803 Ji384 A STUDY OF THE SOURCES AND COMPOSITION OF THE OLD FRENCH LAI D'HAVELOC A DISSERTATION PRESENTED TO THE FACULTY OF BRYN MAWR COLLEGE IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY MAY, 1908 BY EDITH FAHNESTOCK EXCHANOT / X3. A STUDY OF THE SOURCES AND COMPOSITION OF THE OLD FRENCH LAI D'HAVELOC A DISSERTATION PRESENTED TO THE FACULTY OF BRYN MAWR COLLEGE IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY MAY, 1908 BY EDITH FAHNESTOCK THE MARION PRESS JAMAICA QUEENSBOROUGH NEW YORK 1915 e%c** ^t»' o^ Preface The publication of this dissertation, presented, in 1908, to the Faculty of Bryn Mawr College in partial fulfilment of the requirements for the degree of Doctor of Philosophy, has been delayed by various chances culminating in the loss of a manu- script left with a German publisher two years ago, and the necessity of preparing another manuscript for press here. Although phases of the problems concerning Havelok have been discussed in articles that have appeared during this time, nothing, so far as I know, has been published that affects my conclusions, or that would lead to other than slight modifica- tions in my work. The article by Professor Creek in the current number of the Engliscbe Studien^ on the author of the English Lay of Havelok^ in so far as it touches on my subject, points to the conclusions to which my comparison of the French and English versions has led me. I thank especially Mr. Foulet who has given me the benefit of valuable criticism, and Miss Helen E. Sandison who has most kindly assisted me in reading proof. Edith Fahnestock. Vassar College^ May, 1915. 313618 Contents Page I — Introduction . . . . . . . j II — A Consideration of Putnam's Study, 'The Lambeth Version of Havelok . . . . . 13 III — An Investigation of Kupferschmidt's Study, The Haveloksaga in Gaimar and the relation between . this version and the Lai d'Haveloc . . 25 IV — A Comparison of the French Lay and Gaimar's Version .... . . . . 57 V — The Composition of the Lai d'Haveloc . . no I Introduction The Lai d'Haveloc is an Old French poem which tells the story of an exiled prince, Havelok, who married the niece of an English king in whose kitchen he had served as scullion. After many adventures Havelok regained his own realm and also his wife*s English kingdom which had long been withheld from her by her treacherous uncle. Though this tale belongs to Scandinavian saga, it is told in the form of a Lay — that is, a short narrative poem containing a prologue and an epilogue, in which the author alludes to the Bretons and a lay of theirs on the subject of Havelok. Since, however, the story of Havelok is foreign to the Bretons and is yet connected with their lays, it seems probable that an investi- gation of this lay may throw light on the general question of the Bretons and their lays. Aside from its connection with the Bretons, however, the Lai d'Haveloc deserves special study as one of the best and earliest of the Old French lays, and as one of the several very interesting works dealing with the adventures of this hero which have come down to us. It is my purpose in the present study to determine, if possible, the source or sources of the poem, and to examine carefully the author's adaptation of his material to the literary form of the lay. The longest and most important of the versions of the story are the following: I. The Havelok episode in GefFrei Gaimar's Lestorie des Engles^ date 1 147-1 151.^ ^ There are 3 Mss. containing Gaimar's Havelok episode, which are desig- nated in the Hardy-Martin edition as follows : (M) British Museum, Ms. Royal 13 A. XXI; (Z)) Durham Cathedral, Ms.C. IV. 27; (Z) Lincoln Cathedral, Ms, A. 4/1 2 (formerly H. 18. 3). Gaimar's Havelok episode has been published four times, as follows : ( i ) by Sir Frederick Madden in his edition of Havelok, 1828; (2) by Petrie in the Monument a Historica Britannica, London, 1848; (3) by Thomas Wright (Caxton Society), London, 1850; (4) by Sir Thomas DufFus Hardy and C. T. Martin (Rolls Series), London, 1888, XCI, part 1, 1-34. 2 Gaston Paris, Lit.fr. au moye?i-age, 3ded., Paris, 1900, p. 273, dates the Estorie about 1148. He gives as time limits 11 45-1 151 (p. 15). 6 THE OLD FRENCH LAI D'HAVELOC 2. The Lai d'Haveloc le 'Danots^ date about 1 170 according to Gaston Paris/ the second half of the twelfth century according to Deutschbein.^ 3. The English romance called The Lay of Havelok the Dane^ date about 1300 according to Skeat/ a few decades earlier according to Deutschbein.^ 4. The Lambeth version^ an interpolation in the chronicle of Robert of Brunne, who wrote in the first half of the fourteenth century. '° The interpolation is of later date." The story of Havelok is found also in various chronicles in more or less condensed form." Heyman '^ and Brie''^ have al- ready shown that these accounts, generally called the minor versions, are derived from one or more of the principal versions with the addition of other details. Since I have failed to dis-^ cover primitive traits of the story in these minor versions, I shall pass over them in all the discussion of the sources of the 3 In 2 Mss. — (i) Herald's College, Arundel XIV, mentioned as H in the enumeration of the Mss. of Gaimar's Estorie, Hardy-Martin edition, preface, pp. xxix-xxxiii, and (2) Ms. P, Cheltenham, Thirlestane House, Phillipps 3713. Ms. H has been published four times, as follows : ( i ) by Sir F. Madden, in his edition of Havelok y 1828 ; (2) by Fr. Michel, Chroniques anglo-normandes, Paris, 1883, I; (3) by Thomas Wright (Caxton Society), London, 1850; (4) by Sir Thomas Hardy and C. T. Martin, in the work cited in note i. This edition also contains readings from Ms. P, which has never been published. 4 Gaston Paris, op. cit. p. 274. 5 Deutschbein, Studien zur sagengeschichte Englandsy Cothen, 1906, p. 98. 6 In I Ms., Oxford, Bodleian, Ms. Laud Misc. 108, published four times, as follows: ( 1 ) by Sir F. Madden for the Roxburghe Club, 1828; (2) by Skeat, E. E.T. S., ext. ser., IV, 1868, 1889 ; (3) Holthausen, Old and Middle English TextSy London, 1901 ; (4) Skeat's edition revised for the Clarendon Press, 1902. 7 Skeat, ed. 1902, Introd., pp. xxiv-xxv. s Deutschbein, op. cit. p. 97. 9 In I Ms., the Lambeth copy of Robert Manning of Brunne's Story of Englandy published by Madden and Skeat in the prefaces to their editions of the English Lay. Cf Skeat, ed. 1868, pp. xi-xiii, and ed. 1902, pp. xlv-xlvii. ^o Deutschbein, op. cit. p. 96. "See p. 13. ^2 For a list of minor versions see Madden, Skeat, Heyman, and Brie. ^3 Harold E. Heyman, Studies on the Havelok Taky Inaugural dissertation, Upsala, 1903, ch. V. The minor versions. M Fr. Brie, Zum for tie ben der Havelok-sagCy Englische Studien y 1905, XXXV, 359-371. INTRODUCTION 7 French poem, and shall consider only the principal versions as Kupferschmidt and Putnam have done.'^ This study, as has been said, is concerned with the second of the principal versions — the French Lay, and the other longer works will be considered only when necessarily involved in questions relating to the sources of the Lay. Much has been written on the origin and development of the Havel ok saga, as well as on the sources and relationship of the various versions. With the opinions concerning the legend itself, I have, of course, nothing to do ; I am considering the source and com- position of one literary product — the French Lay. Neverthe- less, what has been written concerning the derivation of the other versions, in so far as it affects the French Lay, must be reviewed in its general outlines before an investigation of the Lay itself is undertaken. When the question of sources was first approached by stu- dents of the English Lay, they mentioned the fact that the Havelok story was found in Gaimar's Estorie and in a French Lay, and without close examination ventured the hypothesis that Gaimar had abridged the somewhat longer French Lay. Skeat added that Gaimar must have had some additional source.'^ Petrie, who was skeptical as to Gaimar's authorship of the Have- lok episode in his Estorie^ thought that both this account and that of the French Lay reflected an earlier romance. '^ The first person who seriously considered the derivation of the versions was Kupferschmidt.'^ His study is important be- cause he first attempted to establish the chronological order of the versions, and also because his conclusions have been gen- erally accepted. The two points definitely settled by him were that the Havelok episode was written by Gaimar and therefore composed between 1147 and 1151, and that the French Lay, ^5 These versions contribute interesting points bearing on the history of the English romance. But since these points have not changed the conclusions reached in this study, and since the English romance is considered only when it is neces- sary to a discussion of the sources of the French poems, an investigation of them must be left for another study. ^6 Skeat, ed. 1868, Introd., p. iv, §5. See also Storm, who speaks of the French Lay as composed in the first half of the twelfth century and abridged by Gaimar about 1145. Efig. Stud. 1880, III, 533. =^7 Petrie, Monumenta Historka Britannicay London, 1848, I, 765, n. b. ^^ Die Haveloksage bei Gaimar und ihr Verhdltnis zum Lai d^ HaveloCy Roman- ische Studien, 1880, IV, 411-430. 8 THE OLD FRENCH LAI D'HAVELOC which dated from the beginning of the thirteenth century, could not have been abridged by him. The more recent date given for the French Lay is the second half of the twelfth century,'^ but since Gaimar's work belongs to the first half of the twelfth century, the chronological order of the versions established by Kupferschmidt still holds. Having overthrown the hypothesis that Gaimar abridged the Lay, Kupferschmidt had to decide whether the author of the Lay used Gaimar, or whether both came from a common source. He found some details in Gaimar that were not in the French Lay, and some in the Lay that were not in Gaimar. Since these details were found also in the English Lay, a version ^° which he claimed had developed independently beside the two French versions, he concluded that the only possible theory was that of a common source. Kupferschmidt's system of derivation is, therefore, the following:" Original version. Lost French version. ^Gaimar. English Romance. The Brut. Le Lai d'Haveloc. Interpolation in Robert of Brunne. Gaston Paris," in his brief notice of the article, pronounced Kupferschmidt's conclusions plausible ; and Grober ''^ added to his announcement of the study his opinion that Kupferschmidt's conclusions seemed convincing. This derivation, modified slightly, has since then been generally accepted. Two scholars, however, though they have not affected pre- vailing sentiment, have expressed opinions at variance with ^9 Deutschbein, op. cit. p. 98. 20 Kupferschmidt, op. cit. p. 429, a version **die sich unabhangig neben den zwei franzosischen Versionen entwickelt hat." A reference is given to Skeat, ed. 1868, Pref. §20. 21 Ibid.f p. 430. =^2 Gaston Paris, Romaniay 1880, IX, 480. **Ces conclusions paraissent plausibles." 23 Grober, Zeitschrift fiir romanische Pbilologie, 1880, IV, 466. INTRODUCTION 9 Kupferschmidt's conclusions. Axel Ahlstrom ""^ thought it ex- tremely unlikely that so highly developed a lay as would have been required for Kupferschmidt's hypothesis should have ex- isted as early as 1 1 50, and consequently deemed it simplest and safest to consider oral tradition, and not a finished metrical romance, as the basis of the French Lay. Ward ^^ decided that the French Lay was enlarged from Gaimar. He gives six reasons for his opinion. Of these Putnam says : " His attempt to de- rive the Lay directly from Gaimar cannot be accepted. Every one of his six arguments can be used with equal force in favor of a common source for Gaimar and the Lay." Ward does not answer Kupferschmidt's arguments ; it is possible, as Putnam suggests,^^ that he did not know of Kupferschmidt's discussion of the question in Romanische Studien. The Brut'''' is no longer considered an independent version, and the Lambeth Interpolation has been regarded as a principal version since Putnam derived it from the lost metrical romance instead of from Gaimar.^^ With this change of detail, Putnam's work only confirms Kupferschmidt's results. His scheme of derivation is as follows :^^ Unknown source or sources. Lost French version m r Gaimar. Lay. Lambeth English. Interpolation. 24 Axel Ahlstrom, Studier i den fornfranska lais litter aturen, Upsala, 1892, p. 124. ** These traditions were transmitted to the author of the lay by Anglo- Bretons." 25 Ward, Catalogue of Romances in the Department of Mss. in the British Mu- seum y 1883, I, 437-446. Cf. ** The existing text of the fuller Lay is of the 1 3th century ; but it may possibly, of course, represent an earlier Lay, which has been abridged by Gaimar. But we believe, on the contrary, that the fuller Lay was simply enlarged from Gaimar' s version, and enlarged by a minstrel who probably knew nothing about the Lincolnshire legends of Havelok." Again (p. 440), **The French Lay, then, we regard as a literary offshoot of Gaimar's version." 26 Putnam, The Lambeth Version of Havelok^ Publications of the Modern Lan- guage Association of America y 1900, XV, 9, n. 2. 27 Cf. Brie, op. cit. p. 363. **Heyman hatte demnach recht, wenn er der Havelok- episode im Brut jeden originellen zug absprach." 28 Putnam, op. cit. pp. 1-16. ^9 /^/^.^ p. 16. lo THE OLD FRENCH LAI D»HAVELOC This Kupferschmidt- Putnam derivation^ has prevailed since 1900— 1 90 1, in spite of the different views expressed by several 30 The principal scholars who follow Kupferschmidt or Putnam are : (a) ten Brink, tr. H. M. Kennedy, History of English Literature, New York, i 889, p. 1 8 1 . ** A Norman song of Havelok must have appeared as early as the beginning of the twelfth century ; and upon this was founded both the version of the saga in Gaimar, and a younger poem, probably not much later than 11 50." (b) Holthausen, Haveloky London, 1901, Preface, p. ix. **The existing Old French Lai d"* Haveloc le danois and the Havelok-episode in GefFrei Gaimar' s Estorie des Engles bear only a very remote resemblance to the English version. Cf. Skeat's edition and Kupferschmidt. . . ." (c) Anna Hunt Billings (^A Guide to the Middle Eng- lish Metrical Romances y Tale Studies in English , 1901, IX, 21, 23) refers to Kupferschmidt. (d) Skeat in his edition of 1902, Introduction, p. xlvii, cites both Kupferschmidt and Putnam, as does (e) Heyman, op. cit. p. 146. (f) Grober, Grundriss, II, p. 634, inclines to accept Kupferschmidt for the theory of the lost version, the source of Gaimar, although he does not follow his deriva- tion of the French Lay. Cf. p. 471 : **Ziemlich wahrscheinUch ist, dass schon vor 1 1 50 ein anglofranzosisches Gedicht von dem Geschick des ausgesetzten Konigs- sohns Haveloc erzahlte, der in seiner Erniedrigung eine entthronte Konigstochter, [heiratete] und den Verfolgern zum Trotz Land und Thron zuriickgewinnt.** A reference in a footnote calls attention to Kupferschmidt. Also, p. 473, in speaking of Gaimar' s Estorie y Grober adds: **Erhalten blieb davon die zweite, die Jahre 495-1100 umfassende Halfte . . . begleitet von einer Bearbeitung der Haveloksage . . . nach altrer franzosischer Vorlage gedichtet." (g) Scho- field also follows in part the Kupferschmidt- Putnam theory in his English Litera- ture from the Norman Conquest to Chaucer y London, New York, 1906, p. 267. He says : **In the twelfth century, when the treasure-trove of the Saxons was so extensively appropriated by the Anglo-Normans, were composed, it appears, at least two French poems on Havelok — one, the more primitive, apparently a metrical romance, the other in the likeness of a Breton lay. Only the latter is preserved ; but of the romance we have a summary interpolated in the manuscript of Robert of Brunne's chronicle." (h) Wohlfeil, The Lay of Havelok the Daney Leipzig, 1890, p. 11, does not decide between Skeat's former opinion (ed. 1868) and Kupferschmidt' s. ** Wir werden vielmehr auf jeden Fall ein eng- lisches Original anzunehmen haben, unbeschadet dessen, ob wir mit Skeat der An- sicht sein mogen, dass das franzosische und englische Lay beide unmittelbar aus derselben englischen Quelle geschopft haben, oder mit Kupferschmidt die Uber- zeugung haben, dass das franzosische Lay und Gaimars Bearbeitung der Sage nicht direkt auf dieses englische Original, sondern auf eine verloren gegangene . . . franzosische Romanze zuriickgehen, die ihrerseits erst dem Original entnommen ist." (i) Deutschbein, op. cit. p. 100, hesitates, but decides in favor of Kupfer- schmidt-Putnam. **Der Verlauf der Erzahlung im Lai deckt sich im ganzen mit der Darstellung Gaimars. Man mochte daher auch annehmen, dass der Dichter des Lais Gaimar als direkte Vorlage benutzt habe. . . . Auf der anderen Seite scheint der Lai urspriingliche Elemente der Sage besser bewahrt zu haben — vor allem sind einzelne Szenen starker ausgebildet : ... so halte ich es fiir wahr- scheinlich, dass Gaimar und der Lai aus einer gemeinsamen Quelle, vielleicht einem anglonormannischen Gedicht in kurzen Reimpaaren geschopft haben. INTRODUCTION 1 1 eminent critics. Grober,^' because the Lay — the more recent reworking of the Havelok saga — coincides at times word for word with Gaimar's episode, concludes that the author must have known Gaimar*s account, and that his poem is a free narration of the tale offering only epic commonplaces in addi- tion to Gaimar*s incidents. Suchier,^^ Gollancz,33 and Foulet^^ express the same opinion without giving reasons, and Warnke also inclines to this view.^^ Because these opinions have been given merely as opinions, with no proof of their validity or disproof of the conclusions of Kupferschmidt and Putnam, they have left unshaken this theory of the derivation of the various versions of the Havelok tale, which is generally adhered to in other critical and literary discussions bearing on Havelok. But these opinions, although held by a minority of those who have approached the question, deserve attention. In spite of the fact that much has been already written on Havelok, two considerations seem to justify a new study of the French Lay. If the source of this poem were known, the wie dies Kupferschmidt, Putnam . . . angenommen haben. Die andere Moglichkeit, .dass der Verfasser des Lais von Gaimar ausgegangen sei und mit grosserem Geschick den urspriinglichen Zusammenhang erkannt habe, hake ich fur ausgeschlossen, da er sonst seinem StoiF verstandnislos gegeniibersteht . (j) Zenker, Boeve-Amlethus, Berhn, 1905, p. 92, finds Kupferschmidt' s conclusions plausible, although he mentions Suchier's and Gollancz's diflFering views. 31^ Grober, Grundrissy II, p. 634, speaks of the French Lay as **eine lai sich nennende Neubearbeitung der halbhistorischen Haveloksage . . . die in der eng- lischen Landessprache nach einheimischer Uberlieferung erzahlt wurde, auf der auch die Vorlage von Geffrey Gaimars Dichtung viber Havelok beruhte. Mit Gaimar trifft die Neubearbeitung einigemale im Wortlaut zusammen, so dass ihr Verfasser Gaimar gekannt haben muss. Im iibrigen ist sie freie Nacherzahlung des Stoffes, bietet aber nur "epische Gemeinplatze iiber Gaimar hinaus mehr." 32 Suchier and Birch- Hirschfeld, Geschichte der franzbsischen Literature Leip- sig, 2d. ed. 191 3, p. 124. Suchier says in speaking of the Lay: *'Der Dichter, der am Schluss einen alten (musikalischen) Lai Have he erwahnt, hat gleichwohl die erzahlte Geschichte aus Gaimars Chronik entlehnt." 33 Gollancz, Hamlet in Iceland, London, 1898, Introd. p. xli. *'As regards the Laiy it is almost certainly derived from Gaimar's terser version." 34 L. Foulet, Marie de France et les Lais bretons, Zeitschrift fiir romanische Philologie, 1905, XXIX, 55. 35 Warnke, Die Lais der Marie de France, Bibliotheca Normannica, Halle, 1900, Introd. p. XV. **Das uns erhaltene Gedicht von Havelok selbst mag freilich mit Unrecht ein Lai genannt werden, da es vielleicht von der Darstellung desselben Gegenstandes bei Gaimar abhangig ist." 12 THE OLD FRENCH LAI D'HAVELOC problem of the relationship of the other versions of the story would be simplified for a future investigation such as Vising suggested for the English Lay.^^ Furthermore, as has been stated, definite knowledge respecting the source and composition of such a lay as that of Havelok may contribute something to the subject of the Bretons, their lays, and the Old French nar- rative lays. Before an independent study of the Lai d'Haveloc can be undertaken, however, the articles by Putnam and Kupferschmidt which give all the detail of the previous discussion must be care- fully examined. This examination will form the subject of Chapters II and III. Putnam's study will be considered first, since his results seem to confirm those of Kupferschmidt. 36 Vising, Kritischer Jahresbericht uber die Forts chr it te der rom. Phii., 1905, VII, part 2, 91. **Eine erneute Prufung wird vielleicht mehr Licht bringen." II A Consideration of Putnam's Study/ The h,ambeth Version of Havelok The article by Putnam on the Lambeth Interpolation^ which, as has been seen, is the second important contribution to the development of the Kupferschmidt-Putnam theory, has as its subject the abridgment of the Havelok tale that is found in the Lambeth manuscript^ of the Story of England which Robert Mannyng of Brunne^ (Bourne) completed in 1338. Robert, well known as the author of a didactic treatise full of anecdote called Handlyng Synne, used as his chief sources for the Story of England Wace and Peter of Langtoft ; he reluctantly discarded as unauthenticated the Havelok tale, which he knew from some source. The story, however, was added by a later writer. It consists of 82 lines in rhymed pairs, and the language is, accord- ing to Putnam, such as might have been written at the end of the fourteenth century.