fet, pro-verb; so in

ï On ne for en in Old French cf. Foerster, Z. R. P., I, p. 92, note ; B. L. R., 2« Bé., vol. V, p. 93; Brinckmann, A. S. N. S., LUI, pp. 195-204; Perle, Z. R. P.

II, pp.^ 7-8 ; but cf. also Paris, Rom. VI, p. 304 ; VII, p. 465 ; Paul Meyer, Rom. VII, p. 343. Oberlin, Easai sur le patois lorrain des environs du comté du Ban de la Roche. Strasbourg, 1775, 8°, p. 105 : " Pour en se dit en et ne."

^ Cf. Foerster, Eric et Enide (1890), p. 312, note to 1. 1992. ' Hist. lit. de la Fr., XXX, p. 36, and Rom. XX, p. 149.

* Eric et Enide (1890), p. 332, note to 1. 3955.

• Oeuvres françoises, Paris, 1569, vol. II, fo. 5ro.

36 Notes on the Text.

459-460. Qu^el est en lui bien enploiee; cf. Auc. et Nie. 2, 1. 38 : " Ele fust bien enploiie en lui ; '^ Récits d^un ménestral de Reims : ^ " Et se vous veez que la couronne soie mieuz emploie en un de vous qu'en moi," etc. For qui, 1 SIng. Près, of cuidier, cf. Conquest of Irelandy^ where je cui : Kilhenni ; Roman de Thèbes,^ je qui : merci ; l^ Évangile de Nicodème,^ jeo cui ; Vie de St. Gilles^^ jo qui : lui.

546. estortre is the less fréquent but etymological Old French form ; more commonly, estordre.

553, 556. entî^eprendre de; cf. Berthe au grands Pieds ; ^ De ce vous vueil parler que vous ai entrepris^ and Scheler's note to the passage : '^ '^ B (Ms. B), dont vous ai (leçon préférable ; cf. notre expression : saisir qqn. d'une chose) '^ ; Garin de Loherain : ® mes d'une chose aveis moult eiitrepris.

561. Cf. Godefroy, Supplément p. 594 a: "Le faire = faire Tacte amoureux."

562. The branc in the manuscript is scribal ; cf. brans 598, 759 ; branc 727, 930 ; but branz 730.

860. The rhyme indicates an omission. Judging from the manuscript, it was only a single line. After 1. 853 the scribe repeated 1. 852, and then by compensatory error omitted 1. 860.

864. ai dî'oit; cf. passage in Méon's Nouveau Recueil:^ j'ai droit que j'ai apporté ici, etc., and in Etienne de Fougères.^^

990-994. n'a gaires — ja. The change of construction results from the complexity of the sentence, vos savez bien de quel proece (997) goes back to preuz (995).

1005-1006. Unless the manuscript is corrupt this is a case of the occasional assonance couplets found in rhymed poems. Cf. Ambroise, Estoire de la Guerre sainte, publiée par Gaston Paris, 1897, pp. XV -XVI.

1128. toiere. Godefroy has not found this word else where and suggests the meaning ^ mare boueuse.' He calls attention to its similarity to taier, ^ bourbier,' but taier is a word hardly more fréquent than toiere.

1188. The manuscript reading is oit, but cf. 1. 844, /aie. oit may be sporadic Eastern oi for ai, or simply a scribal slip.

1 Ed. by N. de Wailly, Paris, 1876, p. 148.

2 Ed. by F. Michel, London, 1837, 1. 1310.

3 Ed. by L. Constans, Paris, 1890, 2 vols. {S. A. T. F.\ vol. II, p. 346, Ms. A. *Ed. by G. Paris and A. Bos, Paris, 1885 {S. A. T. F.), p. 149 ; C, 1. 297.

5 Ed. by G. Paris and A. Bos, Paris, 1881 {S. A. T. F.), 1. 188.

« Ed. by Scheler, Bruxelles, 1874, 1. 166.

"^ Idem, p. 131, note to 1. 166.

^Bartsch and Horning, Lang. et Lil.fr., col. 113, 1. 7.

» Vol. I, p. 125, 1. 44.

^^ Livre deà Manières, éd. by F. Talbert, Paris, 1877, 1. 120.

