s with something resembling it on the map. That is what Skene and Glennie did. t ProfessoBu r Jackson submitte lis e scientifio dt th t c scrutiny, found an d

that only one site can be placed with any4 certainty in Scotland, namely, silva Celidonis or Coit Celidon, which is, of course, the Caledonian Forest. But it is hard to see how a general, operating against the Saxons and Jutes abou e yeath t r 500, could have encountered the mStrathclydn i e th r eo vicinity knowe w s , a ,where r therfa o es , wer Germanio en c invaders ther no later. The same objection holds against Urbs Legionis, i.e., Chester. Jackson conclude authentie s on tha e th t c victory liste Nenniuy db thas si Monsf o t Arthuy b n rwo abouBadonis,s wa tt thad i 500f an i t t musi , t have been fought on the eastern border of the Salisbury Plain, near Swindon or Faringdon. The other battle sites, wherever they actually were, may have witnessed the prowess of Arthur, or they may not. A priest of South Wales, living three hundred years later, is no trustworthy witness on matters of detail. Glennie found corroboration for his thesis in all manner of local associa- tions. e maintaineH 5 d e neighbourhoothath tn i "a r o t f o everf o de yon these battle-sites thus identified, we find existing, from the time of our oldest charters, and other documents, to this day, places with Arthur's name, or traditions of Arthur's history." There are two weaknesses in this argument. In the first place, associations which go no farther back than the 12th century—and none of them do—carry little weight, for even before 1100 the renown of Arthur was beginning to spread beyond Wales.

secone th n I d place sucf i , h evidenc valids ei ,Welse Waleth d h san borde r . StuarS . J t1 Glennie, Arthurian Localities (Edinburgh, 1869); also publishe s introductioa d n to Early English Text Soc., No. 36. . SkeneP . W 2 , Four Ancient Books of Wales (Edinburgh, 1868) , 52-58I , . 3 Chambers cit., . . . 1-12f P op ,8 pp . Lot,23 , Nennius VHistoriat e Brittonum (Paris, 1934) , 194-6I , . . ParalE , Legends Arthurienne (Paris, 1929),. f III 8 3 , • 4 Modern Philology, XLIII (1945), 44-57. 6 Glennie, op. cit., p. 108. ' : , " SCOTLAND AND THE ARTHURIAN LEGEND. 3

would hav estrongea r claim Mirabiliae , th for n i , include Nenniusdn i ' book, Arthu s connectei r d wit ha cair Brecknockshirn ni a buria d an el mound in Herefordshire,1 and that is about four centuries earlier than any similar topographical links with;Arthu e recordear r r Scotland r latedfo fo rs A . times, Welsh toponym d folk-traditioyan n matcca n h thos f Scotlano e d as evidence of interest in Arthur. But in neither country does the existence of an Arthur's Oven or an Arthur's Table mean that the historic hero ever bivouacked there. Of course, Arthur might have been a Briton of the North, but he could with equal probability have come from Cornwall or Wales. Indeed, he might have been born in eastern England before it was overrun by the Saxon Anglesd san . That Arthur live gavd Saxone dan e th goosa d thrashing Soute inth Englanf ho d abou yeae seem0 th t 50 r s pretty clear t wherbu , e he was born, no one knows nor is there the slightest probability that anyone ever will know. Some day a pillar stone marking the true place of his buria distincs (a l t fro holloe mth foun k woa t Grlastonburda 1191n yi y ma ) t tha birts bu hi t, h up certificatg bedu baptismar eo l register will como et light is something that even the most sanguine cannot hope for. Another ardent believer in Arthur's activities in the North is Professor Nitze. For him Arthur was a Roman, Lucius Artorius Castus, commander

of the 2 Sixth Legion, who is on record as fighting in Dalmatia and Armorica, havy ma e o foughwh n Britaii d t an wels na l agains Picte Scotsd th t san . But he lived in the 2nd century, and I find it hard to understand how this Roman general, even thoug middls hhi e nam Artoriuss ewa convertes wa , d into the champion of the Britons against the Saxons in the 5th century or slightly later, and how he came to rouse the ardent loyalty of all the descendant e Britons—Welshth f o s , Cornis d Bretons—fohan thousana r d years. Lucius Artorius Castus certainly lived and made history in a small wa Britain yi Arthur r ou Hadrian' n t i no . s wa se timeh t bu , A remarkable early reference, presumably to the Arthur of Nennius, has been brough fore recenn th ei o t t yearst bringi d thoug, ,an sus Arthurt hno , back to the North again. There is a Welsh poem called the Gododdin, dealing with a defeat inflicted on the Britons of Southern Scotland by the Angles.3 This poem s originait n ,i acceptel w formno s Welsi ,y db h scholars as a composition of about the year 600. In it a certain British warrior is sai havo dt e "glutted black ravene rampar th e city n th so f , o tthoug h t Arthur.no s wa " e h slee Thah w, is tman y Anglesprowess hi t s bu ,wa s t equano thao lt Arthurf o t .infee Whaon r n froca t m this fleeting allusion? Not that Arthur was a North Briton, not that he came from the same regio n e warriore Tweemosnortd th th di f f o s tho da s celebratee th n di

1 Chambers . f Lot cit.,. 9 . 216op ,23 p cit.,. . . op ,pp Faral cit.,. op , Ill. ,61 2 Pub. Mod. Lang. Assoc., Ksnv (1949), 585-96. 3 Gododdin, ed. I. Williams (Cardiff, 1938). Discussed by Williams, Lectures on Early Welsh Poetry (Dublin, 1944), pp. 65-70, and by K. H. Jackson in Antiquity, xin (1939), 25-34. 4 PROCEEDING E SOCIETYTH F O S , 1955-56.

poem, but simply that his reputation had reached the North and become established there round the year 600. There is nothing in the Gododdin rejeco forco t t historie s u eth t c probability that Arthu achieved ha r s dhi pre-eminenc campaigns hi y eb s agains Saxone th t Southern si n England. n factI vere t i seem,m y o t ssignifican t thae otheth t r early Welsh poems, which are roughly contemporary with the Gododdin and which deal with historic figures of the same general area, King Urien of Rheged and his son Owein, there is no further mention of Arthur.1 This silence would greay an tn bvictoriee wo r thre Nortunnaturae o d th o eha n hsi tw e h f i l generations earlier. This silence would be quite natural if one thinks of Arthur as one who fed the ravens with the carcases of Saxons in the Thames Valley or on the Berkshire Downs, where about 500 the Saxons would be reputatioe th founds i t I Arthuf . n Norto e th n hri whic attestes hi e th y db Gododdin, not his activity in any particular part of Britain. Even Nennius' specific reference to the victory of the Wood of Celyddon must be discounted on Nennius' own testimony that Arthur fought against Octha, King of Kent. Ther lona betwee p s ei g ga n Nenniunexe th td authosan lino rt k Arthur with Scotland FlemisA . h cleric, Lamber t OmerS wora f o tn i ,k entitled Liber Floridus and dated 1120 makes the following remarkable statement:2 "Ther Britain n Pictsi e lane s eth i f th d palaco a ,warrioe n ,i th f eo r Arthur, built with marvelou varietyd an t whic sn ar exploiti ,s historhi e hl th al sf yo seee b sculpture n ni ward o t an s s i fough e H . t twelve battles againse th t Saxons who had occupied Britain." Lambert then gives a list of the twelve victories of Arthur according to Nennius. What could have suggested this reference to the sculptures in the land of the Picts ? We do not know but we can guess. A Corpus Christi Cambridge manuscript, a century or so later in date than Lambert's work, has a gloss on Nennius which states that Carausius, the emperor, built a round house with polished stones (politis lapidibus) on the bank of the River Carron.3 Dr Steer of the Ancient Monuments Commission point t thaou s t this glos se Romareferth o t sn temple whic knows hwa earls na Ovee 129s ya th o Arthurf s nto 3o a t i r fo , was roun madd d an f polishe eo dy nea e stoneRivela th rd ran s Carron.

