[211] After I had finished what I had to say in the way of introduction to this ballad, there appeared the study of the Trinkhorn and Mantelsage, by Otto Warnatsch: Der Mantel, Bruchstück eines Lanzeletromans, etc., Breslau, 1883. To this very thorough piece of work, in which the relations of the multiform versions of the double-branched story are investigated with a care that had never before been attempted, I naturally have frequent occasion to refer, and by its help I have supplied some of my deficiencies, indicating always the place by the author's name.

[212] The Bibliothèque des Romans, 1777, Février, pp. 112-115, gives an abstract of a small printed piece in prose, there assigned to the beginning of the sixteenth century, which, as Warnatsch observes, p. 72, must have been a different thing from the tale given by Legrand, inasmuch as it brings in Lancelot and Gawain as suppressing the jests of Kay and Dinadam.

[213] The custom of Arthur not to eat till he had heard of some adventure or strange news was confined to those days when he held full court, according to Perceval le Gallois, II, 217, 15,664-71, and the Roman de Perceval, fol. lxxviii. It is mentioned, with the same limitations, I suppose, in the Roman de Lancelot, III, fol. lxxxii, and we learn from this last romance, I, fol. xxxvi, that Arthur was accustomed to hold a court and wear his crown five times in the year, at Easter, Ascension-day, Pentecost, All Saints, and Christmas. The Roman de Merlin, II, lvi^b, or, as cited by Southey, II, 48, 49, says that "King Arthur, after his first dinner at Logres, when he brought home his bride, made a vow that while he wore a crown he never would seat himself at table till some adventure had occurred." In Malory's King Arthur, Kay reminds the king that this had been the old custom of his court at Pentecost. Arthur is said to observe this custom on Christmas, "vpon such a dere day," in Sir Gawayn and the Green Knight, Madden, p. 6, vv 90-99. Messire Gauvain says "à feste ne mangast, devant," etc., p. 2, vv 18-21. Wolfram von Eschenbach's Parzival does not limit the custom to high holidays, ed. Bartsch, I, 331, vv 875-79; and see Riddarasögur, Parcevals Saga, etc., ed. Kölbing, p. 26. Neither does Wigalois, vv 247-51, or a fragment of Daniel von Blühenthal, Symbolæ ad literaturam Teutonicam, p. 465, cited by Benecke, Wigalois, p. 436 f, or the Färöe Galians kvæði, Kölbing, in Germania, XX, 397. See Madden's Syr Gawayne, which has furnished much of this note, pp 310-12; Southey's King Arthur, II, 203, 462. Robin Hood imitates Arthur: see the beginning of the Little Gest.

[214]

[215] See also Brynjúlfsson, Saga af Tristram ok Ísönd, samt Möttuls Saga, Udtog, pp 318-26, Copenhagen, 1878. There is a general presumption that the larger part of the works translated for King Hákon were derived from England. C. & W., p. 47.

[216] That is, the current one. The Samson saga professes to supply the earlier history. Samson's father is another Arthur, king of England. An abstract of so much of the saga as pertains to the Mantle is given by Cederschiöld and Wulff, p. 90f. Warnatsch, p. 73 f, shows that the Rímur and Samson had probably a common source, independent of the Möttulssaga.

[217] By Warnatsch, who gives the text with the corresponding passages of the fabliau in a parallel column, pp 8-54: the argument for Heinrich's authorship, pp 85-105. 'Der Mantel' had been previously printed in Haupt and Hoffmann's Altdeutsche Blätter, II, 217, and by Müllenhoff in his Altdeutsche Sprachproben, p. 125. Of this poem, which Warnatsch, pp 105-110, holds to be a fragment of a lost romance of Lanzelet, written before the 'Crône,' only 994 verses are left. Deducting about a hundred of introduction, there are some 782 German against some 314 French verses, an excess which is owing, no doubt, largely to insertions and expansions on the part of Heinrich, but in some measure to the existing texts of the fabliau having suffered abridgment. The whole matter of the church service, with the going and coming, is dispatched in less than a dozen verses in the French, but occupies more than seventy in German, and just here we read in the French:

But possibly the last verse should be taken with what follows.