^ Putnam notes in the first place that the Lambeth Interpolation has never received the consideration it deserves. Madden,^ he says, attributed it to the scribe, who had made other changes in the manuscript, and considered it an abridgment of the Havelok story apparently copied from Gaimar. Skeat^ repeated Mad- den's opinion, and Kupferschmidt^ also accepted this view with- out further investigation. Putnam next observes that the Interpolation is closer to the French versions than to the English, and that it is more like ^ Edward Kirby Putnam, The Lambeth Version of Havelok^ Publications of the Modern Language Association of America^ 1900, XV, 1-16. 2 The two existing manuscripts of Robert's Story of England are the Lambeth and the Inner Temple. The latter was published by Hearne, 1725. 3 Schofield, English Literature from the Norman Conquest to Chaucer y p. 362. 4 Putnam, p. 2. 5 Madden, ed. for the Roxburghe Club, London, 1828, p. xvii. (Putnam, p. I.) 6 Skeat, ed. 1868, p. xi. 7 M. Kupferschmidt, Die Haveloksage bet Gaimar y etc., Rom. Stud., 1880, IF, 411-430. 14 THE OLD FRENCH LAI D'HAVELOC Gaimar than like the French Lay in the following points: (i) the names agree; (2) the invasion of Denmark is for tribute which is withheld (Lamb. 2-4, Gaimar 410-41 1), but in the Lay it is to demand tribute (27-30); (3) Grim in both poems appears as a mariner, whereas in the Lay he is a baron (Lamb. 14, Gaimar 423, Lay 57); (4) Edulf is defeated in a general battle and not, as in the Lay, in a single combat (Lamb. 70-72, Gaimar 739-742, Lay 940-970); (5) the geographical details agree. In these points Gaimar and the Interpolation are simpler and seem to present the more primitive form of the story and the spirit of the original. Although Putnam's second point is rather a question of emphasis than of vital difference, the other points, especially i and 5, indicate beyond a doubt a very close resem- blance between the Interpolation and Gaimar. But, as Putnam's further discussion shows, differences as well as resemblances exist between the Interpolation and Gaimar, and these very points of difference correspond to traits found in the French Lay. The most striking of the variations between the Interpolation and Gaimar is the order in which events are nar- rated in the opening portions of the two poems. Gaimar tells first of Argentine and England; the Interpolation, however, begins with Havelok and Denmark, and in this corresponds to the French Lay. Other points of resemblance between the In- terpolation and the Lay are as follows: (i) Gunter's enemies plan to shame his family (Lamb. 9, Lay 79-82); (2) in the In- terpolation Edelsi is "of Breton kynde" (Lamb. 31, Lay 200), in the Lay "Bret par lignage" (200); (3) in the Interpolation and in the Lay Havelok and Argentille are married, whereas in Gaimar no formal marriage takes place (Lamb. 47, Lay 377— 380); (4) the statement in the Interpolation that the king brings about the marriage though many are wroth seems to cor- respond to the account given in the Lay of the anger of the barons at the king's violation of his oath (Lamb. 48, Lay 279- 376); (5) the departure of Havelok from Grimsby is described in both the Interpolation and the Lay and antedates Grim's death (Lamb. 27, 56, Lay 157-192, 565), whereas in Gaimar there is no such description and Havelok leaves Grimsby after Grim's death; (6) the fight between Edelsi and Havelok is introduced abruptly by Gaimar (Lamb. 75-76, Lay 1007- 1026), whereas in the Interpolation and in the Lay there is mentioned the gathering of a host by Edelsi. A CONSIDERATION OF PUTNAM'S STUDY 15 Let us investigate each of the points suggested by Putnam. (i) Gunter's enemies plan to shame his family. The Interpo- lation has : " When he was ded they schope brynge al his blod to schame." In the Lay (81 ff.) we read that Hodulf drove Gunter's friends out of the land. Grim and the queen feared he would take the castle and kill Havelok. Grim fled with the boy and the queen for fear of the evil king: (96) "Tost feroit a li deshonur." This shows more fear of Hodulf than any ex- pressed intention on the part of the latter to harm them. This point is, therefore, not of weight. In Gaimar, the queen is afraid and flees. (2) Edelsi is "of Breton kynde." Gaimar has (61) "Li altre rei estait Breton." Both Gaimar's expression and that of the Lay, " Bret par lignage," could be represented in English by "of Breton kynde." ^ This point may, therefore, be set aside. (3) Havelok and Argentille are married in the Interpolation and the Lay; in Gaimar no formal marriage takes place. This statement is erroneous, for, although Gaimar aflirms (174-175) : Honist sa nece, a son espeir, E la dona a son quistrun, yet he shows by two other passages how those lines are to be interpreted (596-598) : E ceste dame iert sa parente. Si cum lui plut la me donad, E ensemble nus espusad. Also, 99-102 : Pur lerite kil coveitat, Sa nece mesmariat. II la donat a un gar9on, Ki Cuheran aveit a nun. (4) The anger of the barons at Edelsi's violation of his oath in the Lay seems to correspond to the statement in the Inter- polation that he brought about the marriage although "were s Cf. Skeat, Eiym. Diet. ed. 1882, p. 315. Kind, nature, sort, character. A New English Dictionary^ I, (i) Birth, origin, descent; Obs. (b) In phrases, through, by, of kind: right of birth, right or position derived from birth. (2) The station, place, or property belonging to one by birth ; one's native place or position. II, A class, group, or division of things. (lob) Used in poetry, with defining word, in the general sense of race ; i 362, Langl. P. PI. A XI, 282: Cristene kynde to kille to dethe. Or kynde can mean ( 1 1 d) descent, genealogy. Cf. also Robert of Brunne's own use of the word in regretting the lack of a source **])at tellis of Haneloc Kynde," ed. Hearne, p. 26. 1 6 THE OLD FRENCH LAI D'HAVELOC manion ful wrothe" (Lambeth 48, Lay 279-376). The ninety- seven lines referred to in the Lay include the entire description of the interviews of the king with the barons and councillors, with his threat and the mention of guards placed in the assem- bly room. After he had stated his purpose, the barons said that they would not permit him to carry out such an intention. The author states : la icust granz coups doncz. Quant il fet venir les armez. There would have been fighting, but soldiers brought in by the king prevented. This resistance seems different from the wrath reported by the author of the Interpolation. One feels that perhaps it is merely the effort of the author of the more courtly poem to protect the reputation of his barons. Both comments could have been made independently. The line in the Inter- polation may possibly have been brought in for the rhyme. He dide hem arraye ful symplely, & wedde togydere bothe. For he ne rewarded desparagyng, were manion ful wrothe. (5) The departure of Havelok from Grimsby. At first sight there seems to be a striking similarity between the Interpolation and the Lay in this respect, but, if the facts in the two cases are compared, the parallelism is less patent. In the Lay Grim sends Havelok to court "pour aprendre sens et avoir querre,"^ where- as in the Interpolation Havelok goes to court because he is a "man of mykel cost," that has to serve "for his grete susti- nance." '° The spirit of the Lay and the reasons for Havelok's departure are totally different from those of the Interpolation; the resemblance is merely in the time of the episode with respect to Grimes death. (6) The fight between Edelsi and Havelok is introduced abruptly by Gaimar. In the Interpolation and in the Lay the gathering of a host by Edelsi is mentioned." This fact seems a most casual detail in the description of a battle, and one that, unless there is an unexpected attack, can be omitted or expressed at will. Gaimar certainly implies that an army had been assem- 9 Lay, 174. 10 Int., 25-26. " Putnam's reference to the description in the Lay includes 1 007-1 026. 1007— loi 8 report the arrival of Havelok's messengers at court and the mockery with which they were greeted. loi 8-1022 describe their return. The assem- bly of the army on the appointed day at Theford is told in five lines. A CONSIDERATION OF PUTNAM'S STUDY 17 bled in (769) "Mult i out homes afolez" (in D and L asem- blez). This line follows the description of the battle. In 761 he had already stated that Havelok "Od sa grant ost la mer passa/* Hence, the fact that the Lay and the Interpolation both mention Edelsi's gathering of a host cannot have weight as an argument to prove the similarity of the two versions. Putnam himself did not attach equal importance to all his arguments, for he adds:'" "It is difficult to imagine that all these resemblances are accidental. The first two and the last might be so, but the others seem to point to details in a source common to both the Interpolation and the Lay. This common source cannot be Gaimar, because in these points Gaimar diffisrs. Moreover, in all three points Gaimar, rather than the other versions, seems to show a change from what must have been the original form of the story." Point 3, that of the marriage, which is one of the most im- portant points, as we have seen, will not hold. Points i, 2, and 6 are vague possible resemblances inadequate for proof, and 4 may be quite as casual. The only point showing definite re- semblance is 5, and even here the likeness consists merely in the time of the departure of Havelok from Grimsby. The cir- cumstances and the tone of the accounts are wholly different. Putnam's subsequent statement is, therefore, unwarranted. He says : " The number of important details common to the Interpolation and the Lay and the exact agreement in the order of the narrative establish a close relationship between the two and a common source independent of Gaimar." But since we have not found a number of important details common to the Interpolation and the Lay that confirm the relationship of the versions which the similarity of the order of events narrated in the two poems seems to indicate, an investigation of this simi- larity becomes necessary. Could anything other than a source in common with the French Lay explain the Danish beginning of the Interpolation? Robert of Brunne in the second part of his chronicle trans- lated Peter of Langtoft.'^ Peter, after telling of RoUo and Alfred and the departure of the former for Normandy, continued : '^ " Putnam, op, cit. p. 8. ^3 Peter mentioned Grim and Grimsby in the earlier part of the chronicle **A val de Grymesby, ke Grym jadis fesayt." Ed. Wright, Chronicles and Memorials of Great Britain, XL, part I, 24. ^4 Ibid., p. 316. 1 8 THE OLD FRENCH LAI D^HAVELOC Alfrede of ses Englays demort en ses regnez. Taunt cum vers le north Alfred est alez, Gountere,^5 le pere Havelok, de Danays ray clamez. Of grant chuvalerye est Engleterre entrez, Destrut ad les viles et arses les cytez. Les Norays s*atyrent le ray les ad maundez. Of le ray Gountere en chaump sunt encountrez. Les clers saint Chuthbert sun cors unt levez Hors del sepulture ou fust entoumbez, VII. aunz parmy la terre le corsaint fii portez, Dunt le rays Alfrede fu sovent confortez. Kant Alfred et Gountere ensemble ount guerreez, Gountere par aide de Deu chaunge voluntez. Le rays Alfred pur veirs de founte li ad levez, XXX. des melx vayllaunz of ly sunt baptizez, Retournent Cristyens ke vindrent renaez. Robert Mannyng'^ followed Peter closely in telling of Rollo and Alfred and of Rollers passing over into Normandy, leaving Alfred "stille here in Inglond/* Then follow the lines: ^it a noJ>er Danes kyng in ]>e North gan aryve. Alfrid it herd, ))idere gan he dryue. Hanelok fader he was, Gunter was his name. He brent citees & tounes, ouer alle did he schame. The incident of St. Cuthbert's clerks is also told. Then, when Gunter and Alfred had waged war a long time, through God's grace Gunter was converted. Cristend wild he be, })e kyng of fonte him lift, E ]>ritty of his knyghtes turnes, ))orgh Gode's gift. This mention of Hanelok (Havelok) makes Robert desire to tell the story of Havelok, but he writes instead the interestingly naive passage (which Skeat has quoted in his Preface '^) in which he regrets his inability to find a trustworthy account of Havelok. He begins: Bot I haf grete ferly, that I fynd no man. That has written in story, how Hauelok this lond wan. Noither Gildas, no Bede, no Henry of Huntynton, No William of Malmesbiri, ne Pers of Bridlynton, Writes not in ]>eir bokes of no kyng Athelwold, etc. ^s Peter, as has been often remarked, confused this Gunter with Gunter the Danish invader defeated by Alfred the Great, who in the Anglo-Saxon chronicle is called Godrum. Cf Skeat, ed. 1902, Pref. xliii. '6 Peter Langtoft's Chronicle (as illustrated and improved by Thomas Hearne), Oxford, 1725, vol. I, p. 25. ^7 Ed. 1868, p. X. A CONSIDERATION OF PUTNAM'S STUDY He says : Thei mak no menyng whan, no in what date. Bot that thise lowed men upon Inglish tellis, and ends with the lines : Sen I fynd non redy, that tellis of Hauelok kynde, Turne we to |)at story that we written fynde. Although Robert sought in vain an authentic account of the '' rhyme," some one else was more successful or less critical and interpolated an outline of the Havelok story in the Lambeth Ms. of the Story of England. From the sense of the passage preceding the interpolation, it is clear that, however his source read, the author of the abridgment was forced to begin his ac- count with Denmark and the Danish invader of England, who had just been named. It is solely because of the mention of Havelok and Gunter that the story is introduced, and to have begun with Edelsi and Ekenbright would have introduced events entirely foreign to the material of the chronicle at that point and emphasized the discrepancy in the time of the two episodes, that of Gurmond and Alfred, and that of the two English kings, the contemporaries of Constantine, the nephew of Arthur. Moreover, a change from a source such as Gaimar would not have been hard to make, nor would it have demanded great skill on the part of the interpolator, who was in this case, as the condensed clear style of the passage shows, not unintelli- gent.'^ The events told by Gaimar first concerned England and Argentine up to the time of Havelok^s arrival at Edelsi's court. Then after the marriage, dream, and return to Grimsby, Havelok's story is told by Kelloc. In the Interpolation the facts of Havelok's life are not told by Kelloc, but are transferred to the beginning of the narrative and precede the account of Argentine. The order of events in each division remains un- changed. One modification was certainly made by the author. Arthur in connection with this part of the story of England was most incongruous, and he was replaced by an indefinite " Breton kyng " who went to Denmark to collect the tribute which " Arthur whylom nam." The change in the order of events in ^8 Cf. Putnam, op. cit. p. 2. ** The style is marked by extreme condensa- tion, an entire incident often being told in a single line. As a result the story appears in a surprisingly complete form." zo THE OLD FRENCH LAI D'HAVELOC the Interpolation may therefore be easily explained, since the conditions of the chronicle furnish an obvious reason for it, and the argument based on the difference in this order of narration needs no longer to stand in the way of the hypothesis that Gaimar was used as the source of the Interpolation. Putnam claims, however, that a comparison of the Interpo- lation with the English Lay confirms this conclusion of his, which we have just found unconvincing, that the Interpolation, the French Lay, and Gaimar all emanated from a common source. This source he had already described, as follows : '^ " Though it is now lost there must have existed at some time a French version of the romance distinct from Gaimar and the Lay. That such a version did exist and was the common source of both Gaimar and the Lay has been effectively proved by Kupferschmidt.'* In a note he reiterates: " Kupferschmidt*s investigation, already referred to, must be regarded as settling the fact that Gaimar and the Lay had a common source written in French octosyllabic rimed couplets." This lost French ver- sion, he says, was also the source of the Interpolation. A com- parison of the Interpolation with the English Lay reveals: (a) traits which, since they exist only in the English version and in the Interpolation, must have been in the source of the English poem and of the lost French version from which the Interpola- tion came; (b) facts which, since they exist in the Interpolation, the English Lay, and in one of the French versions but not in the other, must have been in the source of the E.nglish Lay and of the two French versions, Gaimar and the Lay. Under (a) Putnam enumerates the following points :''°(i) Have- lok's reason for leaving Grimsby (Lamb. 25-26, E. Lay 824- 852); (2) the love which the folk at court feel for Havelok (Lamb. 44, E. Lay 955-958); (3) Edelsi hears that Havelok has come to the coast (Lamb. 75), and in E. Lay Godrich hears that Havelok has come to England (2531-2547). Although points 2 and 3 do not add anything really new to the accounts of the French versions, since Gaimar tells how gladly all gave to Havelok, since the author of the Lay reflects the same idea in the advice Grim gives Havelok to make himself loved, and ^JjL^.t r since in both versions Edelsi must have heard of Havelok's ^^\iy (approach, even if no special mention is made of it, yet point i, ^^%^^-^ '^Uid., p. 9. 2° I is the important point; 2 and 3 are mentioned, p. 12, n. i. r>'' A CONSIDERATION OF PUTNAM'S STUDY 21 on the other hand, is an important connecting link between the two versions. Under (b), traits which exist in the Interpolation, in the English Lay, and in only one of the French versions, he names of those found in Gaimar the following: (1) the humble posi- tion of Grim ; (2) the defeat of the usurper in general battle. Of those found in the Lay : (i) the marriage of Havelok; (2) Havelok finds Grim dead upon his return to Grimsby; (3) the calling out of the host. The points which are common to Gaimar are definite, but only one is found in the Lay, namely point i, for point 2 is common to all the versions and not opposed to Gaimar^' as Putnam supposed, and 3 is a casual mention of a fact in accordance with Gaimar's description, although not es- pecially mentioned by him. But 2 practically resolves itself into the difference in the time of leaving Grimsby, because in Gaimar Havelok left Grimsby after Grim's death, and in the other versions during his lifetime. We now see that the real point of contact between the In- terpolation, the English Lay, and the French Lay, this differ- ence in the time of leaving Grimsby, is also the unexplained point of similarity between the Interpolation and the French Lay,^^ as opposed to Gaimar. The circumstances of the depart- ure, however, are quite different in these last named versions. The significance of the coincidence in the time of leaving Grims- by which we find in the Interpolation and in the English and French Lays is vastly increased by the fact that the conditions of the departure are exactly the same in the Interpolation and the English Lay. It is interesting to note that the point which Heyman''^ mentions, as common to the English Lay and the Interpola- tion, is Havelok's appetite, and the only allusion to it in the Interpolation is the line in question where the reason for leaving Grimsby is given. Must one believe with Putnam that the correspondence of details indicates a common source for the Interpolation and the French Lay ? May not such details, on the other hand, be interpreted as signs of the influence, direct or indirect, of the English Lay P 21 Cf. this study p. 15. 22 Ibid.y p. 17. 23 Heyman, op. cit. p. 145. 22 THE OLD FRENCH LAI D'HAVELOC Putnam rejected as highly improbable the idea that the Lam- beth Interpolation could have resulted from a combination of two or more versions. "Such a combination," he said, "would of course explain anything except itself. An interpolating scribe, for the sake of inserting into a chronicle an episode of less than a hundred lines, is not likely to take the trouble to compare varying versions of a romance, perhaps in more than one lan- guage, and to make out of them a consistent whole. . . . The Lambeth Interpolation tells a straightforward, consistent story, and any lack of clearness is due to nothing more than the ex- treme condensation." ^^ And again he said concerning the epi- sode of Havelok's departure from Grimsby: "As it is extremely improbable . . . that this agreement with the English romance is accidental, it seems to be clear that this must have been an element of the story in an early form."^^ But in opposition to Putnam the following points may be urged : 1. We are not considering, in the case of the Interpolation, the skillful fusion of two sources, since the basis of the account corresponds to Gaimar's story, and since the material taken from Gaimar may have been simply retouched by an author who knew the English Lay, or its source, or some closely related work. The interpolator, as we have seen, probably modified one detail independently when he substituted for Arthur an inde- finite Breton king. He omitted all supernatural incidents, and introduced one statement which was very likely his own inven- tion, since it has been found nowhere else, i.e., that Argentille was given to Havelok because of a "chere" she made him (44). Is it impossible that, under the direct or indirect influence of the English romance, he modified other details ? 2. Robert, according to Skeat, was familiar with the English romance, for he quoted it or imitated it in his Handlynge Synne.''