Did Lambert fin dsimilaa r cops glos hi Nennius— f yo n si 4 a gloss which con- nected this building with Arthur rather than Carausius? Coul thee dh n have converted the polished stones into sculptures and found a subject

. WilliamI 1 Proc.n si of Brit. Acad., xvm (1932) ff0 . 27 , Morris-Jones iCymmrodor,T n xxvm (1918), 64-71, 151-99. 2 Faral, op. cit., I, 256 n. Migne, Pat. Lat., CLXIII, col. 1012. R. S. and L. H. Loomis, Arthurian Legends Medievaln i t (NeAr w York, . 1938)15 . p , Lot, op. cit., I, 165, n. 8. John Leslie, bishop, of Boss, expressed the opinion.in De Origine Moribus

et Rebus3 Gestis Scotorum (1578) tha t e buildeArthuth stona s f o rwa r e house formerly existinr fa t gno from the River Carron. * Glennie, op. cit., pp. 42 f. SCOTLANE ARTHURIATH D DAN N 5 LEGEND.

for the sculptures in Nennius' list of Arthur's victories? These questions canno answeree b t e affirmativth e n negativedi th r o t esuc bu ;h efforts o elaboratt e scantth e y materials available were characteristi f otheo c r authors than Lamber t OmerS f o t . Geoffre f Monmouto y h -was writinga few years later, say in the early eleven-thirties, and it is abundantly clear thae Historiath t Regum Britanniae s largeli y compounde e rougth f hdo stone traditiof so n freely shape adorned dan Geoffrey'y db s imagination. In 1125, five years after the composition of the Liber Floridus, William of Malmesbury complete e Gestadth Regum Anglorum d testifiedan 1 that e subjecArthuth s f fantastio twa r c tale e Bretonss th tol y b d : "Artur Britonuo dequ m nugae hodieque delirant"—very significant testimono yt the role played by Breton story-tellers in the diffusion of the fascinating t wild incoherenbu an d t traditions whic d gathereha h d about Arthur. William did not associate Arthur in any way with Scotland, but he did mention Arthur's nephew Gawain unde name th r e Walwen said an d, that reigned hha e d over Walweitha, tha , Galloway.is t First t shouli , e db 2 observed that the form Walwen is not Welsh; nothing resembling it ever appear Welsa n si moss i h t i textt d probablan , e tha chroniclee th t r heard t froi m Breton lips—a corroboratio testimons hi f no y abou circulatioe th t n of Breton tales about Arthur l this f courseAl o ,. s easi , understano yt d whe realizee non s that many Breton lords fought for Willia Conqueroe mth r

at Hastings and were rewarded with lands, and others followe3 d after. Sir Frank Stenton declared: "There is hardly an English county in which the Breton elemen t somn foundi no d es i tan countie , s influencsit s deewa ep and permanent." For example, there was the great Breton earldom of Richmond in Yorkshire.* Breton lords would be followed, of course, by Breton entertainers and would welcome professional story-tellers to relieve the tedium of long winter evenings when there was no television and no radio. Bedier was clear on the point: after the battle of Hastings "toute la civilisation normande se trouva brusquement transplanted telle quelle dans les chateaux d'Outre-Manche t lee ,s jongleurs armoricain suivirensy t leurs patrons: jongleurs armoricains, mais plus qu'a demi romanises, mais vivan servicu a t seigneure ed s francais t contane , t pour leur plaire. " This

ifaca s f capitao t l importance. Breton story-teller singerd san s woul5 e db bilingual and could address Anglo-Norman audiences as well as Breton. Devoted to the memory of their ancestral hero, they drew heavily on the Welsh store-house of romantic fiction about Arthur. They adapted it to

1 Chambers, op. cit., p. 250. William of Malmesbury, Gesta Regum Anglorum, ed. W. Stubbs, Rolls Series . B. Loomis11 S . , I , Arthuriane , Walesth d an Legend (Cardiff, 1956) . 183-5pp , .

2 Chambers, op. cit., p. 250. William of Malmesbury, op. cit., II, 342. 3 P. M. Stentou, 1st Century of English Feudalism (Oxford, 1932), pp. 24 f., 28. 4 Victoria History of the County of York, North Biding, ed. W. Page (London, 1914), pp. 2 f. Annales de Sretagne, XLIII (1935) ff5 . 26 , 5 Thomas, Tristan, ed. J. Bedier (Paris, 1905), II, 126 f. 6 PROCEEDING E SOCIETYTH F O S , 1955-56.

the taste of Anglo-Normans and Frenchmen and thus spread his fame and Rouns thahi f o td Table throughou westere th t n world.1 The fact, however, that William of Malmesbury made Gawain king of Galloway probably has no significance. In the Middle Ages no principle of historiograph s morywa e solidly established tha e identh a that places took their names from persons. The early history of Rome was sufficient guarantee for it, and so was Vergil's Aeneid. William, therfore, had no difficult believinn yi g that Walweitha, Galloway, too nams kit e from Walwen, Gawain t ther Bu nothins e.i supporo gt t this derivation I hav d ean , else- where presented ample evidence to show that the name Walwen or Gawain had an origin which completely eliminates the possibility of a connection with Galloway. Indeed scholaro ,tw eminencef so . DJ , . Bruc Edward ean d

Brugger,3 agree2 that the association is no better than Geoffrey of Monmouth's linking King Leir with Leicester and Coel with Colchester. r GeoffreAfo s y himself, everyone knowe perpetratoth s s wa tha e h tr e mosth f to successfue oon f l hoaxee world'th n i s s history e Historiath , Regum Britanniae (c. 1136). Though born probably at Monmouth, and for a large part of his life a resident of Oxford, he was almost certainly of Breton extraction. Sir John Lloyd, Sir Edmund Chambers and Tatlock all came to this conclusion in view of his marked bias in favour of the Bretons as againsmanuscripe on Welsh,e n th i t d actualle h an 4t y refer himselo t s s a f Brito.5 That in the 12th century did not mean a Welshman, for he says him- self that after the time of Cadwallader the Welsh "non vocabantur Britones sed Gualenses. e conceiveH " 6 e bolth d d ide f fabricatinao a ghistor f yo Britain from the beginnings, taking advantage of the enormous prestige whic e Bretohth n story-teller build arounp ha su t d their ancestral hero. e phrasth f n WilliaI o e f Newburghmo e eveh , n mad e littlth e e finger of Arthur thicker than the loins of Alexander the Great.7 But he did his wor throug w cleverlo ksa s o h wh y e ver thaw th yf fe o t e Williaon s mwa e shamth . Almost everyone else accepte e Historiadth Regum Britanniae veracioua s a s chronicle. Geoffrey also linked Gawain with a similarScotlan n i d , an dthoug h s Willia a e same y th t f Malmesburywa m,o no e decideH . d that Lothian

1 B. S. Loomis, Arthurian Tradition and ChrMen de Troyes (New York, 1949), pp. 15-22, 27-32. Gottingische Gelehrte Anzeigen, 1890, pp. 788 ff. Kultur der Gegenwart, ed. P. Hinneberg (Berlin, Leipzig,

1909), TeU i, Abt. xi , 11-15,I , 60-65. Zeits. franz.. f Sprache . Literatur,u xx1 (1898), 79-162; XLiv2

(1922), 78-87. 2 Loomis, Arthurian Tradition, . 146-54pp . . DJ .3 Bruce, Evolution of Arthurian Romance (Baltimore, Halle . , n Zeits. 1923)1 2 . franz,f , I ,

Sprache, xxxni2 (1908), 59 f. 4 English Historical Review, LVII, 466 f. Chambers, op. cit., pp. 23 f., 90. J. S. P. Tatlock, Legendary History of Britain (Berkeley, Calif., 1950), 396-402, 439 f., 443. 5 Faral cit.,. op , in, 189, variants. 6 Loomis, Wales and the Arthurian Legend, pp. 181-4. Giraldus Cambrensis says (Opera, ed. Dimock, Rolls Series, vi, 179): "Usque in hodiernum, barbara nuncupatioue et homines Wallensea et terra Wallia vocitatur." ' Chronicles of the Reigns of Stephen, Henry II and Richard I, ed. R. Hewlett, Rolls Series, i, 11. SCOTLANE ARTHURIATH D DAN N 7 LEGEND.