[218] In Hahn, Griechische Märchen, No 70, II, 60 f, a walnut contains a dress with the earth and its flowers displayed on it, an almond one with the heaven and its stars, a hazel-nut one with the sea and its fishes. No 7, I, 99, a walnut contains a complete costume exhibiting heaven with its stars, a hazel-nut another with the sea and its waves. No 67, II, 33, an almond encloses a woman's dress with heaven and its stars on it, a hazel-nut a suit for her husband. In the Grimms' No 113, three walnuts contain successively each a finer dress than the other, II, 142 f, ed. 1857. There are three similar nuts in Haltrich, No 43, and in Volksmärchen aus Venetien, Jahrbuch für r. u. e. Lit., VII, 249, No 12. Ulrich's mantle is worked with all manner of beasts, birds, and sea monsters, on earth or under, and betwixt earth and heaven: Lanzelet, 5820-27.

[219] I cite the text according to Warnatsch. Warnatsch thinks it worth noticing that it is the queen only, in Mantel 771 f, as in our ballad, st. 14, that curses the maker of the mantle; not, as in the fabliau, the gentlemen whose feelings were so much tried. These, like the queen in the ballad, ont maudit le mantel, et celui qui li aporta.

[220] Not even for Ginovere hübsch unde guot, or Enîte diu reine. The queen has always been heedful of her acts, and has never done anything wrong: doch ist siu an den gedenken missevarn, Heaven knows how. Ulrich is very feeble here.

A remark is here in place which will be still more applicable to some of the tests that are to be spoken of further on. Both the French fabliau and the English ballad give to the mantle the power of detecting the woman that has once done amiss, a de rien messerré. We naturally suppose that we understand what is meant. The trial in the fabliau is so conducted as to confirm our original conception of the nature of the inquest, and so it is, in the case of Arthur's queen, Kay's lady, and the old knight's wife, in the ballad. But when we come to the charmingly pretty passage about Cradock's wife, what are we to think? Is the mantle in a teasing mood, or is it exhibiting its real quality? If once to have kissed Cradock's mouth before marriage is once to have done amiss, Heaven keep our Mirandas and our Perditas, and Heaven forgive our Juliets and our Rosalinds! ("Les dames et demoiselles, pour être baisées devant leur noces, il n'est pas la coutume de France," we know, but this nice custom could hardly have had sway in England. Is then this passage rendered from something in French that is lost?) But the mantle, in the ballad, after indulging its humor or its captiousness for a moment, does Cradock's wife full justice. The mantle, if uncompromising as to acts, at least does not assume to bring thoughts under its jurisdiction. Many of the probations allow themselves this range, and as no definite idea is given of what is charged, no one need be shocked, or perhaps disturbed, by the number of convictions. The satire loses zest, and the moral effect is not improved.

[221] Nul femme que [ne] vouloit lesser sauoir à soun marry soun fet et pensé. T. Wright, in Archæologia Cambrensis, January, 1863, p. 10. Mr Wright gives one of the texts of Cort Mantel, with an English translation. We are further told, in Scalachronica, that this mantle was afterwards made into a chasuble, and that it is "to this day" preserved at Glastonbury. Three versions of the fabliau testify that Carados and his amie deposited the mantle in a Welsh abbey. The Skikkju Rímur say that the lady presented it to the cloister of Cologne; the Möttulssaga has simply a monastery (and, indeed, the mantle, as described by some, must have had a vocation that way from the beginning). "Item, in the castel of Douer ye may see Gauwayn's skull and Cradok's mantel:" Caxton, in his preface to Kyng Arthur, 1485, I, ii, in Southey's ed.; cited by Michel, Tristan, II, 181, and from him by Warnatsch.

[222] For this enchanter see Le Livre de Karados in Perceval le Gallois, ed. Potvin, II, 118 ff. It is not said in the printed copy that he sent the mantle [horn].

[223] Another copy, assigned to the end of the 14th century, from the Kolmar MS., Barisch, p. 373, No LXIX (Warnatsch).

[224] Warnatsch shows, p. 75 f, that the fastnachtspiel must have been made up in part from some version of the Mantle story which was also the source of the meisterlied, and in part from a meisterlied of the Horn, which will be mentioned further on.

[225] The Dean of Lismore's Book, edited by Rev. Thomas M'Lauchlan, p. 72 of the translation, 50/51 of the original. Repeated in Campbell's Heroic Gaelic Ballads, p. 138 f, 'The Maid of the White Mantle.' Mr Campbell remarks: "This ballad, or the story of it, is known in Irish writings. It is not remembered in Scotland now." Mr Wright cites this poem, Archæologia Cambrensis, p. 14 f, 39 f.

[226] Cf. Arthur in the Lai du Corn and Fraw Tristerat Horn, a little further on.