^ Skeat thinks Robert's allusion to the "ryme" that "lowed rrien upon Inglish tellis " refers to the romance we know. However that may be, it is not unlikely that the interpolator knew such a work as Robert mentions, and that Robert's words may have brought it to his mind and may have been the cause of his in- troducing into his account one or two points taken from it. 24 Putnam, op. cit. p. 8. =5 Ibid.y p. 12. =6 Skeat, ed. 1902, pp. xliii-xliv. A CONSIDERATION OF PUTNAM'S STUDY 23 3. Perhaps the interpolator himself was not responsible for the combination of different sources. He may have taken his material from some French or English chronicle in which the fusion was already made. Such combinations are found among the minor versions of the Havelok story already referred to,^^ as for example in the Brut, which was studied by Brie, who found an earlier manuscript than that mentioned by Skeat in his introduction to Havelok. ^^ From his investigations Brie knew that Gaimar had served as basis of the account of Havelok in the Brut, but he found that several points in the version correspond to the English romance and differ from Gaimar.^^ Some other work closely related to the English Lay ,3° if not the Lay itself in early form, is therefore reflected in this passage for which Gaimar furnished the greater part of the details. Other manuscripts of the Brut contain the story at greater length and with still more variety of incident. From the Brut the story passed into other chronicles. May it not be that some com- bination in a chronicle which was based chiefly on Gaimar, and which has not come down to us, has served the interpolator as a source .f* Considering once more the Interpolation and Gaimar, we note that the principal events of the story of the Interpolation are all contained in Gaimar, with the exception of the episode of the departure from Grimsby. ^^ The few points that the Interpola- tion has in common with the French Lay alone may be due to 27 See this study p. 6. 28 Skeat, ed. 1868, pp. xiii-xiv. 29 Brie, Eng. Stud. 1905, XXXV, 359-371; and Heyman, op. cit. pp. 146 and 147. 3° I ought to call attention to the fact that the chronicles referred to may not reflect merely the versions which have been considered. There may have been various literary works derived from both Gaimar and the French Lay. 31 Putnam thinks the Interpolation and the English Lay preserve primitive traits in this episode, since it is natural to suppose there was a reason for leaving Grimsby. Grim's death furnished a reason' of the same kind, he being regarded as Havelok's means of support — the provider of the family. I think it less Hkely that Gaimar changed. Gaimar in other cases has primitive traits, and here he has no reason for leaving out a famine and substituting Grim's death. The other writers change and modify statements, as we know. Unless bound to accept the common source theory, I am inclined to regard Gaimar' s as the more primitive account. Conden- sation could not explain Gaimar' s changes. His account is over eight hundred lines in length and the French Lay only eleven hundred, and Gaimar frequently enters into long and unimportant descriptions. 24 THE OLD FRENCH LAI D'HAVELOC chance, and in any case they are vague and general. The Inter- polation also agrees with Gaimar's account in many definite details, as Putnam showed, especially in the geography and proper names. All the names of characters mentioned in Gaimar with the exception of Kelloc occur in the Interpolation. (Eleyne is really the Alvive of Gaimar, as is shown by Ms. D of Gaimar, Alleve, and Ms. Z>, Alleue. The letter u has evidently been read n.) Inasmuch as the evidence derived from the Interpolation seems to point more definitely to Gaimar than to the other versions, and since the differences may be explained as due directly to the influence of the English poem, is it not going too far to see in Putnam's explanation of the Interpolation a confirmation of Kupferschmidt's theory of the lost common source of the French Lay and Gaimar P If such a source existed for these works, the Interpolation may have been derived from it, but there is not evidence enough in the Interpolation to es- tablish such a source. Furthermore it seems that instead of corroborating Kupfer- schmidt's theory, Putnam's solution of the problem of the source of the Interpolation depends on the validity of Kupfer- schmidt's claims. The Interpolation may be explained as deriv- ing from the lost common source of Gaimar and the French Lay, or from Gaimar modified by a few suggestions from the English Lay or a work closely related to it, or from some chronicle in which Gaimar's narrative was already fused with material later than Gaimar. Before we decide to follow Putnam in adopting the first of these hypotheses, the theory of the lost common source, it is expedient to turn to the study in which Kupfer- schmidt arrives at the conclusion that the French Lay and Gaimar emanated from a lost common source. Ill An Investigation of Kupferschmidt's Study, The Haveloksaga in Gaimar and the relation between this version and the h,ai d' Haveloc^ Two questions are discussed by Kupferschmidt in his study, Die Haveloksage bei Gaimar: (i) Was the Havelok episode in Gaimar's Estorie des Engles written by him, or is it an interpo- lation ? (2) What is the relationship between the two French versions of the story of Havelok? The first of these questions is of great importance, since the chronological order of the sev- eral versions of Havelok would be affected should this Havelok episode of the Estorie be an interpolation and possibly of much later date than Gaimar. Kupferschmidt decided that the episode of Havelok in Gaimar's Estorie was written by him. Vising^ has discussed this part of Kupferschmidt's article ; and although he corrects the writer's derivation of Gaimar's manuscripts, and points out inaccuracies, he concludes: "J'admets que la con- clusion ou s'arrete M. Kupferschmidt est la plus probable, lors meme que je ne peux admettre que son raisonnement soit toujours bon ou correct." ^ This recognition of the Have- lok episode as Gaimar's establishes with certainty its date, since Gaimar's Estorie des Engles was written between 1147 and 1151.^ Before considering the second question, it will be necessary to 1 Rom. Stud'.y IV, 410-430. 4!>0 2 Vising, Etude sur le dialecte anglo-normand du XII Steele y Upsala, 1882, pp. 26-34. 3 Cf. ibid.y p. 29. Vising states again (p. 30) that there are slight differences of language and versification between the episode of Havelok and other parts of Gaimar, but that these are attributable to the different sources used, and adds, ** Mais en general ces deux parties se ressemblent etroitement. Seulement il ne faut pas baser une telle conclusion sur les faits rapportes par M. Kupferschmidt, car il a pris ses citations sans aucune critique." 4 Gaston Paris (^Litterature fran^aise au Moyen Age, 3d ed., Paris, 1905, p. 273) places Gaimar's Estorie about 1148, and (p. 145) he gives 1 147— 1 151 as the time limits for its composition. 26 THE OLD FRENCH LAI D'HAVELOC determine whether the Estorie preceded or followed the French Lai d' Haveloc. Kupferschmidt, who studied the rhymes of the latter poem, concluded that the rhyming of ie : e and oit (Lat. -abat): eit (Lat. -ebat) showed that the Lay could not be earlier than the thirteenth century. ^ This result is of great importance since it shows: (i) that the Lai d'Haveloc is not the oldest ex- isting version of the Havelok saga as Madden and Skeat^ had supposed it to be; and (2) that, since the Lay is more recent than Gaimar, it could not have been abridged by the latter in his Estorie^ as Madden, Skeat,^ and Storm ^ had suggested. Recent- ly Deutschbein,^ who had at his disposal the variants of the older manuscript of the Lay (Ms. P),'° found that the rhymes ie: e^ and out\= oit] : eit in the afore-mentioned verb forms of the imperfect, were not used in that manuscript and were therefore not characteristic of the Lay itself, but that they were evidently changes made by a later scribe and found in Ms. H, the basis of the text which Kupferschmidt studied. Deutschbein then corrected the latter's date for the Lay, and placed it in the second half of the 12th century." This, however, does not affect the results obtained by Kupferschmidt, inasmuch as the Lay is still more recent than the text of Gaimar, who conse- quently could not have abridged it. After establishing the fact that Gaimar*s Estorie preceded the Lay, Kupferschmidt cited from this poem thirty-four lines, which correspond almost word for word with lines of Gaimar's Havelok, and which show that some connection must have ex- isted between the two poems. Since Kupferschmidt had settled in the negative the question of Gaimar's having abridged the Lay, there remained only two possible suppositions as to the 5 Cf. Kupferschmidt, op. eit. p. 424. **Eine Untersuchung der Reime des lai d' Havelok jedoch zeigt uns, dass dieses unmoglich vor dem Anfang des XIII Jahrhunderts entstanden sein kann.'* 6 Skeat, ed. 1868, §4. **This version was certainly composed in the first half of the 12th century." 7 Skeat, ed. 1868, §5. 8 Storm, Eng. Stud., 1880, III, 533. 9 Deutschbein, Studien xur Sagefigeschichte Englandsy pp. 97-98. ^° Ms. Phillipps, Cheltenham, the variants of which are given in the Hardy- Martin edition of Gaimar, I, 320-327. " P. 98. **Mithin gehort der Lai d*Aveloc ciner wesentlich alteren Zeit an, und wird wohl in die zweite Halfte des 12. Jhs. zu setzen sein. Suchier erklart den Lais d'Aveloc fur einen der altesten Lais." KUPFERSCHMIDT'S STUDY 27 relationship of the two versions : either that the Lay came from Gaimar, or that both emanated from a common source.'^ Before Kupferschmidt only one person, Petrie/^ had sug- gested a lost version as a source of the two French poems. Petrie observed that, although the two versions have the same story in substance and contain lines exactly alike, at times each has circumstances not found in the other which are necessary to the story. He speaks of the visit to the hermit as an episode~~^ not found in Gaimar's account, and as one which is needed to r explain Argentine's dream. In the latter version, on the other ' hand, there is the description of Sygar's ring, and the record of a feast, both omitted by the author of the Lay. Kupferschmidt "^ criticizes Petrie's arguments, saying that these episodes are by no means essential, and that all that one may conclude from them is that the author of the Lay had used Gaimar. Nothing Petrie said, Kupferschmidt decides, can be taken as proof that there was a romance which served as the source of the Lay and of Gaimar's version. To give such proof Petrie would have had to show that incidents found in each of the French works occurred also in a third work which was totally independent of them both. Then we should know that details found in one of the French poems and in such a work were drawn from the source of all three works, and that they were not the invention of one author. * Until such proof is offered, he says, it is idle to speak of a common source. Having dismissed Petrie's arguments as worthless, Kupfer- schmidt goes on to say that although Petrie did not prove his assertion it is nevertheless possible to do so. An English ver- sion of the story is in existence which developed independently, and we may assume that details belonging to this version and one of the French poems were in the original source of all three.'s For his assertion that the English version developed inde- pendently, and was not influenced by the French versions, Kupferschmidt refers to Skeat, §20.'^ Skeat, it will be remem- ^2 Kupferschmidt, p. 427. ^3 Petrie, Mon. Hist. Brit., London, 1848, I, 765, note b, reprinted by Skeat, ed. 1868, p. xxvii. ^4 Op. cit. pp. 427-429. ^5 Op, cit. p. 429. ^6 Skeat, ed. 1868, §20. 28 THE OLD FRENCH LAI D^HAVELOC bered, made this statement in his edition of 1868. His com- parison of the three works concerned is external and general. At that epoch also much less was known about the general de- pendence of English romances on French originals. Kupfer- schmidt's entire proof therefore rests on the authority of Skeat, who might possibly have modified his opinion in his later edition had it not been for Kupferschmidt's own conclusions which he (Skeat) accepted and quoted, together with the results of Put- nam's study. According to Skeat, in his edition of 1902, the English Lay is derived from an original English source through several stages of development which he represents by Ei, E2, E3, E. This last form, E, the extant poem, cannot have been derived directly from X, the original source of all versions, since there must have been a version Ei probably in the Lincoln- shire dialect, and "if so, belonging rather to the earlier than to the later part of the thirteenth century." Then there must have been E2, "almost certainly written in the South of England by a Norman scribe"; and E3, "which may likewise have been a copy by a Norman scribe, but written in Lincolnshire and add- ing a few local interpolations to bring it up to date, perhaps as late as 1301. Of this E seems to have been an unintelligent copy, made not many years afterwards by a somewhat careless scribe who tried to copy what he had before him." '^ Skeat's solution of the relationship of the versions is, therefore, the following : {r Gaimar Y } Lay of Havelok (^ Interpolation Ei Ez E3 E But if E has passed from the original form through these stages, how can it be asserted that the various forms of the story have never been affected by influence from the French Lay or Gaimar.? Skeat says nothing of influence from these versions; but to prove that the English Lay in the form in which we have it was entirely independent of the French versions, it is necessary that it should have had an independent origin and also that it should have remained free throughout its development. Critics have varied in their opinions regarding the general resemblances of the versions. Skeat, as we have seen, consid- *7 Skeat, ed. 1902, Introd. p. xlviii. KUPFERSCHMIDT'S STUDY 29 ered the English Lay quite different from the French Lay, but Storm '^ observed that the English Lay corresponds to the French Lay in general, although some details and names are different. Kittredge,'^ who quoted Storm, disagreed with him, saying that the English Lay, which was three times as long as the French, was different and that it was really a "gest'* and not a lay. Critics are not more unanimous as to this French influence. Some claim a definitely English origin for Havelok, and make no mention of French influence, whereas others be- lieve that the source was English but acknowledge French in- fluence. Another class of critics refers the poem to a French source. This French source is in its turn considered by some to be the French Lay, and by others to be the hypothetical lost version proposed by Kupferschmidt. Many, however, simply say that the source was French without attempting to determine it. Ten Brink,=° Korting,^^ Wohlfeil,^^ Kupferschmidt,^^ p^t- nam,^^ Billings,^^ Heyman,^^ Brie,=^7 Zenker,^^ and others con- sider the source English. Heyman perhaps gives his reasons most definitely. " It has been a common presumption with scholars that there is no original English romance before Chaucer's time, and it is a fact that a very great number of the Middle English romances are translated from French versions (which are often based, in their turn, on Anglo-Saxon originals). In many cases this can be proved by the existence of French versions which were composed ^8 Storm, Eng. Stud., 1880, III, 533. ^9Kittredge, Amer. Journ. of Phil., 1886, VII, 184, n, 3. 20 Ten Brink, History of English Literature, to Wyclif translated by Kennedy, New York, 1889, pp. 150, 232. 21 Korting, Grundriss der Geschichte der englischen Literatur, 1 9 1 o, p. 1 1 o, §8 8 . 22 Wohlfeil, The Lay of Havelok the Dane, Leipsic, 1890, p. 1 1 . =^3 Kupferschmidt, op. cit. p. 429. 24 Putnam, op. cit. pp. 15—16. 25 Billings, A Guide to the Middle English Metrical Romances, New York, 1901, pp. 21-22. Kupferschmidt' s theory is given, and reference made also to Brandl and Wiilker for suggested French influence. 26 Heyman, Studies on the Havelok -Tale, Upsala, 1903, p. 147. 27 Brie, Eng. Stud., XXXV, 364. Brie proves that an English Havelok poem must have existed before the extant one, and adds : ** Dies scheint mir auch einen schritt weiter zur feststellung eines englischen, nicht anglonormannischen originals der Havelok-dichtung zu bedeuten, fiir das ich mit Heyman (p. 148) gegeniiber den meisten forschern eintreten mochte." 28 Zenker, Boeve Amlethus, Berlin, 1905, pp. 64, 92. 30 THE OLD FRENCH LAI D'HAVELOC previous to the preserved English ones, and by means of other evidence. But a similar presumption lacks proof with regard to Havelok, and there is no reason, but for the general want of early English originals, to assume that the original version was in French. Several reasons may, on the contrary, be adduced in favour of the thorough English or Anglo-Danish origin of the lay. First of all the tale itself is based on traditions which were current in England before the arrival of the Normans. It is hard to say why a poem should not have been actually written already towards the middle of the nth century. The number of different facts stated in the English poem is so great that we are forced to assume that the majority were put in writing at a comparatively early period." Heyman then refers again to several passages which occur in the English romance of Havelok and which occur in only one of the three other versions (the Lay, the Interpolation, and Gaimar), and he claims that the corresponding passages prove the English version to be closer to the original tale. As additional evidence of the English origin of the poem, he cites the English, Scandinavian, and Welsh names of the characters. His conclusion is: "In short, as all this supports the probability that the English romance was not translated from a presumed French version, and owing to the fact that there is no convincing evidence to prove the contrary, it seems tolerably certain that the Lay of Havelok the Dane is an original, English romance." ^^ Vising,3° in his notice of Heyman, comments on this passage as follows: "Das bemerkenswerteste an diesem Resultat ist die Unabhangigkeit der englischen Romanze von der franzosischen Version. So hatte schon Kupferschmidt geurteilt, und ihm folgt Dr. Anna Hunt Billings. Indes sind die Griinde hierfiir schwach und es fragt sich, ob nicht die fur die entgegengesetzte Auffas- sung ins Feld gefuhrten allgemeinen Betrachtungen, die Hey- man zwar seinerseits entkraften will (S. 147), starker sind. Man hat kaum ein sicheres Beispiel von einer romantischen Sage, die als alt- oder mittelenglisches Original vorhanden ware. Auf der andern Seite stellen sich eine betrachtliche Zahl englischer Romanzen oder Romane als Bearbeitungen nach franzosischen (anglonormannischen) Vorlagen aus. Schofield wagt sogar fol- *9 Heyman, op, cit. p. 148. y* Kritischer Jahresbericht uber die Fortschritte der rom. Phil.y 190 2- 190 3, VII, 2, 90-91. KUPFERSCHMIDT'S STUDY 31 gende Behauptung: 'The stones of the English heroes Waldef (Walpeof), Havelok, Guy of Warwick, Beves of Hampton, Hereward, Fulk Fitz Warren, and others, were, it is well known, recorded in French. There is, indeed, as has already been said, no single instance where purely native transmission of an English romance is demonstrable.' ^'^i For King Horn, in the form in which it has come down to us, an English source had been claimed, but Morsbach^^ and Scho- field33 have definitely shown that it is of French origin. Ward expressed the view that the source was English but that French influence was also to be taken into account.^^ The English Lay, he thought, was based on a popular develop- ment of the legend, "though its author was apparently acquainted with the French Lay." Gollancz,^^ in speaking of the English sagas, said of Havelok: "This English romance is independent of the French version, though the author was evidently ac- quainted with the Anglo-Norman poem." Besides Vising and Schofield,^^ who have already been quoted, Wulker,37 Morsbach,^^ Brandl,^^ Holthausen,^" Deutschbein,'^^ 31 Schofield, The Story of Horn and Rimenhildy The Publications of the Modern Language Association of America y 1903, XVIII, 53. 32 Morsbach, Die angebliche Originalitat des friihmittelenglischen **King Horn." Beitrdge zur rom. und eng. Phil. (Festschrift fiir Foerster), Halle, 1902, pp. 297—323. 33 Op. cit. p. 53. **I have given over unwillingly the view I have long had on this point, but it seems to me now impossible to maintain it with good reasons, and cogent arguments are distinctly opposed." 34 Ward, Catalogue of Romances y 1883, I, 440. 35 Gollancz, Hamlet in Icelandy London, 1898, p. xh. 36 Schofield, English Literature from the Norman Conquest to Chancery New York, 1906, p. 268. Of the English Lay he says: **Its relation to the other accounts so far discussed is still obscure. . . . Evidently the English poem stands by itself; but it is not safe to assume that it, rather than the French poems, repre- sents best the original narrative either in substance or in style. Like Horn Childy it is probably a late redaction of early French material, into which new names, new incidents, and new sentiments are introduced.". 37 Wiilker, Geschichte der eng. Lit.y 1900, p. 98. 38 Morsbach, Die angebliche Originalitat des friihmittelenglischen **King Horn," p. 300. 39Brandl, Grundriss der Germ. Phil.y 1893, II, Abt. i, §52. (See Billings.) 