got its name from a certain Loth,1 known in Breton tradition as the father of Gawain.2 So Gawain became for Geoffrey and his translators heir to the lordshi Lothianf po . Nothing much cam thisf eo , however rivaA . l tradition was soon in the field, which made Loth king of Orkney, and as a result Gawaibrothers hi d muce nan sar h more commonly associate romancee th dn i s with the Orkney Isles than with Lothian.3 Geoffrey arbitrarily took over Urien, historic king of Rheged in the late 6th century and ancestor of the present Lord Dynevor,4 made him a con- temporar Arthurf yo represented an , restores a royas m hi dhi o ldt thronf eo Moray by the generous Arthur.5 Likewise Angusel, king of Albania, that is, Scotland between Lothia d Moray nan s restore wa , Arthuy s db hi o t r kingdom.6 Geoffrey later make t quitsi e plain that Urien held Morad yan Angusel held Albany as Arthur's vassals; they were invited to the coronation as " reges subditi." There can be no doubt of the implications of this claim,

e historth d f Scotlanyo an d might have7 been somewhat d differenha t i f i t not been made. For when Edward I in 1301 had his secretaries draw up a statement of his rights to the overlordship of Scotland, they based it largel Geoffrey'n yo s t unlikelSistoria.no s i t yI 8tha Oxfore th t d magister, fataa sittindesks d hi l t ha ,ginfluenc a Scottish-Englisn o e h relationsd an , l don al scribblin y s eb linewa w t i fe s ga wit quils hhi l pen. But Geoffrey's book had a more immediate and romantic effect in Scotland. Characteristically, he had created from the Welsh name for the city of York, Ebrauc, a king of Britain named Ebraucus,9 and placed his reign to synchronize with that of King David of Judea. Geoffrey credited Ebraucus wit e foundatiohth t onl no f nYor yo t als f threkbu o e other towns: Alclud, Castellum Puellarum and Mons Dolorosus—four foundations in all.10 Alclud, defined as lying in the direction of Albany, is easily recognised s Dumbartoa e Clydeth t whan nGeoffreo d bu , di t y mean whe wrote nh e Castellum Puellarum n alternativa e gives a H ?t i s e nam r Oppidumfo e Montis Agned, but that does not help us much, for though Nennius placed Arthur's eleventh battle at Mons Agned, Professor Jackson assures us that impossibls i t i identifo e t Historia e th sitee d yth 1 1an merely implies tha, it t like Alclud "versus wa , s Albaniam." However knoo d we w ,tha 114n i t 1 King David of Scotland came down into England to fight on behalf of his

niece e Empresth , s Matilda, agains e supporterth t f Kinso g Stephen.n I 2

1 Faral, op. cit., in, 225, 237. 1 2 Loomis, Arthurian Tradition, pp. 148-50. 3 Ibid., pp. 71 f. See early editions of Burke's Peerage under Dynevor. 45 Paral, op. cit., Ill, 237. " Ibid. 7 Ibid., in, 242, "... reges etiam et duces sibi subditos ad ipsani festivitatem couvocare. ..." 8 T. Rymer, Foedera, ed. Clerke and Holbrooke (London, 1816), 1,932. Speculum, xxvm (1953), 121 f. 9 Tatlock. cit.,. 12 . op ,p

10 Faral cit.,. op , . Ill97 ,

1 Modern1 Philology, XLIII. 52 , Chronicles of the Reigns of Stephen, etc. ed. Hewlett, III, 75—83. 12 . PROCEEDING SOCIETYE 8 TH F O S , 1955-56.

this enterprise he was allied with his nephew, Robert of Gloucester, and spent several weeks at Oxford. Now Robert was the principal dedicatee of Geoffrey's Historic,1 and must have possessed a copy. Geoffrey was living at Oxford. -It is hard to believe that Robert did not mention the

sensationa historw ne l Kino yt g 2 David cald attentios an hi l , mane th o ynt reference welScotlandn o t s ca l e imaginOn . e tha proposee h t d that such knowledgeabla authoe th s summonee a rb n royae ema th o ldt presencd ean that a dialogue ensued, somewhat like this: "Master Geoffrey, I see that Ebraucus founded three Northe citieth n srecognisI i . e Alclud fortrese th , s Clydee on th t wher bu wha;d e an thi s ti s Castellum Puellarum?" Geoffrey replied: "Sire, it is your own royal fortress of Edinburgh." This dialogue, of course, is vouched for by no contemporary, but it is certain that in the next year, 1142, King David began using Castellum or Castrum Puellarum as an alternative title for his castle of Edinburgh.3 The name remained for centurie offician sa l designatio chroniclen i documentsd t sleasan a n i t d an , two French texts composed withi 1200f o ndecad a o , tw Fergus r e o Doon, d an identificatioe th Chateae th f nPucelleo s ude s with Edinburg clear.s hi 4 The titl always eha s bee nmysterya shouly wh ; d Auld Reeki knowe eb n e Castlth s Maidenf a eo e ChronicleTh 5s ? of Lanercost offere e explanath d - tion that King Edwin foundere th , , place s sevedhi n daughters therr efo safety. recenn I t s beetimeha n t a i surge o t de thadu name s th t ewa

nunner6 y established by St Monenna. But this is a mere guess, for there is

no record of such a house of virgins7 . Furthermore, why would David suddenly start using Castellum Puellaru yeae th rmn i 1142 , whe saine nth t had been in her grave for over 250 years? It is far more likely that the title was derived from Geoffrey. probables i t I , moreover, that Geoffret inven no titl arbie d d th te ydi an - trarily apply it to Edinburgh. For early in the 13th century the author of the Breton lai of Doon not only identified the Chateau des Pucelles with Dane- bore, but he also told a story of a beautiful and proud virgin, mistress of the country, who dwelt there with her maidens.8 She was won, however, by the hero Doon and gave birth to a son, whose story bears a marked resemblance to that of Gawain:9 he is born out of wedlock; after he is grown up, his

1 Chambers cit.,. . 41-44op ,pp . 2 . Ibid.,23 . p 3 A. C. Laurie, Early Scottish Charters Prior to A.D. 1153 (Glasgow, 1905), pp. 112, 123, 146.

G. Chalmers, Caledonia (Paisley, 1887-92), iv, 555-9. Fergus,1 ed. E. Martin (Halle, 1872), p. 106. Romania, vin (1879), 61. 5 On places called Maiden Castle, see B. E. M. Wheeler, Maiden Castle, Dorset (Oxford, 1943), pp. 8-11. J. Bhys, Celtic Folklore, Welsh and Manx (Oxford, 1901), I, 136 f.; Camden, Britannia, rev. E. Gibson (London, 1753), II, 822; Leland's Itinerary, ed. L. Toulmin Smith (London, 1909), iv, 31; v, 147. ' Chronicon de Lanercost, ed. J. Stevenson (Edinburgh, 1839), p. 179; trans. H. Maxwell (Glasgow, 1913), pp. 144 f. Speculum, xvii (1942), 253. W . WatsonJ . , Place Names (Edinburgh, ,1926) . 156p , .

87 Romania, vm, 61-64. .... 9 Loomis, Arthurian . Romania,Tradition,f 2 11 . pp xxxix (1910), . Sommer19--23O . H . , Vulgate Version Arthuriane oth f Romances (.Washington, 1909-16) , 317II , ...... SCOTLAND AND THE ARTHURIAN LEGEND. 9

mother send awam shi y wit recognitioha rina s ga n token meete h ;s hi s father incognit n combati o , overthrows him d onlan , y whe s fathehi n r asks his name, does recognition follow. Now in early French romance Gawain's mother was named Morcades, and Morcades also appears as a quee castla n i ladies. f eo Most importan face th t s thati foun ti r romances

