^{Note 1 in page 407} A not uncommon phrase in Middle English poems:—see, e. g., Sir Orfeo, ll. 121, 175; Erl of Tolous (ed. Lüdtke), 481–82; Ywain and Gawaine, 4011; Seven Sages, 270, 2566; Sir Degarre (Abbotsford Club), 21; also Chaucer's Miller's Tale, 36, and Manciple's Tale, 36; cf. Zielke, Sir Orfeo, Breslau, 1880, p. 16.

^{Note 1 in page 409} Giles's translation, Six O. E. Chronicles, p. 151; cf. San Marte's edition, p. 66.

^{Note 1 in page 410} See Mayor, Thirteen Satires of Juvenal, 4th ed., 1886, i, 238, who remarks: “It was in the year 84, the fourth of Domitian, that Agricola was recalled from Britain, where the work of subjugation remained unfinished.” For references in classical writers to the use of chariots (esseda) by the Celtic warriors see Ludwig Friedlaender, D. Junii Juvenalis Satirarum Libri V, Leipzig, 1895, i, 253, note.

^{Note 2 in page 410} Calvo Neroni, iv, 38; cf. Mayor's note, i, 223 f.

^{Note 3 in page 410} I am indebted to my friend, Dr. R. H. Fletcher, for this suggestion. It is interesting to observe how Geoffrey's reference to Juvenal was misunderstood by Robert of Brunne:

Gode kyng he was, we find in boke;

A boke men calle it Juuenal;

Of stories it spekes alle;

At Gloucester it sais he lies,

And the quene, dame Genuys.

Thus the whole of the developed fictitious narrative of Arviragus's life is definitely attributed to the “stories” in the book called Juvenal. Fable certainly ever clothes itself anew. Citation of authorities evidently does not prove acquaintance with them.

^{Note 1 in page 411} Holinshed adds these interesting remarks (Bk. iv, ch. 3): “And heere to speeke my fansie also what I thinke of this Aruiragus, and other kings (whome Galfrid and such as have followed him do register in order, to succeed one after another). I will not denie but such persons there were, and the same happilie bearing verie great rule in the land, but that they reigned as absolute kings over the whole, or that they succeeded one after another in manner as it is auouched by the same writers, it seemeth most unlike to be true: for rather it maie be gessed by that, which as well Gyldas as the old approved Romane writers haue written that diuerse of these kings liued about one time, or in times greatlie differing from those times which in our writers we find noted. As for example, Juuenal maketh this Aruiragus of whom we now intreat, to reign about Domitians time. For my part, therefore, sith this order of the British kinglie succession in this place is more easie to be flatlie denied and utterlie reproved, than either wiselie defended or trulie amended, I will referre the reforming therof unto those that haue perhaps seene more than I have, or more deepelie considered the thing, to trie out an undoubted truth: in the meane time I have thought good, both to shew what I find in our histories, and likewise in forren writers, to the which we think (namelie in this behalfe, whilest the Romans gouerned there) we maie safelie giue most credit, do we otherwise neuer so much content ourselves with other vaine and fond conceits.”

^{Note 1 in page 412} It was a regular thing for Latin writers to utilize popular songs and stories in their accounts of historical personages. Compare, for example, the way in which Geoffrey's contemporary, William of Malmesbury, wrote (ca. 1142) of Gunhild, daughter of Cnut the Great, who married King Henry, afterwards the Emperor Henry III, in 1036. The particular ballad used by William is moreover of especial interest to us in this connection because it is closely allied to the “Breton lay” of The Erl of Tolous (only preserved in English) which is strikingly like the Franklin's Tale in fundamental theme (see below, p. 437). On the Gunhild story in romantic literature see Child, Eng. and Scottish Pop. Ballads, ii, 37 ff., “Sir Aldingar.” This story was also attached to Matilda, daughter of Henry I of England.