40 Holthausen, ed. oi Havelok y London, 1901, Introd., p. ix. **The English poem is probably a translation of a French one, which is however lost." 41 Deutschbein, Studien zur Sagengeschichte Etiglandsy Cothen, 1906, p. 159. ** Unser Ergebnis hat uns dahingefuhrt, dass der englische Spielmann kaum altere 32 THE OLD FRENCH LAI D^HAVELOC and Suchier'^'' believe that the source may have been French. Of these, Wiilker, Morsbach, Schofield, and Holthausen mere- ly mention a French original. Deutschbein hesitates between the French Lay and Kupferschmidt's hypothetical version. Suchier says : " Der englische Bankelsanger . . . diirfte, un- mittelbar oder mittelbar, auf dem franzosischen Dichter^^ fussen." Gollancz and Ward also have in mind the influence of the French Lay, and not that of an earlier poem. It would seem, then, from these varying opinions that it is impossible to speak of the independence of the English Lay as if it were a definitely established fact, and that an exami- nation of the versions of Gaimar, the French Lay, and the English Lay is absolutely necessary as the next step in an in- vestigation of the source of any one of the three works. If passages clearly resembling Gaimar, and others indubitably resembling the French Lay, exist side by side in the poem, we may agree with Kupferschmidt's belief in the independence of the English Lay. For the detailed study of the three versions I shall place the passages selected from the English Lay opposite the corre- sponding passages of the French poems. If the lines are found only in Gaimar they will be printed in black type ; and if they occur in the Lay only, they will be indicated by an asterisk. G, F, and £, will be used to denote Gaimar's version, the French, and English Lays, respectively. Where the resem- blance is especially close, the lines will be italicized. Occasion- ally lines from the French poems which present interesting differences from £, or which are only remotely connected with £, are given in parentheses. (miindliche) Quellen fiir seinen Stoff benutzt hat ; sondern wie seine ubrigen Landsleute wird er wohl eine franzosische Vorlage gehabt haben. Diese Vorlage konnte nun unser afr. Lai gewesen sein, zu dem das Gedicht in mehreren Punkten weit besser als zu Gaimar passt. Wenn man aber Wert darauf legen will, dass in einem bemerkenswerten Punkte die englische Havelokversion mit der cymrischen Meriaducsage iibereinstimmt, so muss man unser englisches Gedicht (direkt oder durch mehrere Zwischenglieder) auf jenes verlorengegangene franz. Gedicht (bez. dessen Vorlage) zuriickfiihren, das schon Gaimar und dem Verfasser des Lais als Quelle gedient hat. 42 Suchier und Birch-Hirschfeld, Geschichte der franz. tit.y 2d ed., 1913, p. 124. 43 The **Dichter" just mentioned was the author of the French Lay. KUPFERSCHMIDT'S STUDY 33 ENGLISH LAY FRENCH LAY— GAIMAR 13 The English Lay begins with Eng- land and Athelwoid. I I o Of his bodi ne hauede he eyr I I I Bute a may den swi^efayr, 112 pat was so yung ))at sho ne cou|)e Gon on fote, ne speke with mouj^e. \)an him tok an iuel strongy I 5 ])at he zuel wiste, and under-fond, 1 1 6 \)at his deth was comen him on : 44 (The king complains of leaving Argentine, sends for his earls, and, on their arrival, asks them to choose her guardian. They elect Godrich of Corn wayle. The king, with "messebok, caliz, 45 pateyn, corporaus, messe-gere" makes Godrich swear to keep his oath. ) 189 per-on he garte \e erl suere, 190 \>at he sholde yemen hire wel, 191 With-uten lac, with-uten tel, 1 92 Till \at she were tuelf winter old, 193 And of speche were bold ; 194 And )>at she cou]>e of curteysye 195 [Don,] and speken of luue-drurye ; 1 96 And til pat she louen mouhte 197 Hwom so hire to gode thouhte; 198 And j>at he shulde hire yeue 199 pe besteyfayreste, the strangest ok: 201 pat dede he him sweren on ])e bok. And \anne shulde he Engelond A I bitechen in- to hire hond. (Athelwoid) 230 To lesu Crist bigan to calle, 2 3 I And deyede biforn his hey men alle. 202 203 Gaimar's account begins with Eng- land and Argentine. Adelbright and Orewein 208 ^Mes entre eus neurent enfant 209 *Af^j qe vne file bele ; 210 (Argentine out non la pucele). 21 1 ^Rois Ekenbright fut enfermeZy 212 ^Et de grant mal forment greuez, 2 1 3 ^Bien siet nen poet garrir ; *Alsi fet a lui venir, *Sa fille li ad comandee 2 1 6 *Et sa terre tote liueree. 217 ^Primerement li fet iurer, 2 1 8 ^Veiant sa gent, et affier, 219 *^ leaument la nurrireit, 220 *Et sa terre lui gardereit, 221 * Tant qele fust de tie I age 214 215 222 *j^erinne dede hire fede 323 Pourelike in feble wede.*' 47 328 Of Goldboru shul we nou laten, (Birkabeyn had three children, **Deth him tok ))an he best wolde liuen,*':48 Here Athel wold's feeling and actions are repeated by Birkabeyn, who sends for priests to shrive him and knights to select a guar- dian for the children. Godard is chosen and swears on the **messebok" to care for them and to rule Denmark until the son should become a knight when he should give him the country. ) (FRENCH LAY GAIMAR) 237 *Rois Alsi, qui done regna 238 *Et les ii, regnes gouerna, 284 *Z/7 mes chine t quert s a file, 285 *Quc ia estoit creue et grant. 235 DE EUS estoet ore ci lesser. 53 *Gunter auoit vn soen chastel 54 *Sus la marine, fort et bel ; 56 *Dedenz mist sa femme et son fiz ; 57 *A vn baron de la contree 58 *En ad la garde comandee. 61 *Sur totes riens li comanda 46 G has no transition from the death of Ekenbright and the queen to the mar- riage. E Argentine fu norie A Nicole, e en Lindeseie. Si com dit lantive gent, Ele nout nue cheval parent De par sun pere des Daneis. Then follows imme- diately : Oiez ke fit eel felons reis. (91—98.) 47 This episode of the cruel treatment of Goldborough belongs only to the English Lay. It corresponds to the treatment accorded Havelok, and one is a duplication of the other, as Wittenbrinck first suggested in his study Zur Kritik und Rhythmik des altenglischen Lais von Havelok dem Ddnen^ Burgsteinfurt, 1891, p. 5 : ** Die Fabel seiner Geschichte ist diirftig. Zwei grosse Ziige gehen an- fangs parallel und werden dann zusammengefuhrt." The English version, too, shows an entire modification of the early Danish events in order to carry out an extended duplication of the English part of the story. Thus in both England and Denmark the king knows he is going to die, he summons his barons, and entrusts his kingdom and infant heir to an earl who takes an oath and afterwards usurps the kingdom. 48 In /"and G Havelok' s father is killed in combat with Arthur. E has du- plicated Athelwold's illness and death. In G no provision for Havelok was made by Gunter, but in F, before the battle, Havelok and the queen were entrusted to Grim's care. Although the circumstances are different, E and F agree in the provision made by Havelok' s father for him. In G the queen flees with him after Gunter' s death. KUPFERSCHMIDT'S STUDY 35 (ENGLISH lay) (Godard imprisons the children in a tower, starves them and kills the sisters, but spares Havelok. He repents of his clemency later and gives Havelok to Grim to drown. The light emanating from the sack into which Havelok is thrown shows Grim's wife that the boy is no common child. The light 50 is described as in both French versions. But there is the additional **kyne merk " on his right shoulder. Grim saves Havelok but claims reward from Godard for killing him. Godard refuses it and threatens Grim, who sees he must flee or **Heye hangen on galwe-tre."5i Grim sells his grain and animals. ) 706 Hise ship he grey\ede we I inozvy 707 He dede it tere, and ful wel pike, 708 pat it ne doutede sond ne krike ; 709 per-inne [he] dide a ful god mast, 7 I o Stronge kables, and ful fast, 7 1 1 Ores gode, and ful god seyl ; 7 1 2 per-inne wantede nouht a nayl, 7 1 3 pat euere he sholde J)er-inne do ; 7 1 4 Hwan he hauedet grey^ed so, 7 1 5 Hauelok \eyunge he dede \er-inney 5^ 7 1 6 Him and his wify hise sones \rinney 7 1 7 And hise two douhtres, pat /dire wore : (FRENCH LAY GAIMAR) 62 *Son fiz, quil forment ama ; 63 *Qe si de lui mesauenoit, 64 *En bataille sil morroit, 65 *Qa son poeir le garantist 66 *Et fors del pais le meist,49 89 ^Grimfet niefs apparaillefy 90 *Et de viande bien charger. 97 ^^ant sa nieffut apparailleey 98 ^Dedenx fist entrer sa meisnee, 99 ^Ses cheualers et ses serganzy 100 ^Safemme demeine et ses enfanz; 49 Notice that in F Havelok and the queen are put in a strong castle for pro- tection ; in E the enemy places Havelok and his sisters in a strong castle-tower to torment them. In G no strong tower is mentioned. 50 G does not mention the Hght until the night of the marriage. F contains a description of the hght in the part of the story just preceding the embarcation. 51 In Z' Grim takes the initiative and embarks with Havelok and the queen, though not for considerations of his own danger. In G the queen flees to Grim with Havelok and they embark. 52 F and E both describe the embarcation, though they difi^er as to the order in which the family entered the boat. ^J K-tro pt'fi 36 THE OLD FRENCH LAI D*HAVELOC (ENGLISH lay) 723—5 (The *bise' drove them to England. ) 53 In Humber Grim bigan to lende. In Lindeseye, riht at \e north ende. per sat his ship up-on ))e sond. But Grim it drou up to ))e lond ; 737 And ]>ere he made a litel cote 738 To him [and his] , and to hise flote. 733 734 735 736 (FRENCH LAY GAIMAR) 10 1 *La reyne mist el batel, 102 *Haueloc tint souz son mantel ; 103 *I1 meismes aprcs entra, 108 *Mes ne sieuent queu part aler 122 *Tant ont nage et tant sigle 1 23 *Qen vne hauene sont paruenu, I 24 Et de la nief a terre issu. 125 ^Ceofut el North y a Grimesbi. 739 Bigan he )>ere for to erde, ']\o A litel hus to maken of er\e,^^ 741 So fat he [and his] wel wore 742 Of here herboru herborwed \ore ; 743 And for J)at Grim jjat place auhte, 744 p^ stede of Grim \e name lauhte ; 745 So \at Grimesbi [//] calk 746 [He] ]>at ))er-of speken alle;55 749 Grim was fishere $wi])c ^od, 750 And mikel cou|>e on the flod ; 751 Mani ^od fish )>eMnne he tok, 752 Bo])e with net, and [ek] with hok. 753 He tok ]>e sturgiun, and |)e qual, 754 And ))e turbut, and lax with-al, 755 He tok )>e sclc, and [ek] |>e el ; 756 He spedde ofte swi)>e wel : 757 Keling he tok, and tumberel, 758 Hering, and ])e makerei, 759 pe butte, \t. schulle, |>e })ornbake : 760 Gode paniers dede he make, 761 On til him, and ofer ))rinne 53 F and E represent Grim as sailing at random and driven on the shore of England. In G Grim directs his course to England : (581) En Lindeseie volt aler. E omits the attack of the pirates. 54 The house was made by Grim from his boat in both F and G. 55 This information about Grimsby is not given in G. The passage in E strikingly corresponds to that of F. The names of fishes are the same in G and Ey and it is noteworthy that fishes are enumerated in both versions. 793 Li prodoms mansion ileua, 129 // / adresca primes maison 794 *Tut primerement si herberga ; 134 ^Iloec dedenz sest herberge. 139 Plusurs a li sacompaignerent, 1 40 *Sus le hauene se herbergerent ; 141 ^Pur son nony qil eurent oi, 142 *Le liu appellerent Grimesbi. 442 Per un batel ben ^uarisimes, 443 Dunt nostre pere ala pescher. 1 3 5 *Pescher aloit si com il soloit, 795 *Assez nus troua a manger ; 796 Par vendre siel et par pescher, 444 PcJson eumes a manner; 445 Turbuz, salmuns e muivels. 446 Graspeis, porpeis e makerels KUPFERSCHMIDT'S STUDY 37 (ENGLISH lay) 762 Til hise sones, to beren fish inne, 763 Vp o-londe to selle and fonge ; 764 Forbar he neyj>er tun, ne gronge, 765 pat he ne to-yede with his ware ; 766 Kam he neuere horn hand-bare, 767 pat he nc brouhtc bred and sowcl 768 In his shirte, or in his couel ; 7 7 1 And h wan he tok ]>e gretelaumprei, 772 Ful wel he coujje Ipc rihte wei 773 To Lincolne, Ipe gode boru ; (Havelok tries to earn his Hving when twelve. His labor is de- scribed. A famine follows and Grim talks to Havelok, advising him to go to Lincoln to find work. 56 He says : 57 (FRENCH LAY GAIMAR) 839 840 841 843 844 850 851 852 853 854 855 856 857 858 859 *Hauelok, dere sone, I wene that we deye mone For hunger, )7is dere is so strong, Betere is )>at ]>u henne gonge pan pu here dwelle longe ; Betere is ]>at ]>u )?ider go. For }>er is mani god man inne, per ]?ou maght ]>i mete tvinne. But wo is me ! ])ou art so naked. Of mi seyl y wolde were maked A cloth, ))ou mihtest inne gongen, Sone, no cold )>at )>u ne fonge. ' (And the making of the garment is described.) He tok ]>e sheres of ]>e nayl. And made him a couel of ])e sayl. And Hauelok dide it sone on ; (The king's cook, attracted by his strength and work,' takes him to his kitchen to serve. ) 58 448 Eumes pain, e bon peison. 449 Del peissun cangium le pain 450 Horn nus aportout a plain. ( Havelok' s growth and strength are described. Grim rejoices in him, but laments his position and surroundings. Grim then talks to the boy and advises him to go to Lincoln to the court to ** aprendre sens et avoir querre.") 1 66 *" Beau fiz, ' ' fet il, ** entend a moi 167 ***Ici manon mult soutiuement, 168 *** Od pescheours, od poure gent, 171 **e mete to )>e castel, 878 And gat him })ere a fer}>ing wastel. (He pushes aside the 16 porters of the cook and catches up the fish.) 895 He bar up wel a carte-lode 896 Of segges, laxes, of playces brode, (Havelok tells what he can do.) 2 Fir and water y wile yow fete, 3 pe fir blowe, and ful wele maken ; 4 Stickes kan ich breken and kraken, 9 Ful wel kan ich dishes swilen. (FRENCH LAY GAIMAR) 244 Et mult le vist de bon semblant (His duties are described.) 932 lilck?imlol^&wt\\tyWaterup-droWy 933 And filde ]?er a michel so ; 935 Bi-twen his hondes he bar it in, 936 Al him one, to ))e kichin. 945 Of alle men was he mest meke, 946 Lauhwinde ay, and bli\e of speke ; 947 Euere he was glad and bli\ey 949 It ne was non so litel knaue, 951 pat he ne wolde with him pleye ; 952 pe children that yeden in })e weie 953 Of him he deden al her wille,59 954 And with him leykeden here fille. 955 Him ioueden alle, stille and bolde, 956 Knihtcs, children, yun^c and oldc ; 957 Alle him Ioueden )>at him sowen, 958 Bopen heye men and lowe. 959 Of him J}e word ful wide sprong, 960 Hu he was mikel, hu he was strong,^ 961 Hu fayr man God him hauede maked. 245 Merueillous fcs poeit leuer. 246 "^Busche tailler 247 '^Les esquieles receuoit, 248 *£/ apres manger ks lauoit ; 246 "^ewe porter 277 XII. homes ne poeient leuer 278 Le fes que il poeit porter. 253 ^Tant estoit franc et deboneire. 254 ^^e tuz voloit lur pleisir fere. 1 39 Pur CO cstait sj ben amez, 140 E si preisez* c si loez, H* Navcit frans horn en la meison, 142 Si Cuheran en voleit don, 143 Kil ne lui donast volunters; 243 Purceo qe fort le vist et grant, 244 Et mult le vist de bon semblant. 59952 corresponds to 253 of F, but with a change in subject. Havelok did everyone's pleasure in F. In both G and F he gave abundantly to the "valets," and G adds that he was much loved for this, and that there was no one who would not give to him. In F they make fun of him. In E^ however, the description of the attitude recalls the general phrases used for Athelwold (30) : Him lovede yung, him loveden olde, ^< ^<, Erl and barun, dreng and thayn, ^ Kniht, and bondeman, and swain, etc. ^Ey Fy and G correspond in giving Havelok' s characterisdcs. KUPFERSCHMIDT'S STUDY 39 (ENGLISH lay) 962 But-on ]>at he was almest naked : 979 For \anne he weren a lie samen 980 At Lincolney at \e gameriy^'^ 981 And )>e erles men woren alle |?ore, 982 Was Hauelok bi ])e shuldren more 983 pan \t meste J>at ))er kam : 984 In armes him noman \ne~^ nam 985 ptf/ he doune sone ne caste; 986 Hauelok stod ouer hem als a mast. 989 In Engelond [was] non hise per 990 Ofstreng\e \at euere kam him ner^ 99 1 Als he was strong y so was he softe ; g()z pey a man him misdede ofte, 993 Neuere more he him [misseyde], 994 Ne hond on him with yuele leyde. 1 00 1 And he (Godrich) gart komen into ]>e tun 1002 Mani erly and mani barun; 1003 And a lie [men^ pat Hues were 1004 1005 1006 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 In Engelond, ]7anne were ))ere, . pat )?ey haueden after sent To ben )>er at \z parlement. (A description of games follows in which Havelok excels all.) porhut England yede ])e speke, Hu he was strong, and ek [ful] meke ; In the castel, up in |>e halle, pe knihtes speken J)er-of alle. So that Godrich it herde wel per speken of Hauelok, eueri del, Hu he was strong man and hey^ Hu he was strongs and ek \_ful sley-] And )?ouhte Godrich, *))oru ]>is knaue Shall ich Engelond al haue. And mi sone after me ; ^2 (FRENCH LAY GAIMAR) 239 ^Bone curt tint et grant gent; 240 "^A Nicole manoit souent, 262 Li cheualer et li sergant, 264 Desquil seurent sa grant vertu, 265 Deuant eus liuter le fesoient 266 As plus for X homes qil sauoient, 267 Et il trestouz les abatit; 268 Et si nuls de eus le mesdeisist, 269 Par dreite force le lioit ; 270 Tant le tenoit et iustisoit 271 Quil li auoit tut pardone, 272 Et quil restoient acorde. 279 En la curt fut lungement, 280 *Dici qun assemblement, 281 *j^ li baron a la curt vindrent 282 *^/ de Ekenbright lur terre tip- drent ; 2JT, ^Li rots forment sesmerueilloit 274 ^De la force qen lui veoit. 275 *Dis des plus forz de sa meson 276 *Neurent vers li nule fuison ; ^^ In G no special gathering at this time at Nicole is mentioned ; in /" we read of a king's court at Nicole, and the late English author introduces a parlia- ment there. ■ 62 In all this account E is much closer to F than to G. The latter states that the king made Havelok his juggler, and then married him to his niece to shame her. F shows how Havelok wrestled before the knights, and how the king him- ^^ 40 THE OLD FRENCH LAI D'HAVELOC (ENGLISH lay) 1076 For so i wile put it be. 1077 King Ape/zoa/^ me dide swere 1078 Vpon al \t messe-gere, 1079 p^/ J' shulde his doubter yeue 1080 \>e hexte \man\ \>at mihte Hue, 108 1 \>e beste, \e fairest, \>e strangest oh ; 1082 '^at gart he me sweren on \e bok, 1083 Hwere mihte i finden ani so hey 1084 So Hauelok is, or so sky? 1085 ])ouh y souhte he\en in-to Tnde, 1086 Sofayr, so strong, ne mihte yfinde. 1087 Hauelok is \2X ilke knaue 1088 pat shal Goldeborw haue.'^a 1089 pis ])ouhte [he] with trechery, 1090 With traysoun, and with felony ; 1 09 1 For he wende, \d\. Hauelok wore 1092 Sum cherles sone, and no more; 1097 He wende, ]?at Hauelok wer a ))ral,64 (Goldborough is brought to Lin- coln, and the conversations be- tween the king and her, and be- tween the king and Havelok, are reported. Both refuse to marry and are forced to do so by the king with threats. The wedding is also described. )^5 (FRENCH LAY GAIMAR) 354 '^**^uant E ken bright le roi fini, 355 ^**En ma garde sa jille mist, 356 '^**yn serement iurer me fist 357 ** * -^^ pl^^ fort home la dorroie 358 *<*^ el reaume trouer porroie. 359 360 361 362 363 364 365 366 ** 367 ** 368 *' 369 ** 370 ** As sex ai quis et demande, Tant qen ai vn fort troue. Vn valet ai en ma quisine A qui ieo dorrai la me s chine. Cuaran ad cil a non. Li dis plus fort de ma maison Ne se poent a lui tenir. Son giu ne sa liute sufFrir. Veritez est, desqa Rome De corsage nad si grant home. Li garder voil mon serement, Ne la pus doner autrement." 1 6 1 Purhoc sil estait en tel despit, 162 Venuz esteit de gcntil lit: 163 E si li reis saparceust, 1 64 Ne quid ke ja sa nece eust 377 Sa niece lur fet amener. self marvelled at his strength. After the assembly at court where the barons de- manded the keeping of the oath, the king thought of Havelok and arranged the marriage. In E the king noticed Havelok in the games at the time of the parlia- ment held at Lincoln ; he bethought himself of his oath and decided that Havelok should marry Argentille. 63 Lines 1079-1 182 repeat 198-201 of the first part oi E. ^4 This observation of the author of E is not unUke that in G. However, the reflection is a natural one to make on such a situation, and it might perhaps have occurred to the author of F, had he not given nearly all this scene as conversation between the barons and the king. ^5 In F and G neither these interviews nor the wedding are described. KUPFERSCHMIDT'S STUDY 4» (ENGLISH lay) 1 1 7 5 He weren spused fayre and wel. (Havelok leaves with his bride for Grimsby ^^ because he fears the king and that : ) 1 1 9 1 Men sholde don his leman shame, 1 1 92 Or elles bringen in wicke blame ; (They hasten on) : 1202 ** Til he komen to Grimes by, 1203 \>anne he komen ])ere, \anne was Grim ded, 1 204 Of him ne haueden he no red ; 1205 But hise children alle fyue 1 206 Alle weren yet on Hue ; ^7 1247 On |?e niht, als Goldborw lay, 1 248 Sory and sorwful was she ay, 1 2 49 For she wende she were bi -swike, 1250 pat she were yeuen un-kynde- like.^^ I 25 I O niht saw she \er-inne a liht, 1252 A swij)e fayr, a swi]7e bryht, 1253 Al so briht, al so shir 1254 So it were a blase of fir. 1255 She lokede nor}), and ek south, 1256 And saw it comen ut of his mouthy 1257 pat lay bi hire in J)e bed : 1258 Noferlike \ouh she were adred! ^ (FRENCH LAY GAIMAR) 378 Et a Cuaran esposer; 379 Pur lui auiler et honir 380 La fist la nuit lez lui gisir. 556 A Grimesby sen alerent. 557 * Mes li prodoms estoit finiz 558 E la dame qis out nurriz. 559 Kelloc sa fille iont trouee ; 560 Vn marchant lout esposee. 1^7 La nece al rei se compleigneit; 188 Sovent son uncle maldisseit, 189 Ki si laveit desherite, 190 E a un tel horn donee; 43 5 Mult out del sunge grant poour ; 436 Puis out greindre de son seignur 437 Pur la fiambe qele choisit 438 ^ de la bouche li is sit. 66 In F and G the story of the vision and the flame precedes the departure for Grimsby, and leads to it. In E the two leave for Grimsby because they fear the king. The vision and the light are mentioned later. 67 At Grimsby they are well received, but there is no suggestion of Havelok*s story or of a return to Denmark as yet. That night, however, there comes the supernatural revelation to Goldborough. In F and E note that Grim is now dead. In G he had died before Havelok left home. 68 Goldborough' s sorrowful attitude is commented on in both E and G. Her feeling, however, is more bitter in G, and her complaint is directed against her uncle who is not mentioned in E. The resemblance is not close. F does not make any such observation and represents Argentille as reconciled to Havelok almost immediately. There is a difference of tone in the three poems which makes it possible for E and G to record an observadon which, in F, would be out of keeping with the ideal of Argentille, who should not despise Havelok whether he is known or unknown to her. 69 The flame is partly explained in E by the cross (the **kynemerk") and by the angel's voice. The vision is seen by Havelok and not by Argentille, as in G and F, The vision is a different one and is explained by Goldborough. It serves no purpose. In G and F the vision comes first. Argentille awakens in fright, then sees the flame and arouses Havelok, who gives an unsatisfactory ex- 42 THE OLD FRENCH LAI D'HAVELOC (ENGLISH lay) (After the light she perceives the **kynemerk" on his shoulder, and hears the angel's voice tell her that Havelok will be king of Denmark and England.) 