Morcade e wif f th Kino es i s g 1 Lot f Lothian.ho Other facts show that

Morcades is simply a variatio2 n on the name Morgain la Fee.3 knoe Sow w tha e Castlth t Maidenf eo Edinburghs swa knoe W w. that the mistress of the Castle of Maidens was Morgain la Fee. We know that, though she had various lovers and at least two husbands, one of them was King Loth of Lothian. As Queen of Lothian, Morcades or Morgain could have dwel severan i t l castles wit r attendanhhe t maidens t nonbu , e would have e grea welth o suite s s ta r l fortresdhe f Edinburgho s . Onc e traditioeth n was established that Morgain la Fee was wooed and won by Loth, the eponymous king of Lothian, it was a matter of plain logic that Edinburgh wa Castle sth Maidensf eo . Wha earle th t y n happene e I yearth . f it o s e dse thethisI s s nwa a , 12th centur laterd yan , Breton conteurs came acros e Tweeth s fino dt a d •welcome from the Breton and Norman lords. It is not without significance that the first Scottish ancestor of the Stuart line was a Breton, whom Davi dI appointe dapifers dhi r steward o Bretoe Th . n entertainers were onlreado to y localiso yt e their tales wherever they found patronsd an , they have left their traces in three Breton lais centered in Scotland—Doon, already mentioned, Desire Gurund an (preserved onl Norsa yn i e translation).4 They were presumably responsible for linking Morgain la Fee to King Loth, for linking Lot Lothian o hthut r fo s d placinan , g Morgair maidenhe d nan s in Edinburgh castle. I have little doubt that Geoffrey, himself a Breton, had picked up this tradition when he boldly placed the Castellum Puellarum among King Ebraucus' foundations in the North. It is proof of his astute- nes softeo s tha e nh t appropriate s usehi s o t dtradition s which already enjoye certaida n currency. A remarkable confirmatio thif no s inference regardin e connectiogth f no Morgai wite Fe h n a Edinburgl h comes fro ballade mth Queene ,Th of Scotland,5 collected by Peter Buchan, corresponding member of the Society of Anti- quarie f Scotlando s . Scholars recognised over sixty year thao e ag sth t latter par thif to s poem tol dstora y obviously paralle stora o Caradof lt y o c e Shor relateoe th m firsf th Ar tt n di continuatio f Chretieno Troyes'e nd s

1 Romania, xxxix, 19-23. J. L. Western, Legend of Sir Perceval (1906-9), i, 193. 2 Loomis, Arthurian Tradition, p. 114. 3 Ibid., pp. 112-6, 302-6, 451-7. 1 Ibid., . 112-6pp , 269-72 . f Desire, 0 29 , Graelent, Melion,d an . B.ed M. Grimes (New York, 1928), pp. 52-75. Studia NeopMlologica, sry (1942), 1-24. F. J. Child, English and Scottish Popular Ballads, V, 176 f. P. Buchan, Ancient Ballads and Songs Northe oth f 5 of Scotland (Edinburgh, 1828) , 46-49I , . 10 PROCEEDING E SOCIETYTH F O S , 1955-56.

Perceval.1 The ballad, therefore, contained Arthurian material dating back e yeath ro t 1200. What abou e firsth t e tQueen th par f f Scotland?o o t Three point noteworthye sar Queen'e Th . 1 s: abod s "Reekie'ei s towers," i.e., Edinburgh Castle; 2. Her bower has "pictures round it set." 3. She trie seduco st youne eth ghave herow e w vainn seei , No n. that Morgain was the mistress of the Castle of Maidens; three medieval authors give elaborate description murae th f so l painting r palace;he repeatedln se i sh 2 y trie o seducdt e knight f Arthur'o s s rebuffedf thess o wa court l ed Al an .j tradition representee sar Frence th n di h Mort Artu, which places Morgain's castle, where she had tried to seduce Lancelot in vain, within two days' ride of Edinburgh (Taneborc), describes its mural paintings, and tells how Arthur was served at table and attended to his bedchamber by damsels only.

This French romance Scottisth d an he ballad, though separaten a y b d 3 interval of more than six centuries, preserve variant versions of the same tradition associatioe Th . Morgaif no n with Edinburgh was, therefore, very old and very persistent, even though her name was early lost and she became anonymoue th s Quee Scotlandf no . e inconsistencstruce b th y y Onkb ma e f thesyo e traditions aboue th t mistres Castle th tha e betrayef e Maidensb f so eo t th i n deserted Ca dan . d virgin of the lai of Doon, the wife of King Loth of Lothian, mother of Gawain, lustfue th d l an temptres Scottise th f so h ballad were e th originall d an e yon same, were all avatars, as it were, of Morgain la Fee? Anyone who makes e slightesth t investigatio f thino s fascinating figur medievan i e l literature will find that she is an extreme example of Vergil's characterisation of her sex: "varium et mutabile."4 And Mark Twain's Connecticut Yankee, who had made her personal acquaintance, declared: "I have seen a good many kinds of women in my time, but she laid it over them all for variety." Not only was Geoffrey's Castellum Puellarum taken seriously as a name for the Scottish capital, but the third of King Ebraucus' foundations, Mons Dolorosus, was also identified with what was probably the most imposing Roman ruin in all Scotland, the fort of Trimontium in the parish of Melrose. authoe Th 5Fergus,f o r Guillaume le Clerc, writing e earlth n yi 13th century, shows a detailed and accurate knowledge of Scottish geography

from Carlisle and Jedburgh in the south to Dunnottar Castle in the north. 6 1 Modern Language Notes, xm (1898), 417 ff. Romania, xxvm (1898), 214-31. Continuations of

Frenchd theOl Perceval f Chretieno Troyes,e d . Wed . Koach (Philadelphia, 1949) , liiiI , , 169-223. 2 B. S. and L. H. Loomis, Arthurian Legends, pp. 16 f., 24 f. 3 Mort Roie l Artu,. FrappieJ . ed r (Paris, 1936) . 44-47pp , . R. LoomisS . Arthurian,e Walesth d an Legend, . 105-30pp . . CurieJ 54 , Roman Frontier s People, It Postd Forte an th of Newstead Parishe th n i of Melrose (Glasgow, 1911). Brugger in Miscellany of Studies in Romance Languages and Literatures in Honour of L. E. Kastner (Cambridge6 , 1932) . 94—107. Leggepp ,D . M ,. "Some Note ihn so e Roman Fergus,"e d . Dumf.inT d an Gall. A.S., xxvn (1950), 165-7. SCOTLANE ARTHURIATH D DAN N LEGEND. 11

He make t quiti s e plain tha identifiee h t d Mont Dolerous wit hfortressa , overhangin a gdee p ravine d tha n alternativan a ,t e nam s Marowa e s (Melrose).1 When I inquired of Mr Angus Graham, the Secretary of the Ancient Monuments Commissio r Scotlandnfo whao t s a ,t this could mean, he promptly recognise descriptioa s a t di f Trimontiumno . Surely anyone seeking to identify the remains of a city or fortress, allegedly founded in remote antiquity, coul t havdno e done better tha imposine pitco nth t n ho g ruins of Trimontium, even though they did not actually go back to the days of Geoffrey's Ebraucus or the biblical David. Not only did the author of Fergus know Trimontium as Mont Dolerous, but even earlier, in 1171, the abbot of Melrose was referred to as the abbot of Mons Dolorosus,2 obviously because the original site of Melrose Abbey lay not more than a froo greae milmtw th r eo t Roma Tweede n th for n o t . Whereas the connection of King Loth and his wife Morcades or Morgain with Edinburgh was due to the artificial and comparatively late association of Loth with Lothian, there were earlier and rival traditions, and one of them linked Lothia Arthuriae th o nt n hero, Yvai r Ewainno . Chretiee nd Troyes referred about 117Yvain a o Loenele 0t nd d Loenean , s easili l y recognised as a corrupt form of Loeneis, a French name for Lothian.3 Now, as a matter of historic fact, Owein, the original of Yvain, was the son of Urien, King of Rheged, and fought with his father against the sons of Ida, king f Berniciao s century.h 6t authoritiee e ,Th th lat n ei s cannot place

Rheged with certainty,4 t Owein'bu s activities would surely have taken him,

if not into Lothian, a5 t least into neighbouring territory. Apparently a tradition connectin Urief o g n Oweinso wit e Scottise th n th h w whano h s i t Border Country lasted from the 6th to the 12th century. It is found not only in the name, Yvain de Loenel, but also in a Life of St Kentigern,6 written between 1147 and 1164, which records that Ewen, son of King Urien, celebrated by the histriones,7 that is, professional story-tellers, wooed the step-daughter of Leudonus, King of Leudonia (Lothian), surprised her beside a brook, ravished her, and begat on her the future St Kentigern. Leudonu f courseo , is s fictitioua , s character, invente accouno dt e th r fo t name of his kingdom, Leudonia, just as Brutus of Trojan descent was in- vented to account for the name Britannia, and as Albanact was invented by Geoffrey of Monmouth to explain the name Albania. So we have the testimon Lifee th f yt o Kentigerno S f middle thae 12tth th n hi tf e o centur y

Fergus, ed. B Martin, Halle, (1872), p. 121.