^{Note 1 in page 413} With Chaucer's lines quoted above (p. 408), it is not without interest to compare the following from Laamon:

is lond heold Arviragus

& Genuis his quene,

e wifmon wel idone.

a isæh isses ledes king

at him ne derede naming.

us he wunede here

mid blisse twenti ere. (9653 ff.)

‘Arviragus and his queen Genuis, the very fair woman, held this land. Then the king of this people saw that nothing troubled him. Thus he dwelt here with bliss twenty years.‘

In the life of Arviragus the English historian, as usual, greatly expands Wace, his original, (chaps. xii-xvi occupy about 800 lines, 9186 ff.) but here without adding anything really significant.

^{Note 1 in page 414} Chron. of the Kings of Britain, London, 1811, p. 86, n. 2.

^{Note 2 in page 414} Dorigen's lover is called in Chaucer a “servant of Venus.”

^{Note 1 in page 415} Geoffrey says that the city Claudius founded was called after him Kaerglou, that is, Gloucester. In the following sentence, however, he adds another explanation of the name: “But some say that it derived its name from Duke Gloius, a son that was born to Claudius there, and to whom, after the death of Arviragus, fell the kingdom of Dimetia.” He doubtless felt forced to offer this alternative eponymous founder, because of the statement in the Historia Britonum (§ 49), at. the end of the genealogy of Vortigern. Guitolion of Gloui is there said to have been one of four brothers “who built Gloiuda, a great city upon the banks of the river Severn, and in British is called Cair Gloui, in Saxon Gloucester.”

It should be noted that Geoffrey had no more foundation for his statement that this Gloui was a son of Claudius, than for his statements about the marriage of that emperor's daughter to Arviragus. He would have asserted dogmatically that Gloucester got its name from Claudius, whom he chose to represent as his founder, with the assurance that his assertion could not be disproved (for was he not simply translating Archdeacon Walter's British book?), had he not been well aware that his contemporaries knew of the totally different and much more probable explanation in Nennius. So he decided to give both, albeit they were inconsistent; and, to bring them into some sort of harmony, he remarked that if the city was really called after a Gloui, this person was at any rate “a son that was born to Claudius there.” Alas ! for the genealogy of Vortigern, thus put to shame. Alas! for historic truth in the hands of a jesting prelate.

^{Note 2 in page 415} The form Juvenissa (Iuvenissa) in the abbreviation of Geoffrey by Ponticus Virunnius (p. 105) is of course a corruption. The Brut Gruffyd ap Arthur has Gennylles. The Brut Tysilio has no name. See San Marte's Geoffrey, p. 264.

^{Note 1 in page 416} Romania, viii, 61 ff.

^{Note 1 in page 417} Warnke, Lais des Marie de France, pp. 113 ff.

^{Note 2 in page 417} In the oldest Danish version of “Sir Olaf and the Elf,” the latter makes Olaf great offers if he will pledge his troth to her, among other things a horse that would go to Rome and back in an hour; see Child, Ballads, i, 375.

^{Note 1 in page 419} Roman de Merlin, ed. Sommer, pp. 222 ff.; English Prose Merlin, ed. Wheatley (E. E. T. S.), i, 361 ff. (cf. Mead's Introduction, pp. ccxxvi f.); Merlin, Paris, 1528, i, folio 145; P. Paris, Romans de la Table Ronde, ii, 174–180.—With Merlin's exhibition of magic, compare that by Guynebans. See Merlin, ed. Sommer, pp. 261 ff.; English Merlin (E. E. T. S.), i, 361 ff.; Merlin, Paris, 1528, i, folio 168; P. Paris, R. T. R., ii, 196; Livre d'Artus, P (Zt. f. franz. Sp. u. Lit., xvii, § 24). Also the illusions produced by Auberon before Huon, viz. a river created by enchantment (Huon de Bordeaux, ed. Guessard, Paris, 1860, vv. 3275–3284), a tower with battlements (id., vv. 3295–3299), a palace with viands prepared therein (id., vv. 3525–3529; 3592–3605). Cf. further the magic house built by Merlin (Huth Merlin, i, 149).—See also Tristan, ed. Michel, i, 222; ii, 102. I am indebted to Miss Lucy A. Paton for these references.