1 2 66 For Haueloky \at haue\ spuset ]>e, 1267 [Is] kinges sone and kinges eyr ; 1 269 // bikennethy \at he shal 1270 Dene mark hauen, and England al; (FRENCH LAY GAIMAR) 441 442 443 (Havelok dreams and Goldbor- ough interprets the dream, advis- ing Havelok to take Grim's sons to Denmark with him. Havelok tells his tale to Kelloc and the others. ) 7° 440 '* I preie you )>at ye wende with me. And ich may mak you riche men; Ilk of you shal haue castles ten. And ))e lond Jjat J)or-til longes, 444 Borwes, tunes, wodes and wonges.'* 71 (Ubbe beholds Havelok.) 645 Hauelok bi-held he swi\e we I, Hu he was wel of bones maked, Brod in \e sholdres, ful wel schapedy ])uke in \e brest, of bodi long; 649 He semede wel to ben wel strong. 659 * Haueloky haue [)>^a] \i boney 660 And y ful wel rede \e 661 ptf/ \ou comey and ete with me 646 647 648 519 *** 520 *** 521 *«« 522 *** 523 **« 524 *" 525 *" 526 *" 527 **' 528 *" 529 *" Bele," fet il, **ceo qe sunge as De ton baron, tu le verras. II est ne de real lignage, Oncore auera grant heritage. Grant gent fra vers li encline, II serra roi et tu reyne. Demande li qi fut son piere, Et sil ad sorour ne frere ; Puis si meint en lur contree ; Iloec orras la destinee Dont ert nez et dont il est. 639 Argentine, quant ele loit, 641 A eus promet foi et amur ; 643 Grant bien lur f era y ceo dit. 739 Li senescaus ala auant ; 740 Vist Haueloc et creu et grant, 743 Gent cors et bele feturey 744 Lungs braz et grant fur cheure. 745 Ententiuement lesgarday 655 *** Haueloc " fet il, ** beaus amis, 657 ^** En son chastel va herbergery 658 ** Et a sa table va manger y planation. The vision, as explained by the hermit in Fy gives the reason for going to Grimsby. In E the idea of flame, vision, and departure is kept, but the order, circumstances, and relationship of the three are changed, and other matter is introduced. 70 In G and F it is Kelloc who tells Havelok to go to Denmark and take her brothers. 71 A leaf containing 180 lines is lost, and in the next passage Havelok goes to Ubbe's house with a ring. This is not found in G or F. KUPFERSCHMIDT'S STUDY 43 (ENGLISH lay) 1662 To-day y ]>ou, and\ifayre wify 1663 pat )>ou louest also |>i lif. 1 664 And haue J)ou of hire no drede, 1665 Shal hire no man shame bede. 1666 Bi ))e fey y owe to ]?e, 1667 perof shal i [miself] borw be.' (Havelok, though afraid of con- sequences, brings Goldborough. Ubbe loves him.) 1 7 1 4 Hwan it was comen time to e Jirste niht he lay \er-inneyTi 2091 Hise tuify and \eK\ his serganz \rinney 2092 Aboute ])e middel of pe niht 2093 Wok Ubbe, and saw a mikel liht 2094 In )>e bour [j>er] Hauelok lay, 2095 Also briht so it were day. 2096 ' Deus ! ' quoth Ubbe, ' hwat may ])is be? 2097 Betere is i ^o miself, and see: (The light is not understood until the "kunrik" denoting his king- ship is seen. ) 2 1 49 So ]>at he knewen, at |>e laste, 21 50 pat he was Birkabeynes sone, 2151 pat was here king, J)at was hem wone 2152 Wei to yemen, and wel were 2153 Ageynes uten-laddes here. (Ubbe tells Havelok he shall be king. ) 2 1 90 On \t morwen, hwan it was liht, 2192 Vbbe dide up-on a stede 2193 A ladde lepe, and \ider bede 2194 ErleSy barounsy drengesy theynesy 2195 Klerkesy knihtes, bur gey Sy sweynesy 2196 p/7/ he sholden comen a-non 2197 B if or en him sone euerilkon (Ubbe summons all and they come as he bids. He tells the Danish part of Havelok's story and shows the heir of Denmark to them. ) 74 2240 Lokesy hware he stondes her: 2241 In al j>is werd ne haues he per ; 2242 [Is] non so fayr, ne non so long, 2243 Nenon so mikel,nenon so strong. (FRENCH LAY GAIMAR) 828 La nuit en sa chambre gisir. 837 Meismc lure quil dormit 838 De sa bouche le feu issit. 639 E li prodom levat del lit; 640 Quant il i vint la flambe vit 841 Et il en ad Dieu mercie, 842 Qc le dreit heir ad recouerc. 843 *Ses chapeleins fet demander; 844 *Ses briefs escriure et enseeler, 845 Par ses messages les mandoy 846 Et pur ses amis enuoiay 847 Pur ses homesy pur ses parenz. 848 Mult i assembla granz genz. 910 *^ tuz ensemble le mustra ; 911 ** Seignurs, purceo vus ai mandez, 912 ** Qe Dieu nus ad reuisitcz. 73 Since Ubbe has not asked Havelok who he is, the light and the cross serve to reveal Havelok's identity to him and do not merely confirm an opinion already conceived, as in G and F. E is nearer to G in that Ubbe goes to see for himself what the light is. In G Sygar has Havelok watched by servants but goes to verify their report himself; in F the chamberlain brings the report to Sygar but the latter rejoices without going to see for himself. 74 Havelok tells his story to Sygar in the French poems. The bath, robes, hatchet, and horn episodes are lacking in E, KUPFERSCHMIDT'S STUDY 45 (ENGLISH lay) 2 2 46 Bes of him ful glad and bli^e, 2247 And cometh alle hider swi|?e, 2248 Manrede youre louerd forto make, 2250 / shal mi-self do first ]>e gamen, 2251 And ye si))en alle samen.' 2252 O hues ful fayre he him sette^ 2253 Mouhte noj^ing him |)er-fro lette, 2254 And bi-cam his man riht ]>are, 2255 pat alle sawen Tpat ]?ere ware. 2256 After him stirt up laddes ten, 2257 And bi-comen hise men ; 2258 And si\en euerilk a baroun 2259 ]>at euere weren in a I that toun; 2260 And si\en drengesy and si\en thaynes, 2262 So pat, or .))at day was gon, 2263 In all ])e tun ne was nouht on 2264 pat [he] ne was his man bi-comen: 2265 Manrede of alle hauede he nomen. 2268 Grundlike dide he hem swere 2269 pat he him sholden god fey th here 2270 Ageynes alle )>at woren on Hue ; 2271 per-yen ne wolde neuer on striae, 2272 ptf/ he ne maden sone ]>at oth, 2273 Riche and pourey lef and lothy 75 2312 Hwan he hauede manrede and oth 2313 Taken of lef and [ekj of loth, 2314 Vbbe dubbede him to kniht^^ 2315 With a swerd ful swi)?e briht ; (Then follow rejoicing, games, gleemen, and a feast. Grim's sons are knighted also. ) 77 2354 Hwan J)at feste was al don, 2355 A thusand knihtes wel o bon 2356 With-held J)e king, with him to lede; (FRENCH LAY GAIMAR) 913 ** Feez ci nostre dreit heir ; 914 * * * Bien en deuom grant ioie aueir. * ' 9 1 5 Tut primer ain se desafubla, 9 1 6 Pardeuant lui sagenuilla ; 9 1 7 Sis homs deuinty si li iura 918 Qe leaument le seruira 919 Li autre sont apres ale, 92 1 Tuit si home sont deuenu. 922 Puis quil li eurent receu, 923 La nouele fut recontee : 925 De totes parz i accurroient, 926 Et riche et pouere, qui loeient, 927 De lui firent lur auowe. 925 De totes parz i accurroient, 926 Et riche et pouere, qui loient, 927 De lui firent lur auowe. 928 ^A cheualer lont adubbe, (Gaimar mentions a feast after the battle, 1. 757.) 75 This whole passage (2240-2273) is much closer to F than to G. Gaimar has merely (727-734) : Tuz ses homes ad done mande : Lores li firent felte. II meismes sagenulla, De fai tenir laseura. Puis enveia pur les barons. A ki eel reis aveit tencons. Tuz sunt ses homes devenuz, E a seignur lunt receuz. This passage in E is like that of Fy which is, however, more detailed. 76 This is an important point of resemblance between E and Fy since G does not mention the knighting of Havelok. 77 Neither G nor F mentions a feast in just this connection, though G mentions one after the battle (757). 46 THE OLD FRENCH LAI D'HAVELOC (ENGLISH lay) 2360 With hem [ek] fiue thusand gode 2361 Sergaunz, )?at weren to fyhten wode, 2370 Half hundred knihtcs dcdc he calle, 2371 And hise fif thusand sergaunz alle. ( Robert, Havelok's foster brother, meets Godard and calls him traitor and **caynard." Godard takes a knife and smites him through the right arm. His men come to the rescue and are killed by Have- lok's men. Godard is bound and taken to Havelok. The traitor is then duly sentenced, flayed,drawn and hanged.) 78 2512 panne he was ded, }>at Sathanas, 2513 Sket was seysed al ))at his was 2514 In ]>e kinges hand ilk del, 2515 Lond and lith, and o]?er catel, (Havelok leaves the land in Ubbe's keeping, and vows a monastery to Grim. Godrich in England hears that Havelok is ruler of Denmark and is coming with an army.) 79 2548 He dide sone ferd ut \bede^ 2 5 49 p/7/ al \at euere mouhte stede 2550 Ride, J)or helm on heued bere, 2551 Brini on bac, and sheld and spere. (FRENCH LAY GAIMAR) 929 Tant li aida le senescal, 931 Qe merveillous ost assembla. 737 • • • • dcs chevalers 738 Ourent il bien trente n^illers. 96 1 *Haueloc fut de grant vertu ; 962 *Le roi Hodulf ad si fern 963 *Dune hache qil apporta, 964 *Quil labatit, puis ne leua, 965 *Iloec loccist deuant sa gent. 971 *Apres cest fet ad receu 972 *Le regne qa son piere fli 973 *Par la terre bone pees mist, 1023 '^Dedenx le tour qil eurent pris, 1024 *Alsi manda ses amis 1025 *£/ toux ceus quil pout aueir ; 78 F and G differ in the story of the battle. Gaimar states that the enemies fought and that Hodulf was conquered. In F a hand to hand contest between Hodulf and Havelok is described, Hodulf was killed in the sight of his people. E is closer to F in this latter detail and also in the former, for, changed and popu- larized as is the whole account, there is still more reflection of the hand to hand contest in the struggle between Godard and Robert than of the regular battle of Gaimar's description. There is a general fight between Havelok's and Hodulf's men at the end of the contest, but this is no regular battle. The tone of the de- scription is different and Robert challenges Hodulf by calling him names, whereas Havelok calls him to fight to spare the **menue gent." 79 Notice that in F and G Havelok receives his father's kingdom. In E he receives all the personal possessions of Hodulf as well. It is on this occasion that the feast of G is mentioned (756), De lui firent seignur e reis. Grant feste tint, e grant baldoire. In G as in £" the expedition to England seems to follow imme- diately. In F the peace is described, and after four years Argentille commands Havelok to reconquer her kingdom. KUPFERSCHMIDT'S STUDY 47 (ENGLISH lay) 2552 Or ani oJ>er wepne here, 2553 Hand-ax, syj)e, gisarm, orspere, 2554 Or aunlaz, and [ful] god long knif, 2555 paf, als he louede leme or lif, 2556 ])ey sh olden comen [a//e] him to- 2558 To Lincolney per he lay^ 2559 Of Marz Jje seuentenj?e day.2° (After the separate deeds of Roberd, Huwe, and Ubbe have been described, Havelok addresses Godrich and promises him for- giveness if he will yield up her rightful inheritance to Argentille. Godrich refuses. Havelok fights him, conquers and binds him, and sends him to the queen for custody until the knights shall pass judgment on him. ) ^^ 2766 pan J)e EngHshe men J)at sawe, 2767 pat pei wisten, heye and lawe, 2768 pat Goldeboru, })at was so fayr, 2769 Was of Engeland riht eyr, 2770 And })at ])e king hire hauede wedded, 2771 And haueden [he] ben samen bedded, 2772 He comen alle, to crie merci, 2773 Vnto |)e king, at one cri, 2774 And beden him sone manrede and oth, 2775 pat he ne sholden, for lef ne loth 2776 Neuere more ageyn him go. (FRENCH LAY GAIMAr) 1026 ^Nul ni less a remaneir. 1027 ^A Theford les ostz assemblerent 1083 Tote sa terre li rendit 1084 Que Ekenbright tint tant come il vesquit. 965 deuant sa gent. 966 Qe touz li crient hautement, 967 **Sire, merci, qe ni moroms, 968 ** Car volenters te seruiroms.*' 969 Cil se sont a lui tourne, 970 Et il lur ad tut pardone.^^ ^°In the French versions Godrich is challenged by Havelok. Although Gaimar indicates that Edelsi assembled a host (764-770), the longer description is given in the Lay. In this respect E is closer to F. In G no place or time is mentioned for the assembly ; in i^ an indefinite time and a definite place, Theford; in ^ a definite time and place, Lincoln, the 17th of March. ^^ This single combat is found neither in G nor in F. It may be, however, a reflection of Havelok' s single combat with Hodulf In both French stories \ Edelsi makes peace and is forgiven on yielding up the kingdom. Perhaps this is >• reflected in Ey where Havelok offers to forgive Godrich if he will surrender the 3 kingdom. ^2 In F and G the mention of forgiveness is made in connection with the Danish subjects and not the English. The peaceful ending of the affairs in Eng- land in both versions, however, allows this forgiveness to be understood. 48 THE OLD FRENCH LAI D'HAVELOC (ENGLISH lay) (The earls bring in Goldborough, and the English fall on their knees and confess their fault, saying : ) 2800 For Englond auhte forto ben 2801 Youres, and we youre men. 2802 Is non of us, [ne] yung ne old, 2803 pat [he] ne wot, ^at A)?elwold 2804 Was king of [al] ^is kunerike, 2805 And ye his eyr. (Godrich is condemned and burned. 83 Grim's daughters are married to earls, and Bertram the cook is rewarded. Then follows the coronation feast of forty days in London.) ^4 2942 per-after sone, with his here, 2943 For he to Lundone, forto here 2944 Coruney so pat [alle\ it sawe. 2948 ])e feste of his coruning 2949 Lastede with gret ioying 2950 Fourti dawes, and sumdel mo. (Havelok reigns sixty years. The French versions end here with no mention of Havelok's children.) (FRENCH LAY GAIMAR) 093 II nout nul heir si droiturel 094 Come Haueloc et sa muiller. 1087 Mes Haueloc safeste tint 1088 A la cite quant il vint ; 099 XX. anz regna, si en fut rois.^s This comparison shows that the resemblances to Gaimar alone are as follows : 1. An enumeration of the fishes which Grim caught. (^, 751-759 ; G, 445— 446.) 2. A comment on the king's ignorance of Havelok's real position. (£, 1089— 1097 ; G, 161-164.) 3. An observation of Argentine's on her unhappy situation. (^, 1 247-1 250 ; G, 187-190.) 4. The fact that Ubbe himself goes to investigate the light that came from Havelok, as Sygar does in Gaimar. (^, 2096-2097; G, 639-640.) 5. The mention of a feast in Denmark. 83 The punishment of Godrich forms a pendant to Godard's fate. In the I French poems Edelsi dies at the end of a fortnight. 84 A feast is spoken of in F (1087) and in G (806) : Rei Haveloc la tin sa feste. La is indefinite. In the preceding lines Gaimar speaks of the kingdom's extending from Hoiland to Colecestre ; in /^ it is la cite; in £, Lundone, 85 Also in G (817), Vint anz fud reis. KUPFERSCHMIDT'S STUDY 49 6. The placing of the expedition to England directly after the conquest of Denmark. (In F four years elapsed.) 7. The coincidence in the order in which events are narrated in the poem, those of England preceding those of Denmark. The enumeration of fishes does not prove a direct relation between E and G, for the list might have been in F's original too, and reflected in F by the line Asez nos trova a manger, or by some such remark in F; or F's source might have been in- dependently expanded to an enumeration by the authors of G and E. The author of E was especially fond of naming articles of food, and as Skeat^^ says, quoting Madden: "The list of fishes here enumerated may be increased from line 189, and presents us with a sufficiently accurate notion of the different species eaten in the thirteenth century." Gaimar in his list of six gives two not in E, whereas in E many others are added. The enumeration may have been in a source common to all versions, but may also have been condensed in F and expanded directly from F in E, without any reference to Gaimar or Gaimar's source. 2. It seems impossible to decide whether the passages referring to the king's ignorance of Havelok's rank are derived from a passage in the originals of G and E, or whether the authors in- dependently made these casual observations on existing condi- tions. The actual remarks differ. In E the king made his plan treacherously, for he believed Havelok to be some churl's son. In G, although Havelok was in such a plight, he was really of noble birth; if the king had known this he would not have given him his niece. 3- ' The references to Argentine's unhappiness after her marriage may also have been made independently and casually by the two authors. The lines are not alike. In E she lay awake and grieved that she was so betrayed. In G she went to the king and complained often : son uncle maldisseit. 86 Ed. 1902, p. 1 1 1. 50 THE OLD FRENCH LAI D'HAVELOC 4- The fact that Ubbe in E and Sygar in G go in person to see this light seems significant. But an examination of the ac- counts makes it seem less so. In G and F the circumstances are similar. Sygar suspects Havelok's origin and has him watched. His servants in G, the chamberlain in F\ come to tell him of the flame. In G Sygar goes to see for himself, in the more courtly poem, F, he merely gives expression to his joy. A quite diflPerent version is found in E. Ubbe does not know who Havelok is, and, sleeping in a room near him, he is awakened by a bright light. He thinks Havelok must be revelling, and goes to see. Looking through a hole in a board, he sees a bright flame coming from both Havelok and Gold- borough. Then he calls his knights and men to come and see what the wonder is. They all see the cross on Havelok's shoulder and know that he is Birkabeyn's son. Surely the popularizing of the incident, the exaggeration of the brightness of the flame, and the stressing of Ubbe's astonishment give enough motive for his going himself to see the light, whatever the original of this freely developed scene may have been. The description in E contains, moreover, new details of obviously late origin, such as the flame which comes forth from both, and which is brighter than one hundred and seven candles, and the cross which is needed to explain the flame. 5; The feast in Denmark, mentioned in G and E, is not found in quite the same position in the two versions. In £ the feast comes before the battle and is a kind of coronation feast. In G it follows the battle. One cannot be sure that the feast was mentioned in the source of £. It may have been introduced simply to duplicate the coronation feast of England (2344), since events in England and Denmark have been made to cor- respond in other respects. Like the latter celebration, it lasts forty days. The author of E introduced feasts and meals on all possible occasions. (Cf 641-656, 922-926, 17 14-173 5, 1762-1766, 2339-2345.) 6. The fact that the expedition to England immediately followed the peace in Denmark is not important. The author of the KUPFERSCHMIDT'S STUDY 51 Lay probably invented the delay, since in G it is not noted. Even had the English author had F before him he could have omitted the four years' delay, there being no reason for keeping it. \n F it is unnecessary, but serves to make more emphatic the influence of Argentille, who moves Havelok to go to England. Note also that in G Havelok waits for a good wind in Grimsby ; in i^ he sails immediately for Denmark. No one of these arguments seems sufficient to prove a defin- ite relationship between G and E or their sources, since the details seem in each case casual and natural. ^ 7- This last point of similarity — the similar order in the narra- tion of events — has already been touched upon in Chapter 11,^^ where it was noted that two out of the four Havelok versions must have changed the original order of the story, since two open with England and two with the story in Denmark. It seems significant that E corresponds to G in this particular, and although the theory of derivation from a common source appears to be a somewhat simpler explanation of the case than any other, we see that, even according to that theory, not every author, in telling his story of Havelok, followed exactly the order which he found in his source. Furthermore the change was a simple one, involving no internal alteration or adjustment in the passage which could have been merely shifted to another part of the poem. The transition from the English to the Danish part of the story was indicated by the following words, which practically separated the passage and made it easily trans- ferable : Of Goldeboru shul we nou laten. Say we nou forth in ure spelle ! In |>at time so it befelle, Was in }?e lond of Denemark, etc. It seems therefore that this point, although more important than the other more casual resemblances, is not compelling evidence for the independence of the English Lay. Turning now to a comparison between E and F, we note the following points : 87 pp. 13-15. 52 THE OLD FRENCH LAI D'HAVELOC I. The outline of the story corresponds much more closely to F than to G. 2. The following passages offer some resemblances to each other, sometimes significant and sometimes as casual as the resemblances already noted between E and G. (a) I lo-i 1 6. The king's daughter and his illness. (^) 189-203. The oath. (c) 280. The daughter thrives. (^) 706—716. The preparation of the boat and the embarcation of Grim*s family. (e) 725. In F and E Grim arrives in England by chance; in G Grim sails for Lindeseie. (/) 734" ^* ^^ Lindeseye, riht at ]>e north ende ; F, Ceo fut el North, a Grimesbi. (i) 743~74^' Explanation of the naming of Grimsby, lacking in G. (/&) 839-859. The departure from Grimsby is different from that of G, and corresponds in the main to that of F, although the tone of the description is entirely changed and coarsened. (/) 895. The w^ eight he could lift. (y) 912—919, 932. Enumeration of his duties. Very close to F. (i) 947. Glad and bli}>e, £; franc et deboneire, F. (/) 979-981, 1001-1003. The assembly at Lincoln. Gaimar mentions no special gathering at this time at Lincoln. F contains a reference to the king's court gathered there, with chevaliers and barons, and in E we read of a parlia- ment with all the earls. (m) 107 1— 1072. In E Havelok's strength is stressed, and this strength suggested to the king the way to keep his oath and retain England too. Cf. F. G has merely the statement that the king made Havelok his juggler, although elsewhere reference is made to his strength (113— 120). (;7) 1077-1088. Description of the oath again. (0) 1 266-1 270. E is again closer to F in that the information furnished Argentine by the hermit is the same as that given Goldborough by the angel. In G Havelok's own explanation of the dream is the only one given. (/>) 1659— 1662. Ubbe's invitation to Havelok is in the same words used by the merchants in giving Havelok directions. (^) I 714. Time of eating in F and E. (r) 1942— 1944. The seneschal with his knights goes to the fight. KUPFERSCHMIDT'S STUDY 53 (j) 2246. The joy is stressed. (/) 2252-2253. The order in which homage is done Havelok. (2^) 2314. Havelok is dubbed knight. (2/) 2548—2558. The king commands his friends to assemble. 