21 A. O. Anderson, Early Sources of Scottish History (Edinburgh, 1922), II, 275. 3 Chretien de Troyes, Erec, ed. W. Foerster (Halle, 1890), vs. 1707. Modern Philology, xxxvm, 282-6. For various forms of Lothian see ibid., xxii, 186-91.

4 P. Lot, Nennius, I, 73-75, 202, 224. Faral, op. cit., in, 43. 5 Proc. of the Brit. Acad., xvm, 292. Watson, op. cit., p. 156. H. M. Chadwick, Early Scotland (Cambridge, 1949) . 144-60pp , . 6 Lives of St Ninian and St Kentigern, ed. A. P. Forbes (Edinburgh, 1874), 245-7. 7 " Histriomim " is the correct reading of the manuscript, not" historiarum." Romanians (1893), 566. 12 ; PROCEEDINGS OF THE SOCIETY, 1955-56.

stora currens ywa t that Owei e lovee Urienf Ewenr th th no o f s o rn wa so , step-daughter of the King of Lothian. Still another and highly elaborated form of this tradition is preserved in Chretien de Troyes's poem Yvain, composed between 1176 and 1181.* Chretie t onlnno y refer Erecn i s Yvai o t latee th Loenele nrd n i poe t mbu , he tells how Yvain, son of Urien, setting out from Carlisle, came to a spring, kille e champioth d defendeo wh n e springth d d marriean , s widowhi d . This widowe tolds namear wa e ,e daughtew d,th s Laudinf wa o r d ean Duke Laudonez or Laudunet. Is it not obvious that Chretien has preserved name th Yvain'f n eo i s bride, Laudinefatherr than he i f d o t , an ,Laudonez , the very same tradition which the Life of St Kentigern gives us as to Ewen's amour wit step-daughtee hth Kine th Leudonif f go ro a? Thoug nature hth e of the amour differs widely in the two accounts, there can be little doubt that both have their origin in a legend that Owein was the lover of the King of Lothian's daughter. We have two other version Arthurian si n romanc f whaeo t must have been a vigorous tradition about the wooing and winning of the Lady of Lothian botn I authoe . hth substitute s traditionae r ha herth n r ofo ow s dhi l lover Owein. The author of Fergus relates that his hero, like Yvain, set out from Carlisle, soon after met "la dame de Lodien," and after many adventures wedde t Jedburga r dhe h (Gedeorde).2 Significantly puceleso tw , fro mcastla e nea wery b r e invite nuptialse th o dt , presumabl referencyn i e to Edinburgh as the Chateau des Puceles. Thus the romance of Fergus and e Ladth f Lothiayo n givethira s su d e traditio forth f mo n which appears in the Life of Kentigern and in Chretien's Yvain. A fourt foune hb forMalory'o n dt i ms i s tal Garetf eo Orknef ho e th n yi seventh Book of the Morte d'Arthur. Gareth takes the place of Yvain as hero, and his adventures at several points resemble those of Fergus, particularly in the great tournament which precedes the wedding.3 Now Gareth's bride is Dame Lyones, and Lyones is Malory's form of Lothian. Quite clearly, then, a common tradition underlies the love stories of Fergus and "la dam Lodiene ed Garetf o Damd d h"an an e Lyones. It is highly probable that Malory's source for his seventh Book was Anglo-Norman a n romanc Fergus,s a d t leasChretien'es a ol a s d ta eve ol s na s Yvain.* There existed, then, in the 12th century a flourishing tradition, first attache historie th o dt e c th Oweinf Urien o n n wo so ,, e tellinh w gho e Ladhanth f Lothianf yo do e authoe LifeTh th . f o ro f Kentigern refers explicitl e histrionese taleth th f so o y t safely abouma ye t Ewenon d an ,

1 Loomis, Arthurian Tradition, pp. 291, 301-3. Chretien de Troyes, Arthurian Romances, Everyman's • • Lib. . 180-208pp , . 2 Fergus, . edMartinE . ; ... 3 Ibid., pp. 177-89. Malory, Morte d'Arthur, Bk. vn, ch. 27-31. • Loomis, Arthurian Tradition,

. f 5 11 . • pp ... 4 .Loomis,- Arthurian tradition, pp. 439 f. - SCOTLAND AND THE ARTHURIAN LEGEND. 13

infer that this romantic legend, with its variations, was popularised by professional raconteurs, who, like actors, would employ intona'tion and gesture to give life and fire to their narratives. We can imagine that stories of the Lady of Lothian would be listened to with particular attention in the halls of the Anglo-Norman families who settled in the Lowlands of Scotland in the reign of David I. It is no wild fancy that Walter the Breton, ancestor of the Stuarts, with lands in Renfrewshire, was one of those who extended hospitalit histriones,e th o yt als f Bretooo n extraction coulo wh ,d tale telLadm e Ewef lhi th s Lothiao f yd o nan sinr no e g laa Doof th i o d nan mistress of the Castle of Maidens. Chretien followed his Yvain with his last and incomplete poem, Perceval or the Conte du Graal, composed about 1182. After about 6500 lines he interrupted the adventures of Perceval to take up those of Gawain and bordere th brough o t f G-alvoie. so m hi t 1 Gawai warnes dangee nwa th f do r of crossing the border but insisted on proceeding. He came to a river, was ferried across to a magnificent castle inhabited by ladies, and after some strange adventures in this Castle of Ladies, was ferried back across the river and succeeded in vanquishing the knight who guarded the water- crossin f Galvoiego . Now Galvoie is unquestionably Galloway, and one remembers that William of Malmesbury as early as 1125 represented Gawain as King of Galloway. Moreover, Galloway was bounded by the River Nith. The late Professor Ritchie in his very informative lecture on Chretien de Troyes and Scotland* pointe t thadou t Robert Mannin Brunnef go , writin 1328n gi , identifiee dth Castle of Maidens, not with Edinburgh, but with Caerlaverock, and Caer- laverock standmout e Nithe th th t lookt I f sRitchif a h. i o strona s sa d eha g case for identifying Chretien's Castle of Ladies, situated on the border river of Galloway, with Caerlaverock. But there is a stronger case, I believe, against it. If Gawain was con- ceive s Banda f Gallowaygo shoule h , ddifficulto n hav d eha enterinn i y g his own kingdom. Wolfram von Eschenbach, who, though dependent largely in his Parzival on Chretien, had other traditional sources, called the river on which the Castle of Ladies stood Sabins,3 and this has been recognised

1 Chretie Troyese nd , Percevalroman, ed. HilkA . a (Halle, 1932), vss. 6600-8648. 2 R. L. Graerne Ritchie, Chretien de Troyes and Scotland (Oxford, 1952), pp. 10, 23. Compare Loomis, Arthurian Tradition, . 502p , inde Scottisb xsu h Tradition. Though admiring Ritchie's lectur feelind ean g no prejudice against Scottish influence on Arthurian romance, I believe several of his suggestions errone- s derivatioousHi . f Bscladono s from Calathros (the Cars f Palkirks equatioeo hi d an f )Calathro no s with Calati questionablee ar r Loomise Se . , ibid., . 112pp , s 282statemenHi . t that Chretien never referre reaa o dlt perso mistakens ni . Ibid., . 492p . Ther evidenco n s ei e that anyone acquainted with British geography of the 12th century called the Britons of Cumbria Walenses or Waleis. Estregale8 is not Strathclyde but South Wales. Ibid., p. 71. The name Erec does not derive from Rhydderch but from Breton Guerec. Ibid., pp. 70-74. There are three Welsh place-names in Chretien's Perceval; Carlion, Gomeret, and Scaudoue representing scribal corruptions of Caerleon, Gwynedd and Snowdou. Ibid., . 481pp , 484, Roch490e Th Sangui.e ed t likelSanchare b no s o nyi t . Ibid., . 490p . 3 Loomis, Arthurian Tradition, p. 451. 14 PROCEEDING E SOCIETYTH F O S , 1955-56.