^{Note 1 in page 420} In the ballad of William the Conqueror, “written by Deloney, the ballading silk-weaver,” who died in or before 1600 (Percy Folio MS., ed. Hales and Furnivall, i, 151 ff.), we read:

To Dover then he tooke the way,

the castle downe for to flinge

which Aueragus had builded there,

the noble Brittaine kinge. (ll. 17 ff.)

The building of Dover was usually attributed to Julius Caesar (cf. Shakspere, Richard II, v, 1, 2); but Camden speaks of a chart formerly hung up there which stated that Arviragus afterwards fortified it and shut up the harbor. Arviragus plays a part in Cymbeline.

^{Note 1 in page 421} Arviragus seems to be composed of a prefix Ar-, a root -vir-, and a suffix -agus. The suffix appears also as -030s, -acus, -akos. Holder identifies the name (Altceltischer Sprachschatz, Leipzig, 1891, p. 243; of. pp. 59, 423, 1007) with Biracos, Biragos, Pirakos. According to D'Arbois de Jubainville, the form Biracos is a derivative of Birus, Birrus (see Revue de la numismatique françoise, 1860, p. 173, pl. 8, 11; 1861, p. 62; 1868, p. 414; cf. Revue Celtique, xi, 156 ff.). Likewise, Arviragus is a derivative of Arvirus (Arvirius).

^{Note 2 in page 421} In the genitive, Arveri; see Hübner, Inscriptiones Britanniae Latinae, 1873, nos. 1236, 1237 (Corpus Inscriptionum Latinarum, vii) = Ephemeris epigraphica, 7, p. 343, n. 1130 (cited Holder, p. 231).

^{Note 1 in page 422} See Rhŷs, Arthurian Legend, Oxford, 1891, p. 365, note 1; cf. Loth, Les Mabinogion, i, 197, note; ii, 293, 294, note 1.

^{Note 1 in page 423} Who was the original hero of the story, we cannot say. The Welsh Bruts, when translating Geoffrey, substitute for Arviragus the name Gweirydd. This is clearly not a phonetic equivalent of the name it supplanted; but it may well be the late Welsh form of an earlier Gwerid(th), which would correspond to all but the prefix of Arverius. There may then have been a person of this or similar name of whom the story was originally told, and it was Geoffrey who perhaps first identified him (Arverius, Gwerydd) with the chieftain whom Juvenal mentions as Arviragus. On Geoffrey's authority this became the established form of the chieftain's name, and it was used afterwards whenever stories were told of that prince.

Prof. Rhys, as the result of an ingenious series of conjectures (Arthurian Legend, pp. 365 ff.), suggested that Geoffrey's story of the marriage of Arviragus with Genuissa is “only another version of the story of Pryderi marrying a grand-daughter of Gloy w Wallt-Iydan,” as told in the mabinogi of Pwyll. Gloyw is evidently what suggested the combination to Prof. Rhys. But he is wrong in asserting that “Geoffrey of Monmouth has identified a Gloyw with Claudius Caesar.” Geoffrey knew from Nennius (see above, p. 415, n.) that the foundation of Gloucester was attributed to a Gloui (Gloyw), but he discarded the notion, and identified this traditional founder of the place not with Claudius, but with a son of that emperor, a personage of the historian's own creation, whom he represents as born there—and all this obviously, not because he had any story of Gloyw in mind, but merely to avert criticism by providing for the conflicting hypothesis. Thus, even if we accept as sure Prof. Rhŷs's observations that Arviragus is correctly rendered in Welsh by Gweirydd, and that Gweirydd is “probably” another form of Gwri, and that Gwri is an occasional name of Pryderi, it is nevertheless unnecessary to combine this Pryderi with Arviragus simply because the former married a granddaughter of a certain Gloyw, while the latter married a sister (fictitious moreover) of another person of the same name, unless some similarity between the marriage of Arviragus and Genuissa on the one hand, and Pryderi and Kigva on the other can be shown to exist.