3. In many passages, where E is essentially different from F and G, F seems to represent a transitional form of the story between the simple form of G and the changed form of E. Compare for instance the following epi- sodes : (/7) Argentine's Story. Edelsi and Adelbrict are kings of two parts of Eng- land, and brothers-in-law. Adelbrict dies leaving a queen and a daughter Ar- gentine. His queen goes to her brother Edelsi and gives Argentine and the kingdom into his charge. She dies. Edelsi to keep the kingdom marries Ar- gentine to his cook Cuaran, whom he has noticed and made his juggler. Alsi and Edelbright are kings of two parts of Eng- land, and brothers-in-law. Edelbright feels that he is to die, and sends for Alsi and puts in his care Argen- tine and the queen, making him take an oath to guard Argentine until she is of age, and then to marry her to the strongest man and restore the kingdom to her. The queen dies. Alsi keeps the kingdom ; the barons assemble to make him keep his vow ; he remembers Cuaran, whose strength has impressed him, and this Cuaran he marries to Argentine. ( Havelok' s story had already been told.) Athelwold is king of all England ; he becomes ill, and sends for his earls to choose the best guardian for the kingdom ; they choose Godrich. He is made to swear that he will take care of Goldborough and marry her to the strongest man he can find. Athelwold dies. Godrich shuts Goldborough up in a tower and starves her. Then fonows the story of Havelok. G and F seem to have the original form of the story with the two kings. G appears to have the primitive conception in these early events of this narrative, for the queen plays an active part in it. She takes Argentille and goes to her brother. In F the guardianship idea is introduced and the queen is only mentioned. Alsi takes the oath to protect Argentille and marry her to the strongest man, Veiant sa gent (218). This scene is further devel- oped in E where the assembly of nobles chooses the guardian. 54 THE OLD FRENCH LAI D'HAVELOC Gunter is killed by trea- son of Hodulf. Hodulf is left in charge of Denmark. The queen of Gunter fears and flees with Havelok to Grim, the fisherman, who saves them in his boat. The queen is killed by pirates. The rest land in England. Grim supports them by fishing. When he dies, Havelok leaves Grimsby and goes to Ed- elsi's court. {b) Havelok's Story. F Gunter before battle en- trusts the queen and Have- lok to the guardianship of Baron Grim. After the death of the king. Grim, fearing for them, leaves the strong castle and takes them in a boat on the sea. Pirates kill the queen. They land in England. Grim supports them by fishing. A town grows up named Grimsby for Grim. Distressed at the surroundings of Havelok, Grim sends him to court to seek his fortune and pro- vides him with clothes. Birkabeyn of Denmark had three children. About to die, he summons his barons to choose a guar- dian. Godard, when cho- sen, takes an oath to pro- tect them and to give Havelok the kingdom. He shuts them in a tower, kills the girls and gives Havelok to Grim to drown. Grim saves the boy because of the light and the cross which reveal his royal birth. When Godard re- fuses the reward promised Grim, the latter with Havelok and his own family flees from him to the sea. Grim lands in England. A town grows up where he lives which is named for him ; he supports his family by fish- ing. Famine comes ; there are no fish. Grim calls Havelok and sends him to Lincoln to work. He makes him clothes from a sail. The same point just discussed with reference to the queen, Adelbright's wife, is illustrated again in this passage, where the queen's part is even more active. She fears, and, taking Have- lok, flees and finds Grim. The guardianship here introduced by the author of F is even more clearly his own invention than is the guardianship of Argentille, for we know he made Grim a baron (from the poor fisherman of the legend) in order to make him a fit guardian. He changed the tone of the poem, and the estate of Grim. Is it not more likely that he introduced the courtly idea of a guardian for the queen and Havelok than that Gaimar dropped the mention of a guardian and invented the queen's own act? There is a duplication in G, h\ and E, The two queens in G correspond. \n F the two guardians are KUPFERSCHMIDT'S STUDY 55 similarly appointed but differ in their efficiency, one being good and one bad. In £ a further deviation in the story is intro- duced and the guardianship is made to duplicate that of God- rich in England. Godard is appointed in the same manner, takes the oath in the same way, and is in reality the arch traitor (Hodulf of the French versions) in the role of guardian, which Grim represents in the Lay. In this connection the substitu- tion and duplication make evident the artificiality and the lack of primitive traits in both F and E. The primitive character conjectured for Gaimar's story seems attested by a comparison with those versions, and also the intermediate, transitional char- acter of F is shown. Furthermore, the description of the de- parture from Grimsby, to which frequent reference has been made in Chapter II, seems especially to bear witness to the same process of development from G to F and from F to E. In the light of what has already been seen of the three versions, does it not seem probable that Gaimar reported the points of the tale unchanged, when he noted that after Grim died Havelok left Grimsby and appeared at Edelsi's court ^ The author of the Lai d'Have/oCy desiring a more chivalrous scene, and one more in keeping with the traditional departure of a knight, might easily have introduced the episode, and in so doing made Have- lok leave Grimsby during the lifetime of Grim. Grim called Havelok, gave him counsel and presented him with clothes. This change is in accord with the spirit of the Lay. The writer of the English poem, however, composed his work for a differ- ent audience, and the tone of this episode in the English Lay is therefore altogether different. Although the courtly, refined traits of F are lost or coarsened in E, the general outlines of the episode of E' are paralleled. Havelok leaves home during Grim's lifetime ; Grim calls him, gives him advice as he sends him to Lincoln to seek work, and makes him clothes. May not the passage of F perhaps in a later and much modified form have suggested the scene developed in E^ What conclusions does our detailed study of the versions lead us to P Are the slight and rather casual resemblances be- tween G and E, and the fact that they both begin with the English part of the story, sufficient evidence to show that G, F, and E emanated from a common source? Or, on the other hand, are the more numerous and often more definite resem- 56 THE OLD FRENCH LAI D'HAVELOC blances between F and E, the closer correspondence of their stories in general outline, and the seemingly transitional char- acter of F when considered with G and E in certain episodes, enough to show influence of F on E, whether that influence was exerted directly or indirectly, late or early in the develop- ment of E ? A number of possibilities present themselves for investiga- tion, such as the following: Was Kupferschmidt*s supposition the correct one after all, in spite of the fact that he did not off^er adequate proof of it? Was, on the other hand, the French Lay directly or indirectly the source of E ? Was there perhaps a later work related to the Lai d'Haveloc as the romance of Galeran to the Lay of Le Fraisne ? and did the author of E know it ? Or, could a chronicle account which contained the story in which traits of G and F were already blended have served as the basis of £? Fortunately our study does not have to wait for the solution of the question of the source of the English Lay. Unless proof is offered concerning the source of £, we are no longer ham- pered in the consideration of the relationship of F' and G, and the problem of the origin and relationship of the various ver- sions of Havelok may be approached as well from the French as from the English side. Moreover, if definite knowledge of the sources of F and G can be gained, the problems of the English Lay will be simplified. We are now ready to take up the discussion at the point where Kupferschmidt faced the alternatives : did F and G have a common source, or did F come from G? But, no longer bound by the consideration of the English Lay, we are free to take up the investigation of his rejected alternative. Could Gaimar's account have been used by the author of the French Lay? IV A Comparison of the French Lay and Gaimar's Version We have already seen in Chapter I that a certain number of scholars failed to follow the majority in the adoption of Kup- ferschmidt's theory. A few of them, it will be remembered, expressed with more or less conviction their opinion that the French Lay might have been derived from Gaimar's version. Ward alone gave his reasons for this view, and these were set aside by Putnam. There has been, as far as I know, no attempt to make a line by line comparison of Gaimar's account with the French Lay, although the existence of nearly identical lines in these two versions has been pointed out in some detail by Kup- ferschmidt and others. But the relative number of such lines, the kinds of resemblances they contain, and the fact that they occur singly or in long passages, should give evidence quite as important as that offered by the differences existing between the two works, — differences which up to the present have been much stressed. Therefore the first step in considering the French Lay and Gaimar's story is such a comparison. The following pages will contain the text of the Lay, each line of which will be accompanied by a corresponding line of Gaimar's text if one is to be found. Passages from the latter poem which contain a mere suggestion, or occasionally an interesting point of difference, will be given in parentheses. It is much to be regretted that no critical text of either Gaimar or the Lay exists. The best edition of Gaimar, on which this comparison is based, is that of Hardy and Martin ( 1 888-1 889), which was criticized severely by Paul Meyer, who said that Sir Thomas Hardy, although deeply versed in the sources of English history and experienced in Latin texts, "etait evidemment peu familier avec I'ancien fran9ais, et sur- tout avec la critique qu'il convient d'appliquer aux textes ecrits en cette langue." ' ^ Rom. XVIII, pp. 314-318. Cf. p. 315. Paul Meyer comments on the editor's inexact manner of denoting omissions and additions in the other manu- 58 THE OLD FRENCH LAI D'HAVELOC The text of the Lay is also from the Hardy- Martin edition of Gaimar (p. 290). The editor published Ms. //^ and gave for the first time variants from the older Ms. P. These vari- ants are not entirely satisfactory, because they are evidently indicated no more carefully than are the variants of Gaimar's text. On page 10 this statement was made: "Thus far every variation of spelling in Mss. D and L is noticed. It will not be so in the future." In the Lay passages of from six to ten lines are often found with no variant given,^ and some mechan- ical errors are also patent.'* Yet notwithstanding the lack of critical texts, it will probably be possible to draw enough evidence from the comparison of the Lay and Gaimar's version in their present condition to throw some light on the question of their relationship. Do they show the marks of a common source, or does the Lay clearly point to Gaimar as its starting point? scripts, when giving variants ; and he criticizes especially the way in which the text is composed. The editor chose Ms. M of the British Museum for his edi- tion, retained all its defective readings, and gave the correct readings merely among the variants from the other Mss., D and Z, thereby leaving the reader to form his own text. The method is illogical since, when any lacuna occurs in Ms. M, a word, a hne, or a whole passage is inserted from Mss. D and L with different spelling, and with indications not definite enough to show the beginning and end of the lacuna. In concluding his remarks on the text, Paul Meyer adds : **On comprend que je ne puis songer a proposer des ameliorations a un texte qui reste entierement a faire " (p. 317). 2 See chapter I, p. 6, n. 3, for manuscripts. Ms. H is in the Herald's Col- lege ; Ms. P at Cheltenham. 3 Cf. IO7O-IO77, 410-420, 836-843, 930-938. 4 Variant in line 43 reads Tels] tieus P, which would indicate that H reads Tels, P tieus, but H in the text reads Tieus ; 161 atendre] entendre P, al- though H has entendre; 271 tutjomitted L (?) ; 294 should be numbered 293 oit] out, and qe cil] quil, which follows without number, indicating, according to the system used, that the line is also 294, probably refers to 293 ; 295 re- queste]requist P, should be 294; 323 troueroie should be 324; 370 Li]Si, P should be 369; 553 should be 552; 554 should be 553 ; 581 has luijceo P, where ceo stands in the text, and the third part of 1082 should be 1084. In giving the variants I have corrected errors in numbering the lines, in the other cases I have omitted the readings entirely. A COMPARISON OF THE FRENCH VERSIONS 59 THE LAI D'HAVELOC 24 De son piere dirrai auant. 25 Gunter out non,«si fut Danois, 26 La terre tint, si estoit rois, 27 En icel tens qe Arthur regna, 28 Vers Danemarche mer passa, 29 La terre vout souz mettre a soi 30 Et le treu auer del roi ; 3 I Au roi Gunter se combati 32 Et as Danois, sis venqui. 33 Li rois meismes i fut occis, 34 Et plusurs autres del pais. 3 5 Hodulf loccist par traison, 36 Qui touz iors out le quor felon. 3 7 Quant Arthur out finie sa guerre, 38 Hodulf dona tote la terre 39 Et les homages des barons, 40 Puis sen ala od ses Bretons ; 7 41 Qe par destreit, qe par poour. GAIMAR 400 Vus fustes fiz a un bon rei. 403 Li vostre pere out nun Gunter ; s 401 Danemarche out par heritage, 409 Mes CO avint en vostre terre, 410 Li reis Arthur, la vint conqueere. 41 1 Pur sun treu, que li detint ; 413 Al rei Gunter semblad contraire; 414 Juste la mer li tint bataille. 41 5 Oscis i fu li reis Gunter, 416 E dambes parz maint chevaler : 513 Pur son seignur, (Gunter), kil (Hodulf) aveit mort, 517 Pur CO kil (Hodulf) ert traitres e fel,6 417 Ki Arthur volt, dona la terre. 523 Cist reis (Hodulf) ki done ert el pais. 511 Ki done ert reis poistifs Variants of Ms. P 25 Danois) Daneis. 28 Vers) En. Danemarche) Dene- ma rche. 29 vout) volt, souz) omitted. 32 sis) tuz les. 33 occis) ossis. 34 autres) altres. 36 quor) quer. 4 1 destreit) destresce. qe) tant. Note. — To avoid confusion I omit the brackets used in the Hardy-Martin edi- tion to indicate forms supplied in the text of Gaimar from Mss. D and L to fill lacunae in Ms. M. Wherever variants of Ms. P of the Lay are closer to Gaimar than the readings of Ms. Hy they are added to the reading of the text, and in the same way such variants of Mss. D and L of Gaimar as resemble the Lay more nearly are included in the text of Ms. M. The lines of the Lay which are practically identical with Gaimar' s are italicized in my text. 5 The Lay omits details given in Gaimar. The queen is Alvive, the daughter of Gaifer. 6 Gaimar adds that Hodulf killed Gunter (514) : Par la vertu de Artur le fort ; Kil out par treison mande, E eel pais li out done. The author of the Lay does not explain how he killed him. 7 The author of the Lay alone mentions the Bretons (40). Gaimar's account leaves Arthur's departure to be inferred from his investing Hodulf with power. s Gaimar states this fact when commenting on Sygar and his adherents, who held aloof from Hodulf. 6o THE OLD FRENCH LAI D'HAVELOC THE LAI D HAVELOC 42 Hodulf seruirent li plusour. 43 Tieus iout, li quistrent mal, 44 Par le consail Sigar Lestal, 45 Qui prodome fut et riche bier, 46 Et bien sauoit guerroier. 47 Cil auoit le corn a garder,9 48 Qe nuls horns ne pout soner. 49 Si dreit heir ne fust del lignage 50 Sur les Danois par heritage. 5 1 Einz qe li rois Arthurs venist, 52 Ne od les Danois se combatist, 53 Gunter auoit vn soen chastel 54 Sus la marine, fort et bel ; 5 5 De viande estoit bien garniz ; 56 Dedenz mist sa femme et son fiz ; 57 A vn baron de la contree 58 En ad la garde comandee. 59 Grim ^° out non, mult le crei, 60 Leaument lout touz tens serui. 61 Sur totes riens li comanda 62 Son fiz, quil forment ama ; 63 Qe si de lui mesauenoit, 64 En bataille sil morroit. GAIMAR 512 Sur Taltre gent de eel pais, 505 Hoc maneit uns riches hom, D L Lestarle 506 Sigar Estalre aveit nun : 507 Seneschal fu al rei Gunter, 508 E de sa terre justiser. 509 Maes ore ert tels ken peis teneit ; 510 E icel riche rei forment haeit, 5 1 8 Plus en unt tenu conseil 519 Ke ja od li ne se tendrunt, 683 Cel corn aveit Sygar guarde, 684 Li reis Gunter li out livere. 677 Si ke nuls ja corner loist, Dne\ 678 Si rei u dreit air le feist : 679 De Denemarche le dreit air 680 Le pot ben soner, pur vair; 579 (Un mariner, ki Grim out nun,) Fariants of Ms. P 44 consail) conseil. 45 riche bier) riches bcr. 46 guerroier) guere mener. 48 pout) poeit. 49 dreit) dreiz. fust) fut. lignage) li- nage. 51 Arthurs) Arthur. 53 Gunter) Gunters. 54 Sus) Sur. 60 touz tens) tut dis. 61 totes riens) tote ren. 64 sil morroit) ou il irreit. 9 The horn and flame are not anticipated in a description by Gaimar, but his facts in regard to them are the same when they are brought into the story. ^°Grim's part is changed. He is a baron entrusted with the care of the queen and Havelok. He protects them in a castle, and prepares a boat for flight as Hodulf becomes more aggressive. In G the queen fears and flees with Havelok to the mariner Grim, who saves them in his boat. Hodulf is emphasized more than in Gaimar. A COMPARISON OF THE FRENCH VERSIONS 6i 70 71 72 73 THE LAI D HAVELOC 65 Qa son poeir le garantist 66 Et fors del pais le meist, 61 Quil ni fust ne pris ne trouez, 68 Na ses enemis liuerez 69 Li emfes nestoit gaires granz, P dous Nauoit mie plus de vii. anz ; Totes les houres qil dormoit Vne flambe de lui issoit. Par la bouche li venoit fors ; 74 Si grant chalur auoit el cors. 75 La flambe rendoit tiel odour, 76 One ne sen tit nul home meillour." 77 A grant merueille le tenoient 78 Cil de la terre qui la veoient. 79 Puis qe li rois Counter fut morz, 80 Et ses barons et son efforz, 81 Hodulf chai et dechaca 82 Tuz ceus qil sout qil ama. 83 La reyne grant poour out, 84 Et li prodoms qi la gardout, 85 Que le chastel sus eus preist, 86 Et le fiz le roi occeist. 87 Nont mie force a eus defendre, 88 Autre consail lur estoet prendre. ^^ 89 Grim fet niefs apparailler, 90 Et de viande bien charger, 91 Fors del pais sen uout fuir 580 Men menat petit valetun. 625 Si li membrat 627 De la flambe ki ert issant 628 De sa buche, quant ert dormant. 418 Meis la reine, pur la guere, 419 Ne pout en la terre remancir; 420 Si sen fui od ledreit air. 423 (Mis pere aveit mult bon nef ;) Variants of Ms. P 66 fors) hors. le meist) sen fuist. 80 ses barons) si barun. son) ses. 67 ni) ne. 81 chai)hai. 69 gaires) gueres. 85 le) lur. preist) preisseist. 70 vij) dous. 86 fiz) fiiz. occeist) esteit. 7 1 dormoit) dormeit. 87 Nont mei) Nen mie. A eus) pur 73 Par la bouche) De sa buche. els. 74 el) al. 88 estoet) estut. 75-76 Omitted. 89 niefs) sa nef apparailler) apareiler. 78 qui la veoient) kil saueient. 91 uout) uult. " It is interesting to note that these additional attributes of the flame are not found in the older manuscript of F. This may have been through an oversight, but it is also possible that these lines were inserted later. '2 The description of preparations for flight and of the embarcation belongs entirely to F. One line ^42 9) of G, however, refers to the chevaliers et nostre gent J so that the queen was not unaccompanied. 62 THE OLD FRENCH LAI D'HAVELOC 92 93 94 95 96 THE LAI D HAVELOC Pur le droit hoir de mort garrir ; La reyne merra od soi. Pur la doute del felon roi. Qui occis auoit son seignur ; Tost feroit a li deshonur. 97 Quant sa nief fut apparaillee, 98 Dedenz fist entrer sa meisnee, P chamberlencs 99 Ses cheualers et ses serganz, 1 00 Sa femme demeine et ses enfanz ; 1 01 La reyne mist el batel, Haueloc tint souz son mantel ; II meismes apres entra A Dieu del del se comanda. Del hauene sont desancre. Car il eurcnt bon orre ; Le trauers eurent de la mier, 108 Mes ne sieuent queu part aler ^3 109 Ou garder pussent lur seignur. 1 1 o Malement lur auint le iour ; ^4 Car outlaghes les encontrerenty Qui hautement les escrierent. Mult durement les assaillirent, Et cil forment se defendirent ; Mes il eurent poi desforz. La nefunt robe e ma I misey E la Reine i fu ossise. 1 16 Li outlaghe les ont touz morz. 102 103 104 105 106 107 1 1 1 1 12 "3 114 P 1 1 7 Ni remist nul petit ne grant 424 (La raine amenout sucf :) 425 (Vers cest pais lamenout,) 426 *' Quant si avinty cum Deu plout, 427 " De utlagles fumes encontrez: 438 (Nostre nef) D L frainte e malmise 439 Car tute fufreitey e malveise, 440 ^ant la raine fu oscise. 428 En mer furent trestuz ruez 429 Nos chevalers, e nostre gent, 43 I Unc ne guari horsxyfors mun pere. 92 garrir) guarir. 93 merra) mettra. 95 Q.ui)Ke. 97 apparaillee) aparile. 99 cheualers) chamberlencs. 100 demeine) ameine. 104 Dieu) Deu. se) les. 105 sont) se sunt. 106 eurent) aueient. orre) ore. Variants of Ms. P 107 eurent) eurent. 108 Mes il ne seuent quel part turner. 1 10 iour) ior. 1 1 1 Outlaghes) Utlages. 112 hautement) laidement. 1 14 forment) de la nef. 1 1 5 P adds : La nef unt robe e mal mise, E la Reine i fu ossise. 