by Gernianists as the Severn, Geoffrey of Monmouth's Sabrina flumen. The Sever wels lnWa knowe Middl th boundard n ol ni ee Ageth s ya s between Englan d Wales.dan 1 Moreover e Frencth , h romance Questea L , du Saint Graal, confirms this identificatio rivee th tellin y f b rGalahado nw gho , like Gawain warnes wa , d against crossin borderse gth , Rivee camth o ert Severn, where stoo e Castldth f Maidenseo , overcam knighte eth guardeo swh e dth boundary, and was welcomed by the maidens.2 Thus Wolfram, and the author e Queste,oth f writing withi ngeneratioa f Chretieno o s r no , agree thae th t boundary was the Severn, and this is corroborated by the Welsh Peredur, of abou same tth e date, which makes Kaer Loyw, Gloucester abode th , ninf eo e enchantresses,3 and f course,o , Glouceste Severne th convergin e n o Th .s i r g testimon f Wolframyo e Questeth , Saintu d Graal, d Peredur,an pointing towar Severne dth , proves that beside traditioe sth n which localised Morgain's Castl f Maideneo t Edinburgha s , ther anothes ewa r whice h th place n o t di border stream between Englan Walesd dan . Chretien seem havo st e erred, then, in placing it on the border not of Gales (Wales) but of Galvoie (Galloway). For this departure from tradition Ritchie provides an adequate and realistic motive. The patron for whom Chretien undertook the composition

of Perceval Philips wa , Coun4 Flandersf o t supporteo wh , d Willia e Liomth n agains 117n i t d Henr3 senan I yI t troopt himafted Kine ai Bu th r.o st g of Scots was captured in 1174, the sons of Fergus of Galloway revolted against his authority, slew all the strangers whom they could lay hands on, and destroyed the royal castles. For several years the Nith was a frontier between Galloway and the rest of Scotland, and a very dangerous one to s quiti cross t I e . understandable, therefore y Chretiewh , n would have shifted the adventures of Gawain at the Castle of Maidens from the border of e bordeGaleth o f Galvoiest o r described an , tha e lattee o on dth n t s a r knight could pasd expecan s o returnt t Philio T .f Flander po n 118i s 2 e bordeth f Gallowao r y meant much more tha e bordenth f Waleso r f I . Robert Mannin hundree gon fiftd dan y years later identifie e Castlth d f eo Maidens with Caerlaverock, he may well have come to this conclusion on the basi Chretien'f so s poe r sommo e derivative fro. mit Fro 13th-centure mth y evidenc documentw tendence ne t th f ge eo e sw y mako t e Arthurian connections with Scotland marginaA . l notmanua n ei - scrip f Nenniuo t s Arthutellw sho r brought back fro e Holmth y Lann da image of the Virgin which he carried on his shoulders at the victory of Castellum Guinion, and how the fragments of this image were held in great veneration at Wedale.5 A later hand added the information that Wedale 1 Ibid. 2 Queste l Saintde . Graal,PauphileA . ed t (Paris) . 46-51pp , . Malory, Morte d''Arthur, Bk. xm, . 15 , ch14 . 3 Loomis, Arthurian Tradition, . 455p . LothJ . , Mabinogion, . (Parised d , 2n 1913) . f , n5 7 , 4 . Bitchief 7 1 . Kit.,. op ,pp . LotF 5 . cit.,,op , 195. I 8 . ,n SCOTLANE ARTHURIATH D DAN N 5 LEGEND1 .

wa towsa Lodonesin ni a (Lothian milex si ) s fro noble mth e monasterf yo Melrose. Ther a consensu s ei f opinioo s n that this refer e Stoo t th s wn i valley of Gala Water, where the church and a spring nearby are dedicated t Mary.S o t 1 Stuart Glennie cited a document of the year 1293, referring to a "furnus Arthuri r ove"o Arthurf identifiee no h d an , t dwiti buildinge hth , already mentioned, on the bank of the River Carron, which was known centuries later as Arthur's O'on.2 Evidently the notion that any unexplained ancient structure wen tday e bac th Arthuf wel s so o ka t s l establisherwa Scotlann di d as it was in Wales or Cornwall. From 14th-century documents Glennie collected further examples of the same tendency. In 1339 David de Lindesay made a grant to the monks of Newbattle of certain lands bounded on the west by a line starting at the "fons Arthuri," the spring of Arthur.3 From a parliamentary record of 1367 we learn that Dumbarto calles nwa d "castrum Arthuri." 4 mose Onth f teo curiou importan d Arthuriae san th l al f o tn localisations in Scotlan e identificatioth s di f Sinadoo n n with Stirling, whic I hhav e already treated elsewhere.5 This alternative nam Stirlinr efo firss gi t given by Froissart, who was secretary for Queen Philippa of England between 136d 1368 n 1361an I mad.e 5h excursion ea n into Scotland, sedulously gathering material whed an visite,e nh d Stirlingravels wa e gyh informed thatimed ol tKinf thie so th gs n Arthui castl s ewa r called Sinaudond an , ther occasion eo knighte nRounth e th f so d Table havresorted.e w ew No 6 plent recordf yo s about Stirling before 1365 t nowherdiscovee bu , on n eca r any verification of the claim that Sinaudon was an ancient name for Stirling.7 It was, in fact, a name which turns up, not infrequently, in Arthurian romance in various but recognisable forms.8 In Biket's Lai du Cor, an Anglo-Norman poem which goes back to Chretien's time, a King of Sinadoune was present at one of Arthur's feasts. In a continuation of Chretien's Perceval the boy hero declares tha "wase h t bor t Sinadonna . Beroul authoe th ,poea f o rm about Trista 1200). n(c , represented Isolt's squir departins ea g from Tintagel Cornwalln i , comin Caerleoo gt Soutn ni h Wales arrivind an , t Isneldonega , where he found Arthur seated at the Round Table. Le Bel Inconnu, dealing with the career of Guinglain, Ga wain's son, describes Sinadon as a ruinous daughtee th town. Kine t i th Walef n f gI o ro imprisone 9s swa fore th m n di

1 . Glennief R 6 7 J cit., . . Snellop ,pp , King Arthur's Country (London) . 213p , . Faral cit.,. op , 1, 148 f. Glennie, op. cit., pp. 42 f. Chalmers, Caledonia, I, 425.

32 . Glennief 3 8 cit.,. . op ,pp ' Ibid., p. 88. 5 Loomis, Wales and the Arthurian Legend, pp. 1—18.

6 Froissart, Oeuvres, ed. Kervyn de Letteuhove (Brussels, 1867), n, 313. Mackenzie. C . W 7 , Scottish Place-names (London, . 1931)81 . p , 8 Loomis Arthurian,e Walesth d an Legend,. 11 . p

. f Eenau 1 Ibid.,1 . Beaujeue tpp d Inconnu,l P.-William. Be , G . ed s (Paris, 1929). 9 16 PROCEEDINGS OF THE SOCIETY, 1955-56. ;

of a dragon. The hero broke the spell by kissing the monster, she was at once transformed int omatchlesa wero s tw ebeauty, e weddeth e d th 1n dan i city of Sinadon, now restored from its ruinous state. We are told that it was the capital of Wales. If we glance once more at Malory's seventh Book, we discover that Gareth was wedded to Dame Lyones at Kynke Kenadonne, describe sande castlth a cita d sn s whicydean "o a h marched nigh Wales." 2 It requires no Sherlock Holmes to perceive that Malory created the form Kenadonne by mistaking the C of a form Cenadonne for a hard c and, as he did elsewhere, substituting k.3 Such indications as we have, therefore, indicate that Sinadon or Senadon was in Wales, and several great scholars—Sir John Rhys, Gaston Paris, . Bruce—recogniseD . J d an wore th dfora n di Snowdon.f mo 4 They were enoughr thes fa a o r yrighg fa wentt .s ta t theno Afte d ,bu ydi r all, Snowdon mountaina s i towa castlea t name r no no , Th e. Snaudon, howevers wa , current in slightly different spellings as an appellation of the whole region about the mountain, as Snowdonia is to-day. What we should look for is a town in that region which was sufficiently imposing as to impress beholders capitae worths th a e Walef b lo yo t Arthur'n si s time. Gaimar chroniclere ,th , writing about 1150, gives us the clue.5 He says that there were three re- nowned citie firse second f Walesth so an teasile d ar d an ,y namem hi y db identifie s Caerleoa d d Caerwentnan e Romath , n walled town f Souto s h Wales. The third Gaimar called "la cite de Snauedun." This can apply only to the important Roman fortress of Segontium, situated on a hilltop, overlooking the Menai Strait, where at low tide the sandy flats are conspi- cuous. Eve earls na Nenniuss ya legendy da ' begud sha attaco nt h them- selves to the ruins of Segontium, and the Welsh tale, the Dream of Maxen, is partly localised there.6 Just as Caerleon in South Wales became famous as a resort of Arthur, so apparently did Segontium in North Wales under its Anglo-Norman name of "la cite de Snaudon," and in its forlorn state t offerei perfecda t settineerie th r e gdragoe storfo th f yo n princes Walesf so . t whe visite Bu sitne on Segontiu f sth eo m to-day stil,y thouglma e hon look down from it on the sands of the Menai Strait, only the foundations remain. Edward I is known to have removed ashlar in 1283 from the Roman havo t ruine buildin d eth e statel an susen th i f t t ga o di y e castlse e ew Caernarvon to-day. e "cit Sinadone Th ed 7 longeo "n r existed exceps a t