The Manuscripfs. 37

THE MANUSCRIPXa

The sole Middle Age manuscript known to contain the Chevalier à VÉpée belongs to the Bibliotheca Bongarsiana at Bern and is no. 354 of that library. It is a quarto parchment of the fourteenth century ^ and is composed of 274 leaves. The writing is quite legible. From Folio 56 on there is a slight change in the handwriting : ^ the strokes above the letter i no longer hâve the délicate reddish tint, the sign for et always has the transverse stroke, which up to that point it has had only when initial, g initial in a word has the same form as 6r, and the formation of z is frequently modified somewhat. The handwriting in our poem is uniform.

The manuscript contains seventy-seven poems in whole or part, thechief of which are the Voie (V Enfer ^ the Mule sans Frein, the Cort Mantelj 576 lines of the Folie Tristan, a fragment of the Roman des sept Sages, and 9178 lines of the Parceval. The only one of thèse texts that has been critically edited from this manuscript is the Folie Tristan.^ It is used, along with other manuscripts, in F. A. Wulff's édition of the Cort Mantel} Severàl collections published in the first half of this century hâve also drawn from the manuscript.^

The manuscript was once the property of Henri Etienne^ (1528-1598) as is shown by his autograph upon it, and later belonged to Goldast von Heiminsfeld^ (1576-1635). It passed into the library of Bongars^ (1546-1612), with it became part of the Bern library, and is described in Sinner's catalogue® under its présent number (354). In 1809, Méon requested the loan of it through the French Minister of Foreign Affairs. The request was granted and the manuscript was transmitted to the Duke d^Otrante, Minister of the Interior. While in his hands, it disappeared. The

^ Cf. Hagen, Catalogus Codicum Bernensium, p. 338.

^ This différence was noted in the examination of the manuscript made at Bern in 1896. » H. Morf, " la Folie Tristan," Rom. XV, pp. 558-574.

*Ro7n. XIV, pp. 343-380.

'18 ont of the 30 sélections in the first volume of Méon's Nouveau Recueil de Fabliaux et Contes are taken from Bern ms. 354. A. Jubinal, in Lettre au Directeur de l'Artiste, prints no. 32, 33, 34, 36, 39 (Hagen's enumeration) of ms. 354, and in his Rapport à M. le Ministre de l' Instruction publique, he prints no. 35 and 48. Thos. VV^right prints no. 21, 22, 54 in his Anecdota Liieraria, and F. Michel no. 66 in Tristan, vol. I, pp. 215 ff.

^ Sinner, Catalogus Codicum Manuscriptorum Bibliothecae Bernensis, vol. III, p. 375.

' Hagen, /. c, p. 338. ^Jubinal, Rapport l. c.

9/.C., vol. III, pp. 375-380.

38 The Manuscripts,

solicitations of the Bern library, renewed in 1814, 1818, 1820, and 1824, were met by promises of restitution, but nothing came of them. Until the year 1836,^ no trace of the manuscript was found. In the course of that year it became known that Crozet, a bookseller, had bought it among the books of Manuel (1775-1827). Crozet demanded 1000 francs for it, and the Bern Library, rather than run the risk of losing it, bought it back at that price.

Besides the Bern manuscript, there exists an eighteenth century copy of it which was made for La Curne de Sainte-Palaye and is now in the French National Library (Moreau, 1720). The Chevalier à UÉpée occupies fo. 26ro.-47ro.^

The four volumes of Méon's revision of Barbazan's Fabliaux, which he published in 1808, contain numerous additions by him, but noneof them are drawn from the Bern manuscript.^ In 1823 appeared the two volumes of his Nouveau Recueil. The first volume contains 30 sélections, among them the chevalier.^ 18 of thèse ^ occur in the Bern manuscript. A comparison of Méon's text with the two manuscripts proves that he used the Sainte-Palaye copy. A statement made by him concerning the sources of his first volume falls in with this.^ The text of the poem contained in the 1829 édition of Le Grand d'Aussy's Fabliaux ou Contes'^ seems, from a comparison, to hâve been made from the Ste.- Palaye manuscript and the Méon édition. The text of the poem

^ This date and ail the facts relative to the loss of the manuscript are from Jubinal's Rapport and his Lettre au Directeur de V Artiste.