1 17 grant) granz. ^3 In F Grim is represented as sailing at random. In G they may have directed their flight toward Lindsey. ^4 The details of the voyage and of the attack of the pirates are strikingly similar. The only difference is the emphasis placed by the author of the Lay on the valor of the queen's party and the impossibility of escape from the pirates. A COMPARISON OF THE FRENCH VERSIONS 63 THE LAI D HAVELOC 1 8 Fors Grim^ qui ert lur conoissant, 1 9 Sa femme et ses enfanx petiz, 20 Et Haueloc i est garriz. 121 122 123 124 125 126 127 128 129 130 132 133 134 135 Puis qe de eus furetit eschape, Tant ont nage et tant sigle Qen vne hauene sont paruenu, Et de la nief a terre issu. Ceo fut el North, a Grimesbi.^s A icel tens qe ieo vus di, Ni out onques home habite, Ne cele hauene nert pas haunte. II i adresca primes maison ; De lui ad Grimesbi a non. Quant Grim primes i ariua. En a. moitez sa nief trencha, Les chiefs en ad amont dresce ; Iloec dedenz sest herberge. Pescher aloit si com il soloit,^^ 136 5/V/ vendoit et achatoit, 1 3 7 Tant qil fut iloec bien seu, 138 Et des paisanz bien coneu. 139 Plusurs a li sacompaignerent, 1 40 Sus le hauene se herbergerent ; 141 Pur son non, qil eurent oi, 142 Le liu appellerent Grimesbi. GAIMAR 433 Mis per e est ait lur c onus s ant ; 432 Ne nule femme y fors ma mere. 434 Pur CO guarirent li enfant y 435 ^ jOy ^ ^^^i ^ ^^ duifrere, 436 Par la priere de mun pere. 495 ( Tant unt nage e gov erne y^ 437 En cest pais quant arivames, 438 Nostre grant nef par mi tren- ch am es ; 44 1 De nostre nef meison feimes : 442 Par un batel ben guarisimes, 443 Dunt nostre pere ala pescher. 380 Grim vendi sely si fu peschere. 118 121 122 123 124 125 126 127 128 Variants of Ms. P conoissant) conisanz. furent) fu. eschape) eschapee. nage) nagee. sigle) siglee. Qen) Ke a. nief) nef. Ceo) Ce. el) en. A) Mes a. qe) dunt. onques) vnques. nert pas) naent. 29 adresca) dresca. 132 moitez) meites. 135 aloit) alad. soloit) sout. 136 Siel)Sel. 137 seu) seuz. 138 coneu) coneuz. 140 Sus) Sur. 141 eurent) orent. 142 appellerent) apellent. ^5 All the information about Grimsby, the harbor and the growth of the town is found only in F. ^^ This is a second failure to be consistent in the part of Baron which the author of F has attributed to the fisherman Grim. 64 THE OLD FRENCH LAI D'HAVELOC THE LAI D HAVELOC 143 Li prodoms son seignur nurrit,^7 144 Et sa femme bien le seruit. 145 Pur lur enfant tuz le tenoient, 146 Car autre chose ne sauoient. 147 Grim li out fet changer son non, 148 Qe par tant nel conuist lom.^^ 1 49 Li emfes creut et amenda ; ^9 I 50 De cors, de membres efForca ; 1 5 1 Einz quil eust gaires de ee 152 Ni trouast il home barbe, 1 5 3 Sencontre lui liuter vousist, 154 Qe li emfes ne labatist. 1 5 5 Mult fut forz et vertuous, I 56 Et enpernant et airous. 1 5 7 A merueille sen esioit 2° I 5 8 Grim le prodome, qi le nurrit ; I 59 Mes de ceo out le queor dolent 1 60 Quil nert nurri entre tiele gent 161 Ou il puist auqes entendre (li prodom) 588 Ki me nuri, e mult mama : 589 II e sa femme me nurirent, 590 E mult me nurirent e cherirent. 1 1 3 Naveit valet en la meison, 1 1 4 Si lui feseit ahataison 115 E sur lui comencast mellees, D L ruast 1160 nel rueit jambes levees : 1 1 1 Mes pur CO que hardi estait, 117 (E quant il ben se corucout) Variants of Ms. P 143 prodoms) prodome. nurrit) norri. 155 fut) par ert. 144 bien le seruit) mult le cheri. 145 enfant) enfanz. le) les. 153 liuter) luter. vousist) volsist. 158 nurrit) noriseit. 159 queor) quer. 160 nurri) noriz. tiele) tel. ^7 These lines (143-144) also resemble those of G (453-455 ff-)* '^*°^ cum vesqui, il e ma mere Bien vus nurrit, mielz que mi frere. In F seruit shows additional respect for Havelok as the heir to the throne. ^8 Gaimar does not comment on the concealment of Havelok* s real name. It may be considered as implied, however, since he did not know his name until told it by Kelloc (421-422). Co estes vus, si cum jo crei, Danz Haveloc, le fiz le rei. That he was considered one of Grim's family and considered himself as such is shown by lines 365-376, and especially by line 367 : Tu es ma suer, jo sui tis frere, Ambure de pere e de mere. Grim fud mis pere, un peschur, Ma mere ot nun Sebrug, sa uxor, etc. ==9 Gaimar' s description of Havelok follows the mention of his arrival at court. Physical strength and valor are emphasized in it also, but the whole conception is more rudely expressed. F contains another description of Havelok at court cor- responding more closely to Gaimar' s. 2° The following account, which explains Havelok's departure from Grimsby, belongs to /^ only. Gaimar simply states that after Grim died Havelok left the place. A COMPARISON OF THE FRENCH VERSIONS 65 THE LAI d'haVELOC GAIMAR 162 Et afetement apendre ; 163 Car il quidoit en son corage 164 Qe encore aueroit son heritage. 165 Grim lappella vn iour a soi ; 166 **Beaufiz," fet il/'entend amoi; 167 **Ici manom mult sou tiuement, 168 ** Od pescheours, od poure gent, 169 **Qui se garrissent de pescher; 1 70 ** Tu ne siez rien de eel mester ; 171 **Ici ne poez sauer nul bien, 172 **Tu ni gaigneras ia rien. 173 ** Va ten, beau fiz, en Engleterre 174 ** Aprendre sens et auoir querre ; 175 **Tes freres meine ensemble od toi; 176 **En la curt a vn riche roi 177 * *Te met,beau fiz, souz les serganz. 178 *«Tu es forz, parcreuz, et granz, 179 **Si porras grant fes porter. 180 ** A tote gent te fai amer, 181 "Si tabandoune del seruir 182 ** Quant tu porras en liu venir; 183 *«Et Dieu te dount si espleiter 184 **Qe auques i puissez gaigner." 185 Quant li prodoms lout enseigne, 186 Et de draps apparaille, 187 De lui le fist partir a peine. 188 Les ii. valez od li ameine.^^ 189 Tuit troi quidoient estre frere, 1 90 Si come lur auoit dit lur piere. 191 Tant ont le droit chemin tenu, 192 Quil sont a Nichole venu. 193 A ICEL tens qe ieo vus di,22 Variants of Ms. P 162 Afaitement e sens aprendre. 174 auoir) auer. 164 heritage) corage. . 177 beau) beus. serganz) seruanz. 166 beau) Beu. 178 Tu es mult fors e cruz e granz. 168 pescheours) pechurs. 179 porras) poras ben. 169 de) par. 181 del) de. 170 siezj sez. eel) lur. 186 draps) noues dras. 1 7 1 poez) poez tu. 1 89 Tuit troi) Tuz treiz. 172 Ne ia ne guainerez ren. 191 le) lur. 2^ The only resemblance to Gaimar's account in the whole passage is the fact that Havelok takes the two foster brothers with him. In F Grim advises him to do so. 22 Gaimar mentions Adelbrit first, and Li altres refers to Edelsi (49). In F Alsi comes first. 66 THE OLD FRENCH LAI D'HAVELOC THE LAI D HAVELOC 194 Vn roi qert nome Alsi 195 Tenoit en la terre en sa baillie, 196 Nicole et tote Lindesie, 197 Cele partie vers le north ; 198 Et Rotelande et Stanford's 1 99 Out cil Alsi en heritage ; 200 Mes il estoit Bret par lignage.24 201 Le roiaume vers les Surois'S 202 Gouernoit vns autres rois ; '6 203 Ekenbright out cil rois a non, 204 Mult out en lui noble baron. 205 II out la sorour Alsi 206 (Compaignon furent et ami), 207 Orewen, vne dame vaillant ; 208 Mes entre eus neurent enfant 209 Mes qe vne fille bele ; 210 Argentine out non la pucelc. 21 1 Rois Ekenbright fiat enfermez,^? 212 Et de grant mal forment greuez. 2 1 3 Bien siet nen poet garrir ; GAIMAR D L Edelsi 49 Li altres out nun Edelsie ; D L Lindesi 50 Sue ert Nicole e Lindeseie, 51 52 61 62 53 Des Humbre desken Roteland Ert le pais en son comant : Li altre rei estait Breton, Ki Edelsi aveit a nun : Li altre ert reis de la contree 54 Ki ore est Nortfolc apelee. D L Achebrit ot 47 Adelbrit aveit a nun li uns des reis ; 48 Riches horn fu, si ert Daneis : 55 Edelsi dona sa sorur D L Achebrit 59 A Adelbrit, eel riche reis, 63 Sa sorur out nun Orwain ; 64 Mult ert franche, e de bone main. 65 66 79 De son seignur out une fille, Ke lom apela Argentille. D Achebricht, L Achebrit Mes done avint ke Adelbrict fu mort. Variants of Ms, P 1 94 AlsiJ Edelsi. 197 Cele)E le. 198 Stanford) Estanfort. 199 cil Alsi) ausi. 200 Bret) Brez. 201 roiaume) realme. Surois) Suriens. 202 Gouernoit) Gouernout ores. 203 Ekenbright) E Sehebrit. 207 Orewen) Orwein. 208 entre eus neurent) il naueient nul. 209 Fors vne sule file bele. 210 pucele) damaisele. 2 1 1 Echebrit chai en enfermete. 212 forment) fu mult, greuez) greue. 213 poet) pora. «3 A difference in geographical detail is to be noted, as well as the statement that Alsi held his kingdom en heritage. 24 In G Edelsi is BretuUy as Adelbrit is a Dane. In F he is Bret par lignage. 25 This geographical detail varies in the two accounts. 26 The author of F does not mention that Ekenbright is a Dane. 27 This is an important difference. Gaimar states merely that Adelbrit died. The author of F represents him as about to die and sending for Alsi, to whom he confides the queen, the kingdom, and Argentille with conditions which Alsi binds himself by oath to fulfill. A COMPARISON OF THE FRENCH VERSIONS (^1 214 215 216 217 218 219 220 THE LAI D HAVELOC Alsi fet a lui venir, Sa fille li ad comandee Et sa terre tote liueree. Primerement li fet iurer, Veiant sa gent, et affier, Qe leaument la nurrireit, Et sa terre lui gardereit, 2 2 1 Tant qele fust de tiel age 222 Que sufFrir porroit mariage ; 223 Quant la pucele seit granz, 224 Par le consail de ses tenanz, 225 Au plus fort home la dorroit 226 Qe el reaume troueroit ; Ms.P (Ken la terre trouer poreit.) 227 Quil li baillast ses citez, P (Pus li bailla les fermetez,) 228 Ses chasteus et ses fermetez, P (Les chasteuz e les citez,) 229 Sa niece en garde et sa sorour, 230 Et tuz les homes de lonur. Mes la reyne enmaladit ; P Echebrit Puis qe Ekenbright finit, Hastiuement refut finie, 234 Lez son seignur fut enfouie.^s 235 De EUS estoet ore ci lesser. 236 De Haueloc voil auant conter. 237 Rois Alsi, qui done regna 238 Et les ii. regnes gouerna, 239 Bone curt tint et grant gent; 240 A Nicole manoit souent. 241 Cil Haueloc a sa curt vint 242 Et vn keu le roi le retint. 231 232 233 89 Car la raine ert enfermee, 90 Ne mais vint jurs ad duree L Achebrit 9 1 Apres Albrict : quant fu finie, 92 Unt la raine ensepelie ; 153 II ert issi, en la meison. Variants of Ms. P 215 fille) nece. 217 Primerement) Premerement. li) le. 221 tiel) tele. 222 porroit) pout. 223 seit) serreit. 225 dorroit) dureit. 226-228 As in the text. 232 Ekenbright) Echebrit. 235 ci) omitted, lesser) laisser. 236 conter) traiter. 239 et) mult ot. 242 Et vn keu) Vn des ecus. 28 According to G, the queen goes to her brother after Adelbrit's death, and puts the kingdom into his charge before her own death. In F, since Adel- bright had already taken this step, it is stated merely that the queen died soon after her husband. 68 THE OLD FRENCH LAI D'HAVELOC THE LAI D HAVELOC 243 Purceo qe fort le vist et grant, 244 Et mult le vist de bon semblant. 245 Merueillous fes poeit leuer, 246 Busche tailler, ewe porter. 247 Les esquieles receuoit, 248 Et apres manger les lauoit ; 249 Et quantquil poeit purchacer 250 Piece de char ou pain enter, 2 5 1 Mult le donoit volentiers, 252 As valez et as esquiers, 253 Tant estoit franc et deboneire, 254 Que tuz voloit lur pleisir fere. 255 Pur la franchise qil out, 256 Entre eus le tenoient pur sot ; 257 De lui fesoient lur deduit,29 258 Cuaran lappelloient tuit ; 259 Car ceo tenoient li Breton 30 260 En lur language quistron. 261 Souent le menoient auant 262 Li cheualer et li sergant, 263 Pur la force qen li fu ; 264 Desquil seurent sa grant vertu, 265 Deuant eus liuter le fesoient 266 As plus forz homes qil sauoient, 267 Et il trestouz les abatit ; 268 Et si nuls de eus le mesdeisist, 269 Par dreite force le lioit : 270 Tant le tenoit et iustisoit GAIMAR 154 Esqueler a une quistron. 106 Mes mult par ert bel valetun. 136 Feseit sovent mult larges dons, 135 E as vallez de la meisons. 1 1 1 112 113 114 115 116 117 118 119 120 121 Mes pur co que hardi estait, E volunters se combateit, Naveit valet en la meison. Si lui feseit ahataison E sur lui comencast mellees, Kil nel rueit jambes levees : E quant il ben se corucout, De sa ceinture le liout: E si cil done naveit guarant, Bien le bateit a un vergant. E nepurhoc tant frans esteit. Variants of Ms. P 244 le vist) esteit. bon) bel. 245 Merueillous) Merueiles. 247 receuoit) receiueit. 248 les lauoit) de co serueit. 251 Mult par les dona volunters. 254 Que) A. 255 qil) quen li. 256 pur) a. 258 Cuaran) Curant. 260 quistron) pur quistrun. 263 fu) feu. 264 Des) Pus. seurent) sorent. 29 Havelok's position at court is made more humiliating by the author of F. Gaimar's Havelok is generous and beloved by his associates. In F his knightly generosity and amiability are scorned by his associates. His duties are enumer- ated in F, but not in G. 30 This allusion to the Bretons is found only in F. A COMPARISON OF THE FRENCH VERSIONS 69 THE LAI D HAVELOC 271 Quil li auoit tut pardone, 272 Et quil restoient acorde.31 P Li Reis memes mut souent Le fit luter deuant sa gent, 273 Li rois forment sesmerueilloit 274 De la force qen lui veoit. 275 Dis des plus forz de sa meson 276 Neurent vers li nule fuison ; 277 XII. homes ne poeient leuer 278 Le fes que il poeit porter 279 En la curt fut lungement, 280 Dici qun assemblement,32 281 Qe li baron a la curt vindrent 282 Qui de Ekenbright lur terre tind- rent ; 283 Et lors tenoient de Argentille 284 La meschine, quert sa fille, 285 Que ia estoit creue et grant, 286 Et bien poeit auoir enfant. 287 Le roi en ont a raison mis, Et de sa niece lont requis. Qa tiel home la mariast Qis meintenist et conseillast, Et si gardast son serement GAIMAR D LYi 122 Si lui vallez li prometteit 123 Ke pur ico mains nel amast, 124 Ignelure le deliast. I 2 5 Quant il se erent entrebaisez, 126 Done estait Cuharan haitez; 166 De lui son jugleur feseit. 2»» 289 290 291 Variants of Ms. P restoient) esteient. Li Reis memes mut souent, Le fit luter deuant sa gent, A grant merueille le teneit De la force ken lui aueit. Nen aueient a lui fiiisun. 277 poeient) poreient. 278 que il poeit) quil sul soleit. 272 273 276 279 fut) fu ben. 280 Dici qun) Deska un. 283 lors) ore. 284 quert) qui fu. 285-286 in P follow 287-288. 285 creue) cruz. grant) granz. 286 enfant) enfanz. 290 Qis) Ke les. 31 The accounts of the wrestling contests of Havelok in F and G differ in tone and detail but represent the same events. In 269 of F lioit corresponds to Gaimar's De sa ceinture le liout. In F the knights know of Havelok and have him wrestle for them. Ms. P, however, is closer to G. The king himself has Havelok wrestle for him. In G, too, the knights know of Havelok, for they make him presents (i 41-142). 32 In F there is a special assembly of the barons in behalf of Argentille, at Alsi's court. The description of this scene between Alsi and the barons is found only in F (279-376). 70 THE OLD FRENCH LAI D'HAVELOC THE LAI d'haVELOC GAIMAR 292 Quil sen aquitast leaument. 293 Li rois oit qe cil disoient, 294 Et la requeste qe cil fesoient ; 295 Vn respit lur en demanda, 296 Et dist quil sen conseillera ; 297 Sauer voudra et demander 298 A qui il la porra doner. 299 Terme lur mist et ior noma, 300 A repairer les comanda 301 Quant il se serra conseillez ; 302 Et il si flit mult veziez. 303 A ses priuez en ad parle. 304 Et son corage tut demustre ; 305 Consail lur quist et demanda, 306 De ceus qi requeroient ia 307 Qa sa niece donast seignur 308 Qjs maintenist a honur; 309 Mes il vout mielz sufFrir lur guerre 3 I o Qil ne soit dessaisi de la terre. 3 1 1 Ceo li dient si conseiller : — 312 ** Fetes la loignz enmener 313 ** En Bretaigne, dela la mer, 314 ** Et a vos parenz comander ; 315 **Nonaine seit en vne abbeie, 316 **Si serue Dieu tote sa vie," 317 "Seignurs, tut el enpense ai, 318 **Tut altrement men deliuerai. 3 19 **Rois Ekenbright, quant il fina 320 ** Et sa fille me comanda, 321 **Vn serement me fist iurer, 322 ** Veianz vus touz, et affier, 323 * * Que au plus fort home la dorroie 324 **Que en la terre troueroie. 325 ** Leaument me pus acquiter ; Variants of Ms. P 292 Quil sen aquitast) Ke il les tenist. 312 enmener) en veier. 293 oit) out. 313 Bretaigne) Bretaine. 294 qe cil) quil. requeste) requist. 314 Et) Sil faites. 296 dist quil) pus il. 317 tut el) fet il. 299 et ior) ior lur. 318 Tut) Ke. 301 se) omitted. 319 Quant Echebrit li Reis fina. 302 Et il si) E issi. veziez) ueissez. 320 fille) terre. 304 tut) lurad. demustre) mustre. 322 vus touz) sa gent. 308 Qis) Ke. a honur) eus e lonur. 323 Que au) Cil al. dorroie) dureit. 309 Mes il uoleit meuz sufFrir guere. 324 troueroie) trouereit. 3 1 o Qil ne soit dessaisi) Ke dessaisi estre A COMPARISON OF THE FRENCH VERSIONS THE LAI d'hAVELOC GAIMAR 326 **A Cuaran la voil doner, 327 **Celui quest en ma cuisine ; 328 **De chauderes serra reyne. 329 ** Quant li baron repaireront, 330 **Et la requeste me feront, 331 ** Oianz touz lur voil mustrer 332 * * Que a mon quistron la voil doner, 333 ** Qu^ fort est et de grant vertu, 334 **Ceo sieuent cil qi lont veu. 335 **Sil ni ad qui le contredie, 336 **Ne qi le mattourt a vileinie, 337 **Dedenz ma prison le mettrai, 338 **Et au quistron cele dorrai." 339 ENSI ad li rois diuise. 34.0 Au ior qil out a ceus nome, 341 Apparailla de ses priuez 342 En sa chambre vii. vinz armez ; 33 343 Car il quidoit auer mellee, 344 La ou ele serroit esposee. 345 A la curt vindrent li baron ; 346 Li rois lur mustra sa raison ; 347 * * Seignurs, " fet il, * * or mescotez, 348 '* Puis qe ci estes assemblez ; 349 ** Vne requeste me feistes 350 **Lautrer, quant a moi venistes, 351 ** Qa ma niece seignur donasse 352 ** Et sa terre li otriasse. 353 ** Vus sauez bien, et ieo le vus di, 354 ** Quant Ekenbright le roi fini, 355 **En ma garde sa fille mist, 356 ** Vn serement iurer me fist 357 ** Qs^ P^us fort home la dorroie 358 ** Qe el reaume trouer porroie. 359 **Assez ai quis et demande, 360 ** Tant qen ai vn fort troue. Variants of Ms. P 326-332 omitted. 349 requeste) queste. feistes) feites. 335 Sil ni ad) Se nul ia. 350 Lautrer) Lautre ior. 337 Dedenz) en. le) la. 351 donasse) donace. 338 cele) la. 352 otriasse) comandasse. 339 Ensi) Eissi. 358 reaume) realme. 344 ele) celle. esposee) donee. 360 qen) ore en. 347 mescotez) escutez. 33 The author of F evidently tries to defend his barons by showing the im- possibility of any resistance because of Alsi's precautions. 72 THE OLD FRENCH LAI D'HAVELOC THE LAI D HAVELOC 361 **rn valet at en ma quisine 362 **A qui ieo dorr at la me s chine. 363 **Cuaran ad cil a non. 364 **Li dis plus fort de ma maison 365 **Ne se poent a lui tenir, 366 "Son giu ne sa liute sufFrir. 367 "Veritez est, desqa Rome 368 "De corsage nad si grant home. 369 "Li garder voil mon serement, 370 "Ne la pus doner autrement.** 371 QUANT LI baron ont escute 372 Qui! out dite sa volente, 373 Entre eus dient en apert 374 Qe ceo nert ia par eus sufFert. 375 la ieust granz coups donez, 376 Quant il fet venir les armez. 377 Sa niece lur fet amener,34 378 Et a Cuaran esposer ; 379 Pur lui auiler et honir 380 La fist la nuit lex lui gisir. 381 Quant couche furent ambedui, 382 Cele out grant home de lui, 383 Et il assez greindre de li ; 384 As denz se geuty si se dor mi; 385 Ne voloit pas qele veist 386 La flambe qe de lui issist ; GAIMAR 174 Honist sa nece, a son espcir, 175 E la dona a son quistrun, 175 Ki Cuheran aveit a nun. 100 Sa nece mesmariat. 1 01 II la donat a un garcon, 102 Ki Cuheran aveit a nun : 103 Pur CO kabeisser la voleit. 167 Pur la terre Albrict tolir, 168 Feseit sa nece od lui gisir. \jj Cil ne saveit ke femme estait, 178 Ne kil fere li deveit : 179 Treskil unkes el lit veneit, 1 80 Adenz giseity si se dormeit. 181 Argentine ert en grant purpens; 182 Pur quel il giseit si a denz ; 183 E mult forment sesmerveillout, 184 Ke unkes vers lui ne se turnout, 185 Ne ne la voleit aprismer, 186 Com home deit fere sa muller. Variants of Ms. P 365 poent) poeient. 376 fet venir) demande. les) ses. 367 desqa) deci ka. 377 lur) omitted. 368 grant) fort. 380 La fist) Lad fet. 369 Li) Si. 381 ambedui) amdui. 372 volente) uolunte. 383 greindre ) greinur. 374 ceo) omitted. 384 As denz) En pels, si se) e si. 375 coups) copz. 386 La flambe qe) ke la flambe. 34 This passage contains the general details of Gaimar's description. As it is a good example of the different manner in which the two authors use their materia], nearly all of the passage of G is given, even where it differs from F. Occasion- ally there is remarkable similarity in the words used in the two passages. A COMPARISON OF THE FRENCH VERSIONS 73 THE LAI D HAVELOC 387 Mes puis sasseurerent tant, 388 Et par parole et par semblant, 389 Quil lama et od lui geut 390 Come od sespouse fere deut. 391 La nuit qe primes enparla, 392 Tiele ioie en out qil lama, P sen 393 Q^i^ s^ dormit et oblia ; 394 Enuers se geut, ne se garda ; 395 Et la meschine sendormi, 396 Son braz getta sus son ami. 397 Iceo li auint en auision /'od 398 Qele ert alee a son baron 35 399 Outre la mier en vn boscage. 400 La troeuent vn vrs sauuage ; 40 1 Goupilz auoit en sa compaigney 402 Tut fut couerte la champaigne ; 403 Cuaran voleient assaillir, 404 ^uant dautre part virent venir P porz 405 Chiens et senglers qui le defen- doient. 406 Et des goupilz mult occioient. 187 La nece al rei se compleigneit ; 188 Sovent son uncle maldissoit. 191 Tant kil avint a une nut, 192 Kil firent primes lur deduit. 193 Apres ico si sendormirent : 1 94 Mult sentreamerent, e joirent. 195 La fille al rei, en son dormant. 196 Songat kele ert, od Cuherant, 197 Entre la mer e un boscage y 198 U conversout un urs salvage, 203 Od lurs aveit as ex gopillXy 204 Ki puis le jur ourent perilz : D Cuaran 202 Ki 36 [lurs) voleit Cuheran manger. 199 Devers la mer veait venir 200 Pors e senglersy prist asaillir 201 Icel grant urs, ke si ert fier, 202 Ki voleit Cuheran manger. 206 Mult em>i {goupilz) destruistrenty e oscistrent. Variants of Ms. P 387 tant) i tant. 388 par, par) de, de. 389 geut) iust. 390 Cum il od sun espuse dust. 391 en)i. 392 Tiele) Tel. qil) e tant. 393 se) sen. 394 se garda) senturna. 396 getta) ieta. sus) sur. 398 a) ** a " in the text, but corrected to **od" in the margin of/*. 399 vn) omitted. 400 troeuent) trouerent. 402 Tut) Tote. 403 voleient) uoilent. 405 Chiens) porz. defendoient) de- fenderent. 406 mult occioient) le guaranterent. 