We should observe in this connection that nearly all of what is peculiar to Geoffrey in his account of the marriage of Arviragus, the emphasis he lays on its unusual felicity, the statement that Gloucester was founded as a monument of it, the explanation that the Gloui after whom Geoffrey admits it may have been called was a son of Claudius born there, the remark that Arviragus was feared by the Romans more than any king of the time, the quotation from Juvenal in support of all this, etc., is not only not in the most remote manner suggested by the tale of Pryderi, but is even not to be found in the Welsh Brut attributed to Tysilio (translated San Marte, in his edition of Geoffrey, 1854, p. 517 f.).

Prof. Rhŷs's remark, moreover,—“The mythic element still further betrays itself in his narrative, when it describes Gweirydd helping to bring Orkney and the other islands into subjection to Gloyw (Claudius) ”—is of little consequence when we remember that Nennius, from whom Geoffrey borrowed, although he never mentions Arviragus, says of Claudius (§ 21): “He next sailed to the Orkneys, which he likewise conquered, and afterwards rendered tributary.”

The following triad (translated by Loth, Les Mabinogion, ii, 283) evidently does not antedate Geoffrey: “122 (Myv. 403. 24). Trois principaux rois de combat de l'île de Prydein: Caswallawn, fils de Beli; Gweirydd, fils de Cynnelyn Wledig; Caradawe, fils de Bran ab Llyr Llediaith.”

El païs ot un damisel, In Armorik, that called is Britayne,²

fiz a un conte, gent e bel. Ther was a knight that loved and dide

De bien faire pur aveir pris his payne

^{Note 1 in page 425} Works of Chaucer, iii, 481.

^{Note 2 in page 425} Note that the French lay begins in like manner:

Veritez est qu'en Neüstrie,

Que nus apelum Normendie (7–8.)

It was the regular way to begin a Breton lay, after the conventional short prologue, which is also in Chaucer; cf. “En Bretaigne jadis maneit” (Le Fraisne, 3; Yonec, 11); “En Bretaigne maneit uns ber” (Bisclavret, 15); “En Seint Mallo en la cuntree” (Laustic, 7); “En Bretaigne a Nantes maneit (Chaitivel, 9); ”En Bretaigne ot un chevalier“ (Eliduc, 5).

^{Note 1 in page 427} See Skeat, Works of Chaucer, v, 388; cf. Koeppel, “Chauceriana,” Anglia, xiv, 258.

^{Note 1 in page 428} In Eg., 141 ff., is an interesting passage beginning “Amurz n'est pruz, se n'est egals,” in which is shown the unwisdom of him who “Vuelt amer par seignurie,”—with which should be compared the passage in Chaucer beginning “Love wol nat be constreyned by maistrye” (36 ff.).

^{Note 1 in page 432} Some one, indeed, after observing its great similarity in particular passages to parts of several of Marie's lays, might possibly suggest that this only evinced Chaucer's familiarity with Marie, and that he deliberately put together a new story of which the various parts are simple echoes of her poems, and for this reason termed a “Breton lay” what was really his own invention. But such an hypothesis is manifestly untenable. It not only runs counter to all that we know of Chaucer's methods, but violates every probability based on other studies in literary history. Inasmuch, however, as the Breton lay of Arveragus and Dorigen does show such striking likeness to Marie's lays, it is not impossible that she was the author of the poem Chaucer had before him; but on this point we have no evidence, and such purely conjectural matters are perhaps hardly worth consideration.

^{Note 1 in page 433} Originals and Analogues of some of Chaucer's Canterbury Tales, No. 16, pp. 289 ff.