[227] Wolf at first speaks of the lai as being made over into the fabliau, in regular court style, ganz nach höfischer Weise, about the middle of the 13th century; then goes on to say that even if the author of the fabliau followed another version of the story, he must have known the jongleur's poem, because he has repeated some of the introductory lines of the lai. This excellent scholar happened, for once, not to observe that the first fourteen lines of the lai, excepting the fourth, which is questionable, are in a longer metre than the rest of the poem, in eights and sevens, not sixes, and the first three of the lai, which agree with the first three of the fabliau, in the eight-syllable verse of the latter; so that it was not the author of the fabliau that borrowed. Warnatsch (who has also made this last remark) has noted other agreements between lai and fabliau, p. 61. Both of these acknowledge their derivation from an earlier dit, estoire, not having which we shall find it hard to determine by which and from what the borrowing was done.

[228] Montpellier MS.

[229] Perceval exhibits agreements, both as to phrase and matter, now with the lai, now with the fabliau, and this phenomenon will occur again and again. This suggests the likelihood of a source which combined traits of both lai and fabliau: Warnatsch, pp 62-64.

[230] So amended by Zingerle from Syrneyer lant. A third copy is cited as in the Kolmar MS., No 806, Bartsch, Meisterlieder der Kolmarer Handschrift, p. 74 (Warnatsch). A remarkable agreement between the French lai, 94, 97, 99-102, and Wigamur 2623-30 convinces Warnatsch that the source of this meisterlied must have been a Middle High German rendering of some form of the Drinking-horn Test closely resembling the lai. See Warnatsch, p. 66.

[231] The king of Spain, who is again the poorest of all the kings, p. 206, line 32, p. 214, line 22, is addressed by Arthur as his nephew, p. 207, line 11, and p. 193, line 30. Carados is called Arthur's nephew in Perceval (he is son of Arthur's niece), e.g. 15,782, and Carados, his father, is Carados de Vaigne, II, 117. It is said of Kalegras's amie in the 'Mantle Rhymes,' III, 59, that many a lady looked down upon her. This may be a chance expression, or possibly point to the poverty which is attributed to the royal pair of Spain in Fastnachtspiele, Nos 81, 127, and in Frau Tristerat Horn. In Der Lanethen Mantel, Laneth is Arthur's niece, and poor: see p. 261.

The fastnachtspiel has points in common with the fabliau, and the assumption of a source which combined features of both lai and fabliau is warrantable: Warnatsch, pp 66-68.

[232] This is a thoroughly dissolute piece, but not ambiguous. It is also the most humorous of the whole series.

[233] Warnatsch shows that Heinrich cannot have derived any part of his Trinkhornprobe from the Perceval of Chrestien, characteristic agreements with Perceval being entirely wanting. There are agreements with the lai, many more with the fabliau; and Heinrich's poem, so far as it is not of his own invention, he believes to be compounded from his own version of the fabliau and some lost version of the Horn-test: pp 111-114.

[234] The principal variations of this name, of which the Welsh Caradoc is assumed to be the original, are: Craddocke (English ballad); Carados, Caradox (Cort Mantel); Karodes (Scalachronica); Caraduz (Crône, 2309, elsewhere) Karadas; Carigras, Kaligras (Rímur); Karodeus, Caraduel (Perceval, 12,466, 12,457, 12,491, but generally), Carados, -ot, or; Caraduel (Messire Gauvain, 3943); Garadue (Lai du Corn); Karadin (Möttuls Saga). Garadue probably == Caraduel, which, in Percival twice, and once in Messire Gauvain, is used for Carados, through confusion with Arthur's residence, Carduel, Cardoil. So Karadas is twice put in the Crône, 16,726, 16,743, for Karidol == Cardoil. Might not Karadin have been written for Karadiu?

[235] Tristan of Hélie de Borron, I, 73 verso, in Rajna, Fonti dell' Orlando Furioso, p. 498 ff. So in Malory's King Arthur, Southey, I, 297, Wright, II, 64. The Italian Tristan, La Tavola Ritonda, ed. Polidori, XLIII, pp 157-160, makes 686 try, of whom only 13 prove to be innocent, and those in spite of themselves. Another account exempts 2 out of 365: Nannucci, Manuale, II, 168-171.

[236] Un vasello fatto da ber, qual già, per fare accorto il suo fratello del fallo di Ginevra, fe Morgana: XLIII, 28; un bel nappo d'or, di fuor di gamme, XLII, 98. The Orlando concurs with the prose Tristan as to the malice of Morgan, but does not, with the Tristan, depart from prescription in making the women drink. Warnatsch observes that the Orlando agrees with the Horn Fastnachtspiel, and may with it follow some lost version of the story: p. 69.