1 For the history of this motif see Studi Medievali, xvn (1951), ,104—13, and Tllrich von Zatzikhoven, Lanzelet, trans. K. G. T. Webster (New York, 1951), pp. 224-6. 2 Malory, Morte d'Arthur, . SommeO ed. H . r (London, 1889). , f 213 I , 9 26 , 3 Loomis, Arthurian Tradition, p. 116, n. 49. 4 Bhys, Celtic Folklore, , 562II . Histoire Litieraire a France,l e d 'xxx . Bruce(1888)n 4 17 ,, Evolution, n, 196. 5 Gaimar, Estorie des Engles, ed. Hardy and Martin, Kolls Series (London, 1888), I, 285. • Loomis Arthurian,e Walesth d an Legend, 3-9. pp . Mabinogion, . JonesT d trans,an . Everyman'.G s Lib., pp. 79-85. ' Cymmrodor., xxxm (1923), 94. , SCOTLAND AND THE ARTHURIAN LEGEND. 17

heaps of rubble and grassy mounds. In a generation or two nobody had idey an a wher t wasei anyond ,an wante o ewh coulo dt e d Th clai . mit Scots were the first to put in their claim; in fact, they put in two claims. Barbou Bruce,s hi n ri Boo , saykIV s that Kildrummy Castle, near Aberdeen, was called Snawdoune. 1 1375n i Thayears n Te wa .t s earlier Froissart heard the other claim at Stirling. this swa latte t I r claim whic reverberates hha d dow yearse nth . William of Worcester in the 15th century asserted: "Bex Arturus custodiebat le round table in Castro de Styrlyng, aliter Snowden West Castell." Sir David

Lindsay Complainte th n ,i of the Papingo (after 1530) carrie pleasane th n do t 2 illusion:3 Adew, fair Snawdoun, wit towriy hth s hie, Thy Chapell-royall, park, and Tabyll Round.

r WalteSi r Scott recalle e Ladyth n dLakei e oth f that "Stirling's towef o r yore the name of Snowdoun claims." At this very day, as I have been informe Angur M y sdb Graham title th Snowdo,f eo n Heral customarils di y bestowe Lyoe th n do Kin retirement s Armf go hi n so titlee Th ,. mentioned firs 1448n i t , doubtless derives fro romantie mth c nam f Stirlineo g Castle, t ultimatelbu t ygoei sAnglo-Normae bacth o kt n Romae namth r efo n fort overlookin Menae gth i Strait, "la cit Sinadon.e ed " Stuart Glennie mentioned, of course, the most familiar testimonial to e voguth f Arthuriao e n romance, namely, Arthur's Seat d carriean , e dth name back to the end of the 15th century.4 Kennedy in his fly ting with Dunbar refers to "Arthur's Sate or ony Hicher Hill." If there has ever beelegeny nan d linke thoso dt e magnificen y t comm no t o s ecragst ha t i , notice. Of one thing we can be sure: the name has no evidential value for Arthur's campaigning against the Angles in the neighbourhood of Edinburgh. A famous figure associated with Arthu s Merlin1150y i rb d ,an whe, n Geoffrey of Monmouth wrote the Vita Merlini, the wizard was already represente madnesf o resortins da Caledoniae t fi th a o st n gi n Forestd an , until very recentl s supposeyhi e see b t dDrummelzier.na o t grav s wa e

It is possible, moreover, that betwee5 n 850 and 1150 a Welsh poem was composed, Afallennau (The Appletrees), in which Myrddin speaks of his miserable state in the Coed Celyddon. But the connections of Merlin with Scotland and the Scottish legend of Lailoken are far too complex to be discussed . I musN"Chadwickd here K .d an . tan , refeM readee . th rH , o t r The Growth of Literature, I, pp. 105-14, 123-32, 453-57, and to a chapter by Mr Jarman in A History of Arthurian Literature in the Middle Ages, to be published in 1958, for the best-informed opinion on the subject.

1 Bruce, . 181vs Bk, . .IV 2 Gleiinie, op. cit., p. 57. 3 . Ibid.,58 . p 4 Ibid., p. 53. . f 5 2 Ibid.,7 . pp VOL. LXXXIX. 18 PROCEEDINGS OF THE SOCIETY, 1955-56.

Though, far into the 15th century, the Scots were eager to share with England in the glories of Arthur, it seems that with the accession of the Tudors Arthur's prestige was so successfully manipulated in the interest dynaste oth f y thamode on tattacf eo Englann ko debaso t s dwa e Arthur and to undo the work of Geoffrey of Monmouth. In 1527 Hector Boece, first Principal of the University of Aberdeen, published the Scotorum, Historia, and proved himself a patriot indeed, but as a historian he stooped e Oxforlevee th th f o lo d t magister. s methoHi s quitdi e obvious. First, Arthur was a bastard. To quote from the translation by Bellenden,1 "the treuth is, that liter gat him on ane othir mannis wife." Loth, on the other hand, is King of the Picts, and, as husband of liter's legitimate daughter Anna "richs wa , t commovit that Arthur, gottir adultrien i , sul preferrie db t s childrinhi o t , gotti lauchfuln ni l bed jusd ,an t heritouri e crowth f f o sno Britane."2 It was Arthur who, it is said, instituted the practice of gormandis- inthirteer gfo n days after Christmas consequenca s a d an , arms ehi y became so effeminat d sofan et tha r manfo t y years the littld ydi e "displeasure" to the Saxons. When, at the great battle which Boece does not name but which is easily identifiable as that of Mount Badon, the Saxons under Colger Childrid nan k fle surrenderer do Arthuro dt victore credie th th , r tfo y is not his but belongs to the Picts under Loth. Doubts, amply justified of course, are cast on Arthur's conquests. When, after Loth's death, the Britons recognised not Modred but Constantine as Arthur's heir, Modred protested, lawfua n i s l quarrelwa rebellionn t i i t d /no an t Arthue , thame th e n i rh t fatal battle Humbere e CamblaTh th . t no , nwitd Riverre h n bloodra , ; Arthur, Modred d Walwaan , n were slain; Guanorar r ladieshe he ,d an , knights •were captured and brought to the castle of Dundee. At Meigle is her sepulchre, held in special reverence of the people, "as the title writtin thairupon schawis: 'All wemen that strampis on this sepulture sail be ay barrantfruty thaif on eo t r ,bu wamb , siclik Guanors ea a was.'" Apparently, then locaa , l traditio alreads nwa existencn yi e before Boece y wroteb d an , the time that the poet Gray visited Meigle in 1765 and saw the tomb, the women of the place were prepared to assert that Queen Wanders "was riven detho t staned-horsey eb gude na er tha sdid.fo e tsh " 3 intereste Thuth n si s of Scottish nationalism Arthur and his fair spouse were loaded with ignominy and Modre whitewashes dwa exaltedd dan , muc Richars bees ha ha n I dII in recent times. Up to this point my paper seems to consist mainly of negatives. Arthur knowe w s himselfa , r nevefa o s r, crosse Tweede dth . Gawain never reigned