^{Note 1 in page 435} Previously told by him in his youthful work Filocopo (Bk. v).

^{Note 1 in page 436} Yet Landau says (Quellen des Dekameron, 1884, p. 94): “Chaucer hat wahrscheinlich auch Boccacio's Novelle benutzt.”

^{Note 2 in page 436} Op. cit., iii, 480–81, 484.

^{Note 1 in page 437} See the last quotation, p. 407, above, and the second, p. 408.

^{Note 1 in page 438} “True” and “troth” echo throughout the poem. The phrase “To plight one's troth,” occurs in ll. 210, 219, 276, 504, 550, 583; “troth,” with another verb, in 282, 294, 635; “true” in 43, 53, 216, 226, 236, 313, 506, 568, 592, 917, 928, 935, 985, 1023, 1037, 1056.—(Ed. Lüdtke, Berlin, 1881).

^{Note 1 in page 439} Lady Guest's translation, 1849, vol. iii; Loth, Les Mabinogion, 1889, i.

^{Note 1 in page 440} This lay presents the classical story of Orpheus completely transformed. All the changes made are in the direction of Celtic tradition. See Kittredge, “Sir Orfeo,” Amer. Journal of Philology, vol. vii.

^{Note 2 in page 440} Ed. Zielke, 461 ff.

^{Note 3 in page 440} Libeaus Desconus, ed. Kaluza, 171 ff.

^{Note 4 in page 440} Le Bel Inconnu, ed. Hippeau, 214 ff.

^{Note 1 in page 441} Ed. Skeat, E. E. T. S., 1865, ll. 1673 ff.

^{Note 2 in page 441} Ed. Bechstein, 1869, vv. 13108 ff.; cf. Miss Weston's translation, ii, 33 ff.

^{Note 3 in page 441} Ed. Kölbing, 1882, ll. 1805 ff.

^{Note 1 in page 442} The abduction of Guinevere is but another variant of this theme. The version of the story recorded by Hartmann von Aue in his Ywein is the nearest like the episode in which Tristan figures. Arthur, having promised an unknown knight (Milianz, Meleagant) an indefinite boon, felt obliged to give up Guinevere when she was demanded of him. She was, however, rescued by one of the king's followers specially devoted to her, here possibly Gawain.—Other more or less divergent accounts of the adventure are given by Chrétien, Ulrich von Zatzikhoven, the author of Diu Krône, Malory, etc.; see G. Paris, Rom., xii; Wend. Foerster, Introd. to Der Karrenritter; Weston, Legend of Sir Gawain, pp. 67 ff.; Legend of Sir Lancelot du Lac, 46 ff. Naturally, the king should never be the rescuer.

^{Note 2 in page 442} Bk. viii, ch. 29 ff. Cf. Löseth, Roman en Prose de Tristan, § 43.

^{Note 1 in page 443} It is even as Sir Walter Scott long since remarked in his edition of Sir Tristrem (p. 322):

“Good faith was the very corner-stone of chivalry. Whenever a knight's word was pledged, it mattered not how rashly, it was to be redeemed at any price. Hence the sacred obligation of the don octroyée, or boon granted by a knight to his suppliant. Instances without number occur in romance, in which a knight, by rashly granting an indefinite boon, was obliged to do, or suffer, something extremely to his prejudice.”

^{Note 1 in page 444} Skeat points out also (v, 387 ff.) slight borrowings from Persius, Dionysius Cato, Ovid, Boethius, and the Roman de la Rose.

^{Note 2 in page 444} 329–30, 401–406, 426–7, 517–527; 542–565.

^{Note 1 in page 445} Ed. Paulin Paris, Paris, 1875, p. 5.

^{Note 2 in page 445} Legend of Good Women, ll. 422–23.

^{Note 1 in page 446} G. de Machaut, Oeuvres [ed. Tarbé], 1849, pp. 11 ff.