Before leaving these drinking-tests, mention may be made of Oberon's gold cup, which, upon his passing his right hand three times round it and making the sign of the cross, fills with wine enough for all the living and the dead; but no one can drink s'il n'est preudom, et nes et purs et sans pecié mortel: Huon de Bordeaux, ed. Guessard et Grandmaison, p. 109 f, vv 3652-69.

[237] The Myvyrian Archæology of Wales, II, 13, triad 54 == triad 103, p. 73; p. 17, triad 78 == triad 108, p. 73.

[238] See the story in Le Livre de Carados, Perceval le Gallois, Potvin, especially II, 214-16, vv 15,577-638. "The Rev. Evan Evans," says Percy, Reliques, III, 349, ed. 1794, "affirmed that the story of the Boy and the Mantle is taken from what is related in some of the old Welsh MSS of Tegan Earfron, one of King Arthur's mistresses." This aspersion, which is even absurd, must have arisen from a misunderstanding on the part of the Bishop: no Welshman could so err.

[239] Myvyrian Archæology, III, 247^a, No 10, pointed out to me by Professor Evans. The story of the 'Boy and the Mantle,' says Warton, "is recorded in many manuscript Welsh chronicles, as I learn from original letters of Llwyd, in the Ashmolean Museum:" History of English Poetry, ed. 1871, I, 97, note 1.

[240] The horn is No 4 in Jones's list, and No 3 in a manuscript of Justice Bosanquet; the knife is 13th in Jones and 6th in the other; the mantle of invisibility is 13th in the Bosanquet series, and, under the title of Arthur's veil or mask, 1st in Jones. The mantle of Tegau Eurvron does not occur in the Bosanquet MS. Jones says, "The original Welsh account of the above regalia was transcribed from a transcript of Mr Edward Llwyd, the antiquary, who informs me that he copied it from an old parchment MS. I have collated this with two other MSS." Not a word of dates. Jones's Welsh Bards, II, 47-49; Lady Charlotte Guest's Mabinogion, II, 353-55.

Lady Charlotte Guest remarks that a boar's head in some form appears as the armorial bearing of all of Caradawc's name. Though most anxious to believe all that is said of Caradawc, I am compelled to doubt whether this goes far to prove that he owned the knife celebrated in the ballad.

[241] Heinrich seeks to put his wearisome invention off on Chrestien de Troyes. Warnatsch argues with force against any authorship but Heinrich's, pp 116 ff.

[242] Gawain had failed in the earlier trial, though he had no fault in mind or body, except that he rated his favor with women too high: 1996-2000.

In the first two probations a false heart is the corpus delicti; something is said of carnal offences, but not very distinctly.

The scope of the glove is of the widest. It takes cognizance of rede und gedanc in maids, werc und gedanc in wives, tugent und manheit, unzuht und zageheit, in men. One must have known as little what one was convicted of as if one had been in the hands of the Holy Office.

[243] Fastnachtspiele aus dem fünfzehnten Jahrhundert, Zweiter Theil, p. 654, No 80.

[244] From Vulpius's Curiositäten, II, 463, in Erlach, I, 132, after a printed copy of the beginning of the 16th century: Wolff, Halle der Völker, II, 243, from a Fliegendes Blatt of the 16th century. Two copies are cited by title in Mone's Anzeiger, VIII, 354 b, No 1; 378, No 165. Wolff prints Asion.

[245] A man must be "clear as beryl." One of the knights is tumbled into the water for having kissed a lady; but this is according to the code, for he had done it without leave. We learn from Perceval that kissing is permissible; marry, not without the lady be willing. 'Die bruck zu Karidol' is alluded to in 'Der Spiegel,' Meister Alswert, ed. Holland u. Keller, p. 179, vv 10-13. (Goedeke.) A man who has transferred his devotion from an earlier love to the image of a lady shown him in a mirror says the bridge would have thrown him over.

[246] Florimel's girdle is a poor contrivance every way, and most of all for practical purposes; for we are told in stanza 3 that it gives the virtue of chaste love to all who wear it, and then that whosoever contrary doth prove cannot keep it on. But what could one expect from a cast-off girdle of Venus?

[247] Nightingales in Grundtvig, No 274, A, B: see p. 64. See, also, Uhland, Zur Geschichte der Dichtung, III, 121 f.

[248] Neither the Sanskrit Shukasaptati nor Nakshabi's Persian version, made early in the fourteenth century, has been published. The Turkish version is said to have been made in the second half of the next century, for Bajazet II. Kadiri's is probably of the seventeenth century. An English and Persian version (Kadiri's), 1801, has the tale at p. 43; Small's English, from a Hindustani version of Kadiri, 1875, at p. 40.