1 Boece, trails . BellendeJ . n (Edinburgh . ,Fletcher 1821)H . , E 66-86 II ,e , Se Arthurian. Material Chronicles,e th n i Studies Notesd an Philologyn i Literature,d an x (1906), 246. 2 Already Fordun (1384-7) had made this point, but justified the choice of Arthur ia preference Galwanuo t Modred san dictates da necessityy db , sinc lattee ewerth o rtw e mere children when TJther died. 3 Correspondence of Thomas Gray, ed. P. Toynbee and L. Whibley (Oxford, 1935), II, 891. SCOTLANE ARTHURIATH D DAN N 9 LEGEND1 .

in Galloway. His father Loth had no connection with Lothian except through the accidental similarity of the two names. The identification of the Castle of Maidens with Edinburg s nothin a falshwa t e gbu inferenc e froe mth supposition that Morgai a Feenl s Kin,a g Loth's wife, must have resided principae inth l fortres Lothianf o s . Sinadot Stirling cite no th ys t nwa bu , of Snowdonia, the Roman fort of Segontium. Gawain's adventures at the perilous border rive f Gallowao r y had, originally connectioo n , n wite hth Nith but with the Severn. But I have shown that the probabilities are strongl favou yn activitiee i th f historie o r th f so c Owein Urienf o n wels ,so ,a l as the romantic affairs of Yvain, having extended into regions included in the old Lothian. After all these denials and doubts, I may conclude on a more positive, more affirmative note. Tristan or Tristram was a historic personage, a King Pictse othath fd an ,t means coursef o , , that Scotlane on s claia n dca m m hi of its most famous sons. Probably no statement about him in the great romance of the Middle Ages is true; he did not live in or near Arthur's time; even his tragic love for Isolt is the invention of a later age. But that Drust, son of Talorc, was a Pictish king of the Dark Ages and that he was the original of the Tristan of romance is not a private opinion of mine, but is now the widely accepted vie Arthuriaf wo n specialists. First broache Heinricy db h Zimmer, the theory has been adopted by Ferdinand Lot, Joseph Loth, Deutschbein, Brugger, Bruce, Bedier, and Mrs Bromwich.1 I do not know scholafamiliay s i oan fo rwh r evidence wit th texte hrejectd th o san ewh . sit e Chronicleth Picts e n th I f Talorcana Drustf o o n so , s recordei , s a d reigning abou e yeath t r 780. Apparentl s reiguneventfus yhi nwa d an l

s demonstrabli t i shortt Bu . e tha romantia t c Perseu e 2 storth f d yo an s Andromeda type grew up about him which is reflected, not only in the Irish saga of The Wooing of Emer, but also in the French romance of Tristram.3 As Bedier long since worked out the itinerary,4 the tradition passed from Scotland through Wales, Cornwall and Brittany into France and England. TalorcanDrustf o n so , , appear Welsn si Drystans ha Tallwchf o n e so , H . become lovee s th Esyllt f o r , wif Marchf eo eventualld an , ycontemporara y of Arthur's. Miss Schoepperle proved tha tragie th t c tal adulterf eo s ywa elaborated unde influence th r Irise th hf o esag Diarmaif ao Grainne.d dan

From Wales the legend passed on to th5 e SW., King Mark became King of

1 Zeits. f. franz. Sprache u. Literatur, xm 1 (1891), 58 ff. Romania, xxv (1896), 15. Comptes Rendus de I'Academie des Inscriptions, 1924, p. 128. Beiblatt zur Anglia, XV (1904), 16-21. Modern Philology, xxxin, 231. J. D. Bruce, Evolution, i, 178 f. Thomas, Tristan, ed. Bedier (Paris, 1905), II, 105-8. K. Bromwich in Trans. Hon. Soc. of Cymmrodoriou, 1953, pp 38 f. 2 A. O. Anderson, Early Sources of Scottish History (Edinburgh, 1922), I, cxiii, cxxvii, n. 2. 3 Beiblatt zur Anglia, XV, 16—21. Thomas of Britain, Romance of Tristram and Tsolt, trans. B. S. Loomis (New York, 1951), p. XX. 4 Thomas, Tristan, Bedier. ed , n , G. Schoepperle, Tristan and Isolt, II, 395 ff. J. P. Campbell had already noticed the relationship

in hi5 s Popular Tales of the West Highlands (London, 1890-3), iv, 240. See also Romania, LIII (1927), 92-95. 20 PROCEEDING E SOCIETYTH F O S , 1955-56.

Cornwall castls identifies hi d ewa an , d with Tintagel, wher shatteree eth d walls, the cliffs, and the little cove are forever associated with the star- crossed lovers. Thence to Brittany, where we find a Tristan, lord of Vitre, firse th llte t n th hali hf o fcentury. 1 Ther Secone store th e th f yo d Isolt, Isol f Brittanyo t e deats addedth f Tristrawa ,d ho an , s localised;mwa 2 there 19te latth s ha ,s e a century , vestige traditioe th f so n survivee th n di form of folk-tales.3 In the course of the 12th century the long romance of Tristra Isol d fulls man twa y formed popularises wa , Bretoy db n conteurs, and takepoety b remarkablf p sno u e 13t e powerth hy centurB . y thers ewa no brance Arthuriath f ho n cycle more familiar throughout Europed an , Tristram and Isolt had become proverbial as ideal lovers.4 e migration spitth I f e legeneo th f no d through Wales, Cornwald an l Brittany place-nameo tw , s embedde Frence th n di h romances still remained suggeso t s Scottisit t h origin. From these French texts through Malory comes the familiar association of Tristram with Lyonesse, the land of his birth. havs Mosu f eo t been brough believo t p u t e that Lyonessa s ewa region lying lon d Wf Land'an o g. d sincEn s e sunk beneat e Atlantihth c waves r BivarM . , however s penetratinhi n i , g article, "Lyonessee th : Evolution of a Fable," has demonstrated that this notion arose in the

17th century as the resul5 t of curious confusions, and Arthurian students have been awar morr efo e than fifty years tha tMalory'f Lyonesso e on s s ewa spellings of Leonois, and that Leonois or Loonois was a common French nam r Lothianefo . Though f courseo , , Drust e Pictisth ,h 8t he kinth f go century, coul t havdno e known Lothian excep troublesoma s a t e neighbour, then settled by the Angles, it is probable that the association of Tristram with that region means that the derivation of the romance from Scotland was not completely forgotten.6 .This probability is strengthened by the fact that the poems of Eilhart von Oberg and Beroul state that when Tristram and Isolt were banished from King Mark's court, they fled to the forest of Morrois, and the only likely identification of Morrois is Moray— a very natural retreat for an exiled Pictish hero. I believe that the great majorit f Arthuriao y n scholars would agree with Bedier that Scotland wat onl birthplace sno yth historie th f eo c Drust legends t alshi f bu ,o o . By the early 13th century the romances of the Round Table, in oral and written form, had spread throughout Latin Christendom from Iceland e Holth yo t theLandw yNo for. integran ma l par Westerf o t n culture. But they are peculiarly the heritage of the English-speaking peoples, and

this heirloo deepls mha y impressed e marupoth Scotlandf t kno i . Unless Revue1 de Bretagne, xvm, 435—9.

2 Thoma Britainf so , Romance of Tristram, trans. Loomis . XXVIpp , . If 3 Annales Bretagne,e d LVI (1949), 203-10.

popularitr Fo ' . LoomisTristaf yo H . L nd , romancArthurianan . S . medievaB n e i e se Legends. t ar l 5 Modern Philology, (1950)L , 162-70.

Romania, xxv . Bruce(1896)D . J if6 .,1 ,Evolution, , 179I ,. 18042 . n , 6 SCOTLANE ARTHURIATH D DAN N 1 LEGEND2 . the late Professor Ritchie and I are completely mistaken, this Scottish impress on Arthurian literature is mainly due to professional reciters and singer 12te th hf so centur founo y wh welcomd a gavScotlano n ei ewh d dan their tales and their lais a setting that would appeal to their auditors. Even when they returned to England or to France, they retained the Scottish localisation. May I be so bold as to suggest that a translation into English of the unfamiliar and almost inaccessible (except in large libraries) Breton lais— Desire, Doon Gurund an (Norse)—an romance th f do Fergus,f eo accompanied bintroduction ya commentaryd nan , woul servica e db Scotlando et , even perhap profitablsa e publishing venture?