^ As the Sainte-Palaye copy is the basis of ail previous éditions of the poem its variations from the Bern manuscript are noted hère. Line 40. ouvrée for ovree. 59. qu* il for que il. S7. d'où for do. lOQ. s' escovri for sel covri. 147. devant for avant. 185. quil for quel. 188. savons for savon. 191. nos . . . vos for vos . . . nos. 227. tiers for tierz. 242. apprester for apresttr. 252. entrefroie for tnù^es tote. 253. damoiselle for damoisele. 27 d. briez for bries. 284. qu'il for qu el. 331. créante for crante. 342. fêtiez for feriez. 423. talent for talant 444. aporterent for aportenl. 452, la for de. 481. coûtes et de for coûtes de. 495. il i vit for il vit. 509. 6ms for braz. 510. solas for solaz. 552. as apris for à apris. 625. arriéres for arrière. 647. bras for braz. 672. doint for doin, 673. troves for trives. 688. erraument for erranment. 697. qu for que. 709. vivrez for vivez. 761. viendroit for vendroit. 874. osle for ostes. 982. merveilloit for mervelloit 983. qu il for qu el. 983. porpensoit for porpensot. 1053. l escondroit for les toudroit. 1087. sachez for sachiez. IIOS. merveille for mervelle. 1105. guerpi for guepi. 1115. pis for piz. 1122. le for lo. 1136. / abandone for i abandone. 1185. o for con.

' He gives the sources that hâve been used in this collection save for four fabliaux, and thèse do not occur in the Bern ms. * pp. 127-164.

5 In the Bern ms. ( Hagen's numbering) 8, 58, 70, 67, 3, 7, 55, 69, 51, 28, 37, 45, 63, 5, 20, 63, 65, 9. In the Nouveau Recueil, no. 1, 2, 5, 6, 7, 8, 10, 11, 12, 15, 16, 17, 18, 19, 20, 22, 24, 29.

•Méon, l. c, vol. I, p. V: "Une partie de ces Contes ayant été prise sur des copies faites pour M. de Sainte-Palaye, on y trouvera quelques mots évidemment mal copiés, et qu'il a été impossible de rectifier faute du manuscrit original."

'' Vol. I, Appendix, pp. 3-15.

The Manuscripts. 39

published by Jonckbloet ^ is taken from Méon's édition, evidently unsupplemented by the original manuscript or the raanuscript copy.

The customary abbreviations occur in the manuscript. Where there is no doubt as to how they should be resolved, I pass them over without notice. Et and moût are nowhere written ont in the poem, but I bave thus solved the signs tbat represent them. ol + cons. is soraetimes written o, sometimes ou, so it is impossible to décide absolutely whether to write moût or mot. G stands frequently for Gauvain or Gauvains. That mt with a bar over the m should be solved by -me7it and not -mant is shown by passages where the form is written out 130, 146, 150, etc.

Nasal consonants. ïhe bar over a vowel preceding a labial consonant I always solve by n since it frequently occnrs written out in the text {sanb- 465, 473, 1125, 1126; chanh- 486, 497, 512, 685, 815 ; conh- 1204) ; the same writing is used before a final consonant (tens 832 ; dedenz 392, etc.). Before m, n is written for Latin n ; cf. erranment 385, 688. For the abbreviation (before m) not representing Latin m, the manuscript gives no light, since feme is the only form of this class that occurs written out, but as the combination mm occurs nowhere in the manuscript, while the combination nm does occur, and as n before labials in the manuscript stands for Latin m, I solve the abbreviation as Prof. Suchier in his édition oï Aucassin and Nicolete does, by n. Thus ainme 1, conme 653, conmander 197, etc. There is no support in the raanuscript for not writing the first of thèse nasals. The bar before n is always solved n, painne 1176, enmôinne 137, etc. The final nasal representing Latin m is written out once, hom 620 ; so I write hom 190, probably kept from becoming hon by the oblique home. In the same way aime probably kept m in the first person, so I write aim 1069. The customary sign for com^ con occurs and is solved con,

The u of the manuscript is solved by v in avraiy savrai; cf. Foerster, Eric et Eiide, 1890, pp. 301-302, note to 1. 248, and Gaston Paris, Rom. XX, p. 148. a; is solved by us.

^ Roman van Walewein door Penninc en Pieter Vostaert, uîtgegeven door W. J. A. Jonckbloet. 2 vol. Leiden, 1846-1848, pp. 35-74.

MM

40 Previous Discussion.