[249] In the Contes à rire, p. 89, a sylph who loves a prince gives him a flower and a vase which will blacken upon his wife's proving unfaithful: Legrand, 1779, I, 78. I have not seen this edition of the book, but presume that this tale is entirely akin with the above.

[250] Cf. the King of Spain, at pp. 261, 263. The agreement may, or may not, be accidental.

[251] All these examples of the probation by flowers, shirt, or picture are noticed in Loiseleur Deslongchamps, Essai sur les Fables Indiennes, p. 107 ff; or in Von der Hagen's Gesammtabenteuer, III, lxxxiv ff; or in an article by Reinhold Köhler, of his usual excellence, in Jahrbuch für romanische und englische Literatur, VIII, 44 ff.

[252] Köhler, as above, p. 60 f.

[253] There is a stone in the Danish Vigoleis with the Gold Wheel which no one could approach "who was not as clean as when he came from his mother's body." Gawain could touch it with his hand, Arthur often sat upon it, and Vigoleis was found sitting on it. Nyerup, Almindelig Morskabslæsning i Danmark og Norge, p. 129, a chap-book of 1732. The stone is not quite so strict in the German Volksbuch, Marbach, No 18, p. 13 f, Simrock, III, 432 f. In the German romance no man less than immaculate in all respects can touch it: Wigalois, ed. Benecke, p. 57, vv 1485-88.

[254] Georgii Codini Excerpta de antiquitatibus Constantinopolitanis, in Corpus Scriptorum Historiæ Byzantinæ, XLV, 50 f, cited by Liebrecht, Germania, I, 264; De Originibus Constantinopolitanis, cited by Lütcke, Von der Hagen's Germania, I, 252, referred to by Liebrecht: both anecdotes in Banduri, Imperium Orientale, Anonymus de Ant. Const. p. 35, 96, p. 57, 162. The statue again in a note of Nic. Alemannus to Procopius, Arcana, 1623, p. 83: cited by Mr Wright, Archæologia Cambrensis, as above, p. 17. Mr Wright also makes mention, p. 16, of the blind dog that quidam Andreas (evidently a merry one) was exhibiting in the seventeenth year of Justinian, which, among other clever performances, ostendebat in utero habentes et fornicarios et adulteros et avaros et magnanimos—omnes cum veritate: Historia Miscella, Eyssenhardt, p. 377 f, l. 18, c. 23; Cedrenus, in the Byzantine Corpus, XXXIII, 657, Theophanes, in XXXVIII, 347 f.

[255] The Meisterlieder and the Indian tale are cited by Warnatsch. Virgil's statue was circumvented by an artifice which is employed in this tale of the Shukasaptati, and in other oriental stories presumably derived from it; and so was the well-known Bocca della Verità, Kaiserchronik, Massmann, pp 448 f. The Bocca della Verità bit off the fingers of perjurers, but took no particular cognizance of the unchaste. A barley-corn [grain of wheat], again, which stood on end when any false oath was sworn over it, Jülg, Mongolische Märchensammlung, Die Geschichte des Ardschi-Bordschi Chan, pp 250-52, cited by Benfey, Pantschatantra, I, 458, and referred to by Warnatsch, does not belong with special tests of chastity.

[256] The phrase looks more malicious than naïf, whether Austrian or Spanish, and implies, I fear, an exsufflicate and blown surmise about female virtue; and so of the Indian 'Volksglaube.' The candle-test is said to be in use for men in Silesia: Warnatsch, citing Weinhold, p. 58.

[257] These are all noted in Liebrecht's Dunlop, pp 11, 16, 33. The spring, says the author of Hysmine, served as good a purpose for Artycomis as the Rhine did for the Celts; referring to a test of the legitimacy of children by swinging or dipping them in the Rhine, which the "Celts" practiced, according to a poem in the Anthology: Jacobs, II, 42 f, No 125; Grimm, Deutsche Rechtsalterthümer, p. 935 (Warnatsch).

[258] Besides sources specially referred to, there may be mentioned, as particularly useful for the history of these tests. Legrand, Fabliaux, 1779, I, 60, 76-78; Dunlop's History of Fiction, 1814, in many places, with Liebrecht's notes, 1851; Grässe, Sagenkreise, 1842, pp 185-87; Von der Hagen's Gesammtabenteuer, 1850, III, lxxxiv-xc, cxxxv f.