35 These passages of G and F are noteworthy for the identical expression of whole lines. 36 Ki (202) of G refers to the bear. "il En (206) of G refers to the foxes. 74 THE OLD FRENCH LAI D'HAVELOC THE LAI D HAVELOC P vcncuz, 407 Quant It goupil furent venuy 408 Fn des senglers par grant vertu 409 Ala vers lours, si lenuait, 410 Iloeqes loccit et abatit. 411 Li goupil qi od li se tindrent, 412 Fers Coaran ensemble vindrent, 413 Deuant li se mistrent a terre, 414 Semblant Jirent de merci querre ; 415 Et Coaran les fist lier,'^^ 416 Puis vout a la mier repairer. 417 Mes li arbre qi el bois erent 418 De totes parz li en diner ent; 419 La mier crut et fiot monta 420 De si qa lui ; grant poour a. 42 1 Deus leons vist de grant fierte ; 422 Vers lui vindrent tut efFree, 42 3 Les bestes del bois deuoroient 424 Celes qen lur votes trouoient. 425 Coaran fut en grant effreiy 426 Plus pur sarnie qe pur sei ; 427 Sur vne halte arbre mont erent, 428 Pur les leons qil doterent ; 429 Mes li leon auant aloienty 430 Desouz larbre sagenuilloient. 207 Quant li gopi I furent destruit, 209 Un sul sengler, Jier e hardi, 2 I o Lad par son cors sul asailli. 208 Cel urs, ke demenout tel bruit, 2 1 1 Tel lui dona del une dent, 212 En dous meitez le quer li fent. : 3 Quant lurs se sent a mort feru, ; 4 Un cri geta, puis est chau : 5 E li gopil vindrent cor ant y D L Cuarant 6 De tutes parxy vers Cuherant, 7 Entre lur quisses lur cuetes, 2 1 8 Les chefs enclins, agenuletes ; 219 E funt semblant de merci quere. D L Her 221 Quant il les out feit tuz lever, 222 Envers la mer volt repairer. 223 Li grant arbre, ki el bois erent, 224 De totes parz lenclinerent. 225 La mer montout e li fioz vint, 226 De si kal bois ne se tint. 227 Li bois si chaeit, la mer veneit, D L Cuaran 228 Cuheran ert en grant destreit. 229 Apres veneient dous leons: 230 Si chaeient a genullons. 231 Mes des bestes mult oscieient 232 El bois, hi en lur veie estaient. 233 Cuheran, pur pour kil out, D L un grant 234 Sur un des granz arbres montout : 235 E les leons vindrent avant, D L larbre 236 Envers eel arbre, agenullant. 407 venu) vencuz. 408 vertu) vertuz. 409 Vers lui ala si len wai. 410 Mes Cuarant locist e uenqui. 420 Di CO ele grant pour a. 421 Kar dous liuns vit par grant ferte Variants of Ms. P 426 Plus) Tant. qe) cum. 427 monterent) andui munterent. 428 doterent) duterent. 429 aloient) alouent. 430 sagenuilloient) sagenulouent. 38 Lier (415) of F is found in Mss. D and L of G. Ms. P shows by 393, 398, 405, 407 that the passage was originally much closer to G than it is in Ms. H. A COMPARISON OF THE FRENCH VERSIONS 75 THE LAI D HAVELOC 431 Semblant li firent damour, 432 Et quil le tenoient a seignur. 433 Par tut le bois out si grant cri 434 Q^e Argentine sen esperi. 435 Mult out del sunge grant poour; 436 Puis out greindre de son seignur 437 Pur la jiambe qele choisit 438 Q^e de la bouche li is sit. 439 En sus se trest, et si cria 440 Si durement qe le esueilla: 441 ** Sire,'' fet ele, ** vus ardez. 442 **Lasse! tut estes allumez.** P trait 443 Cil le braca et estreinst vers soi ; 39 237 Par tut le bois out si grant cri, D L esperi 238 Ke la dame sen eveilli: 243 Pur la pour ses oilz overit, 244 Une Jiambe vit, ki issit Fors de la buche son marri. Ore entendez kele dit. Tant I'embrasca e trest vers sei, Kil sesveilla * * Sire, * ' fet ele, * * vus ardez : ** Esveillez vus si vus volez. ** De vostre buche une flambe ist : 245 250 256 251 252 444 445 446 447 448 449 450 451 452 453 ** Bele amie,''' fet il, ^*pur quoi **Estes vus issi efFree ? ** ^i vus ad issi espoentee /"' **Sire," fet ele, *Meo sungai; " Lauision vus conterai." Conte li ad et coneu, Del feu li dist qele ad veu Qui de sa bouche venoit fors, Ele quidoit qe tut son cors Fust allume, pur ceo cria. 255 Tant P embrasca e trest vers sei, 256 Kil sesveilla, e dist: "Purquei, 257 * * Pur quei mavez eveille bele amie : 258 * * Pur quei estes espontie ? ' ' 259 Tant la preia, e tant la blandist, 260 Kele li conta tut, et regehit 261 De la Jiambe, e del avision 262 Kele out veu de son baron. Variants of Ms, P 43 I firent) feseient. 432 quil) ke. 434 sen esperi) sessperi. 435 Del sunge ot grant pour. 436 greindre) plus. 438 li) sun seignur. 439 trest) traist. 440 le esueilla) sesueilla. 442 Alas ia tut estez alumez. 443 estreinst) trait. 445 issi) si. 446 espoentee) espunte. 448 Lauision) ma uisiun. 450 Tel fu li dit kele out veue. 453 allume) en lumme. 39 These descriptions are almost word for word the same, and the few touches that have made F vary in an occasional detail are very evidently- the author's own : cf. 426. Cuaran is not cowardly; his fear is more for Argentille ; lines 431 — 432 only elaborate the statement of 430; in 443, instead of Cuaran's being awakened by Argentille' s embraces, the author of jp represents him as startled by her cry of fright, whereupon he embraces and soothes her; lines 452-453 explain Argentine's cry of fright. Cf. 454 of F, reconforta, with Gaimar's respondi; cf 259 of G with 447-448 oi F in which Argentille begins to tell her dream without being urged. 1^ THE OLD FRENCH LAI D^HAVELOC THE LAI D HAVELOC 454 Cuaran la reconforta, P **Dame" 455 * * Beky ' ' fet ily * * ne dot ex rien ; 456 ** Cest bon au vostre vs et au mien. 457 **La vision qe auez veue 458 ** Demain poet estre conue. 459 ** Li rois doit safeste tenir, 460 ** Toux ses barons i fet venir. 461 *^Veneison i aver a as sex; P des granz lardez 462 ** leo dorr a i hastes et lardex 463 **As esquiers a grant plente, 464 **Et as valex qui mont ame, 465 **Li esquier sont li goupily 466 **Et li garcon qi sont plus vil ;4o 467 **Et li ours fut des hier occisy 468 ** Et en nostre quisine mis.41 469 **Deus tors fist hui le roi beiter, 470 *^Pur les leons les pus center; 471 **Les ploms poom mettre pur mier 472 **Dont le feu fet lewe monter. 473 **Dite vus at lauision ; 474 **Ne soiez mes en suspecion. 475 **Le feu qi ma bouche getta,42 476 ** Bien vus dirrai qui ceo serra ; 43 477 '* Nostre quisine ardera, ceo crei ; GAIMAR 263 Cuheran len respondi: 267 **Damey'^ dist il, ** co serra bien, 268 **Anbure a vostre oesy e al mien. 270 *^Li reis tendra demain safeste; 271 "Mult i aver a de ses barons. 272 "Cerfsy e cheverelsy e veneisons, 277 **Des bons lardex e de braunsy 276 **Les esquiers ferai manant 279 "Zr/ esquier me sunt aclin, 281 **Cil signefient li gopil 283 **E lurs est morty hier fu oscis ; 285 **Dous tors i ad pur les leons ; 286 **E pur la mery pernum les pluins, 287 **[/ lewe monte come mery **Damey la vision est dite.''^ **Uncore avant me dites, sire, **Quei icel fu put espeleir, ** Ken vostre buche vi ardeir ? '* "Dame," distil, **nesaike dait: ** Mes en dormant si me deceit." 290 292 293 294 295 296 454 reconforta) conforta. 455 Bele) Dame. 456 al, al) au, au. vs) oes. 462 hastes et) des granz. 464 qui mont) que mult. Variants of Ms. P 467 hier) er. 469 fist) fit. hui) omitted. 471 ploms) plums, mier) mer. 474 soiez) seez. 40 This line is perhaps added for the sake of the rhyme. Gaimar has : Dunt vus songastes ; co sunt il : possibly for the same reason. 41 Line 468 o( F is probably the invention of the author, since G has En un bois fu salvage pris. 42 The latter part of the description varies a little. The author of/' omits here, as elsewhere, the urging on the part of the hearer with which Gaimar introduces information. Cf. 292-294 of G. Then, before following in general the reply found in Gaimar about the flame, the author of F adds Havelok's alle- gorical interpretation of the fire to complete his natural explanation of the dream. 43 From 475 to 538 /'again difi^ers entirely from G. This passage is gener- ally attributed to the author of Fy who in this way avoids, in some degree, the abruptness of the question as to Havelok's home and parentage found in G. A COMPARISON OF THE FRENCH VERSIONS 11 THE LAI D HAVELOC 478 **Si en ert en peine et en efFrei 479 **De porter fors nos chaudrons 480 **Et nos pieles et nos ploms ; 481 **Et nepuroec ne quier mentir, 482 *'De ma bouche soelt feu is sir 483 '*^ant ieo me dorm, ne sai pur- quei; 484 '^Issi mauienty ceo peise mei." 48 5 DEL SOUNGE lessent atant, 486 Puis se endorment li enfant ; 487 Mes lendemain la matinee, 488 Quant Argentine fut leuee, 489 Vn chamberlenc qui fut od li, 490 Qui son piere auoit nurri, 49 1 La vision dist et conta ; Icil a bien la tourna. Puis li ad dit, ** En Lindesie ** Estoit vns homs de seinte vie ; **Heremite fut, en hois manoit. *' Sa lui parlast, il lui dirroit **Del sounge, quel ceo porroit estre ; *' Car Dieu lamoit, si ert prestre. " **Amis," fet ele, **mult te croi. ** Pur amour Dieu ! vien od moi. **A eel heremite voil parler, "Si tu i voels od moi aler." Cil li otrie bonement 504 Que od lui irra priueement. 505 Vne chape li afFubla, 506 Al heremitage la mena, 507 Al seint home la fist parler, 508 Et son corage tut mustrer 509 Del songe dont ele out poour. 492 493 494 495 496 497 498 499 500 501 502 503 297 < < Treskejo dorm, ma buche esprent, 298 **De la Jlambe nient ne me sent. 300 **Ke CO mavient en dormant.''^ 299 Veires jo en ai hunte mult grant. Variants of Ms. 478 E io serrai en efFrei. 491 479 fors) hors. 492 481 Et) mes. quier) quer. 493 482 soelt) soleit. 494 483 me) men. 495 485 atant) aitant. 497 486 se endorment) sen dormirent. 498 487 lendemain) el demein. matinee) 500 matine. 502 489 qui) omitted, fut) fui. 508 490 Ke li Reis sun piere norri. p La) Sa. la) li. Pus li cunta quen Lindesie. Estoit) Dut. seinte) hone. Heremite) Hermites. quel) que. porroit) put. Dieu) Deus. ert) fu. vien) ven. tu i voels) uoleies. E sauenture recunter. 78 THE OLD FRENCH LAI D'HAVELOC THE LAI D HAVELOC 510 Et de la bouche son seignour 5 1 1 Dont ele auoit le feu veu, 5 1 2 Mes ne sauoit qui ceo fu. 5 1 3 Par charite li quiert et prie 514 Qil la conseilt, si len die 5 I 5 Son auis et sa volente. 5 1 6 Li heremites ad suspire, 5 1 7 A Dieu comence soreison, 5 1 8 Puis li dist de lauision : 519 «*Bele," fet il, **ceo qe sunge as 520 **De ton baron, tu le verras. 521 **I1 est ne de real lignage, 522 ** Oncore auera grant heritage, 523 "Grant gent fra vers li encline, 524 * ^^^ ^^ ^^ dirrai,S5 596 " La verite ten conterai. 597 ^ * Ton pier e fut Gonter li roisy^^ 598 **j^z sire fut sur les Dariois ; 599 '*Hodulf loccist par traison, 600 ** Qui tuz iors out le queor felon. P Odolf 601 <*Li rois Arthur Hodulf fefFa 602 **Et Danemarche li dona. P nostre 603 ** Grim vostre piere senfuit, 604 ** Pur toi garrir terre guerpit. 605 **Ta miere fut en mier perie ; 606 ^*Car nostre nieffut assaillie P encontrerent 607 **Z)(? outlaghesy qi nus saisirent. P tuerent 608 ** Li plus de nostre gent i perirent. 609 ** Nous eschapames de la mort,57 GAIMAR 360 (E Argentine vint avoc. ) 400 **Fus fustes Jiz a un bon ret, 403 <*Z,z vostre per e out nun Gunter ; 401 ^^Danemarche out par heritage. 523 Cist reis (Odulf ) ki done ert el pais. 417 **Ki Artur volt, dona le terre. 423 ** Mis pere aveit mult bon nef ; 424 ** La raine amenout suef : 428 ** En mer furent trestuz ruez 430 ** E la raine ensemcnt. 583 De uthlages sumes asaillix, 428 *'En mer furent trestuz ruez 429 ** Nos chevalers, e nostre gent. 586 Jo guari, ne sai en quele guise ; 587 E li prodom en eschapa. Variants of Ms. P 593 Ore fai ta feme sauenir. 595 le) omitted. 596 ten) vus. 597 Tes peres fu Gunters li reis. 598 sire) sires. 599 Hodulf) Odolf. 600 Quant sur li uindrent li Bretun. 601 fefFa) ama. 603 vostre) nostre. 606 nief ) nef. 607 saisirent) encuntrerent. 608 i perirent) tuerent. 54 593 takes the place of 360 of G, which introduced Argentille earlier. 55 Kelloc's speech in F is better ordered than in G. She tells Havelok's origin and early history first and then of her plan for him. In G she tells that Grim is not his father, and after saying that a ship has just arrived from Denmark and advising him to return vv^ith it, she tells him his story. 56 The close similarity of these two passages is to be noted. 57 This account of the pirates' attack is found twice in G and three times in F. The first description of it in the latter poem is more detailed (106-120), and corresponds to G (425-433) where Kelloc states the same facts briefly. 606 and 609 of F echo 583 and 587 of Havelok's speech to Sygar in G. It is also interesting to note that 609 of F repeats in the first person i 2 1 of the same poem, and that 613 and 614 repeat exactly 140 and 136. 82 THE OLD FRENCH LAI D'HAVELOC THE LAI D HAVELOC 610 **Ci ariuames a cest port. 6 1 1 **Ne vout mon piere auant aler, 612 **Ici li estoet demorer. 613 **Sus cest hauene se herberga, 614 **Sile vendit et achata, 61 5 **Mult se pena de toi nurrir 616 **Et de celer et de couerir, 617 " Pouerement estoit vestuz, 618 *'Qe ne flissez aparceu. 619 **Nout si hardi en sa maison 620 ** Qui osast dire ton droit non ; 621 **Haueloc auez a non, amis.s^ P El pais 622 **Si aler voillez en vostre pais. 623 '*Mon seignur vus i conduiera,59 624 "Dedenz sa nef vus passera. 625 "Lautrer en vint, nad mie vn mois ; 626 **Assez oit qe li Danois 627 * * Vus voudroient entre eus tenir ; ^o 628 '* Car mult se fet li rois hair. 629 630 61 Fn prodome ad en la terre Qui touz iors ad vers li guerre GAIMAR 437 **En cest pais quant arivames. 380 **Grim vendi sel, si fu peschere. 421 "Co estes vus, si cum jo crei 422 **Danz Haveloc, le fiz le rei. 388 **Si vus volez od els aler, 389 **Jo quid kil irrunt el pais 390 ** U sunt vos parenz, e vos amis. 459 * * En Danemarche fud le autreer. 460 461 462 527 528 505 **E a plusurs oid preier, '*Sil vus trovast ke venissiez. **E le pais chalengissez. II out a nun Odulf le reis ; Mult fud haiz de ses Daneis. Hoc maneit uns riches horn. Variants of Ms. P 620 Vostre dreit nun nus chang- ames) E Cuarant vus appellames. 621 auez) ad. 622 Se aler volez el pais. 623 Mon seignur) mis sires. 624 Dras e uiande vus dora. 625 mie) pas. 627 Vus) Te. 629 prodome) riche home. 630 Vers ki li reis ad tuz iorz guere. 58 This point was insisted upon by the author (147—148). The additional detail of 617— 618, that Havelok was poorly clad in order not to attract attention, belongs wholly to F, 59 Some details of the advice about the return to Denmark vary in F. 623— 624 show that Kelloc's husband himself is to escort them. In G (383 ff. ) we find : Hier arivat leus al port, Un grant kenart, e bon e fort. Pain e char menied, e vin e ble ; Di eel unt il mult grant plente. Ultre la mer volent passer. Si vus volez od els aler, Jo quid kil irrunt el pais U sunt vos parenz, e vos amis. Si vus volez od els aler, Nus les vus purrum bien aluer. In 463 she advises him to take the two valets and promises that she and her husband will follow if he sends for them. The tone of F is more courtly in this passage. ^ Here also the ideas are similar. A COMPARISON OF THE FRENCH VERSIONS 83 THE LAI d'hAVELOC GAIMAR P Estal D L Lestarle 631 ^^ Sigar Lestal est appelkzy^^ 506 Sigar Estalre aveit nun : 632 '*A lui looms qe vus alez ; 633 ♦* II ad vne vostre parente, 634 ** Que pur vus est souent dolente 635 ** Qele ne poet nouele oir. 636 **Car desqa eus porrez venir, 637 **Oncore auerez vos heritez ; 468 **Si Deu vus rent vos heritez." 638 *^Ces n. valez od vus merrex.'*^ 398 **Vos dous vallez od vus merrez; 639 Argentine, quant ele loit,^^ 640 Mult durement sen esioit ; 641 A eus promet foi et amur; 642 Si Dieu la mette a honur, 472 Si Deus nus rent nos heritez : 643 Grant bien lur fera, ceo dit. 470 ** Nus vus rendrum mult bonluer; 644 Puis ni out gaires de respit ; 645 Lur nief tost apresterent, 646 Vers Danemarche mer passerent. 647 QUANT IL sont el pais venu, 496 Ken Denemarche sunt arive. 648 Et de la nief a terre issu/3 649 Li marchant qis amena ^4 Fariants of Ms, P 631 Lestal) Estal. 641 A eus) Assez. 632 looms) loum. 642 la mette a) les met a nul. 633 vostre) tue. 643 mult lur fra grant ben co dit. 634 pur vus) pur tei, 644 ni out) iout. 636 Car) Si. porrez) poez. 645 Lur nef e lur eire aturnerent, 637 vos) vostre. 647 el pais) a la terre. 638 merrez( menez. 648 a terre) omitted. ^'f In G no mention of Sygar is made to Havelok before he lands in Denmark. In F he is described in 43-46, and then again in 628-637. The description, however, does not differ in fact from that given in G later, although 630, 632- 636 belong only to the author of F. 62 Two lines in F are devoted to Argentine's joy. Gaimar gives a description of the ship, provisions, the clothes given Havelok and his wife, and the journey. ^3 This seems to be a favorite phrase. Cf. 123—124. 64 The merchant's part in F seems to be a combination of that of Alger and the merchant of G. Kelloc told Havelok that her husband (the merchant of 560) would accompany him to Denmark. Her husband has no name in F. In Den- mark the merchant of F, while fulfilling the part of Gaimar' s merchant, still has traits of Alger. He provides Havelok with clothes, a thing which Alger and Kelloc had done in G. He gives them minute instructions, telling them to go and eat at Sygar's table. G has only an obscure reference to a meal at Sygar's later in the poem. In F the scene (670-682) is foreshadowed in the directions of the merchant. This entire speech takes the place of geographical and historical information about Denmark and Hodulf in G, a part of which had already been given in i^ (625-631). 84 THE OLD FRENCH LAI D»HAVELOC THE LAI D HAVELOC 650 De bons draz les atourna, 651 Puis lur enseigne qil feront 652 Et a quiele ville il turner onty 653 A la cite eel seneschal, 654 ^e lorn appelle Sigar Lestal 655 **Haueloc," fet il, **beaus amis, 656 ** Quant tu vendras a son pais, 657 **En son chastel va herberger, 658 ** Et a sa table va manger, 659 **Par charite quier le conrei ; 660 **Ta femmemeine ensemble odtei, 661 **Assez tost te demanderont, 662 ** Far la beaute qen lui verront, 663 ** Qui tu es, et de quiele contree, 664 **Et qi tiele femme tad donee.'' 665 II senpartent del marchant, 666 Si tienent lur chemin auant. 667 Tant ont trauaille et erre, 668 Quil paruienent a la cite 669 La ou le seneschal manoit ; 670 Au chastel alerent tut droit ; 671 Le riche home en la curt trouerent, 672 Far charite li demanderent 673 Le conroi qil lur otriast, 674 Et qanuit les herbergast. 675 Li senescaus le lur granta ; 676 Dedenz la sale les mena. 677 Quant fut houre del manger, 678 Et qe tuz alerent lauer, 679 Li prodoms a manger sassist, 680 Les iii. valez seeir i fist,65 GAIMAR 498 A une vile sen alerent 505 Hoc maneit uns riches horn, 507 Seneschal fu al rei Gunter, D L Lestarle 506 Sigar Est a Ire aveit nun: 501 Les marchanz sunt tuz remes, 502 Od lur herneis, es dous nefs, 498 A une vile sen alerent. 505 Hoc maneit uns riches horn, 507 Seneschal ^M al rei Gunter. Variants of Ms. 650 bons) noueaus. 671 651 enseigne) enseigna. 674 653 cite) curt, del) al. 675 654 Que lom) Kem. 656 a) en. 676 657 En) A. 677 659 quier) quer. 679 661-662 Transposed in P. 680 668 paruienent) par uindrent. riche home) seignur. la) sa. qanuit) ke la nuit. senescaus) senechal. le) omitted. granta) ottreia. Dedenz) En. mena) enueia. Tant ke ore fu de digner. prodoms a) sire a sun. seeir fist) asser fit. 65 This scene belongs to F only, but the facts on which it is based may be gathered from Gaimar's description of the attack. In 531 he says the six valets A COMPARISON OF THE FRENCH VERSIONS THE LAI D HAVELOC 68 1 Argentille lez son seignur ; 682 Serui furent a grant honur. 683 Li bacheler et li esquier 684 Qui seruirent au manger 685 La bele dame ont esgardee 686 Et sa beaute forment loee. 687 En vne part se turnent sis ; 688 Ensemble ont lur consail pris 689 Qau valet sa femme toudront ; 690 Sil sencoruce, si le bateront. 69 1 QUANT IL leuerent del manger, 692 Li valez se vont herberger. 693 Li senescaus les fet mener 694 A vn ostel pur reposer. 695 Cil qui la dame ont coueitee, 696 ^i mult ert bele et ens eigne ey 697 Apres eus vont en vne rue ;^^ 698 Au valet ont sa femme tolue ; 699 Od eus leussent enmenee^ 700 ^ant Haueloc ad recoueree 701 Fne hache trench ante et dure, 702 Ne sai par quele auenture 703 Vn de ceus la tint et porta. ^7 704 II li tolit, si sen ala, 705 Les cink en ad tue et occis.^^ 533 Sis bachelers done lasaillirent, 5 3 I Pur sa moillery ke trop ert bele. 5 4 1 Cels ad ateint en la ruele. 534 Pris tent la dame, lui ferirent ; 537 Si cum il sen unt od sarnie, 538 Danz Have Iocs en out envie : 539 Prent une hache mult trenchant, 540 Ken une meison trova pendant; 543 Treis en oscist, dous en tua. Variants of Ms. P 684 au) a icel. 685 bele dame) meschine. ont) vnt mult. 686 loee)loe. 687 En) A. se) omitted. 689 femme) mie. 690 sil le) cil. 691 il leuerent) leue furent. 692 valez) enfant, se vont) sen wnt. 693 senescaus) seneschals. 699 enmenee) tute menee. 700 quant) Mes. 702 par quele) cum par fete. 704-707 E il li tolli e sen uenga. Les sis ad mort e afolez, Li uns esteit eschapez, Le destre poing out coupe. assailed Havelok : Pur sa moiller, ke trop ert bele ; and an obscure suggestion of a meal may be found in 669—671, where Sygar says to Havelok, Kore vus aim plus ke ne fis hier. Quant vus asis a mon manger. 66 541 of G refers to Havelok's pursuit of the valets, whereas 697 of F denotes their attack on Havelok. The phrases en une rue and en la ruelle, however, are the same. The description of the whole attack corresponds closely to G. 67 Gaimar states that the hache was found hanging in a house ; the author of F that Havelok took it from an assailant. 68 These two lines resemble each other strikingly. 706 prepares the way for 724. 86 THE OLD FRENCH LAI D'HAVELOC THE LAI D HAVELOC 706 707 708 709 710 711 712 713 Li vns est eschapez vifs, Mes qe le poign out coupe. ^9 Le cri lieue en la citey Cil senturnerent en fuiant, A vn mouster vindrent currant ; Pur garison i sont entre, Les huis ont sus eus ferme. Haueloc monta en la tour y7° 714 Cil del burg lasseent entour ; 715 De totes parz lont assailli, P mult bien 716 Et il bien se defendi ; 717 Desur le meur la piere prent,7i 7 1 8 Aual la gette ignielement. 7 1 9 La nouele vint au chastel 720 Au seneschal, qui n'est pas bel, 721 Qe cil quil auoit herberge 722 Cine de ses homes out tue, 723 Et li sistes est afolez, 724 Et il sen est eschapez; 725 En la tour del mouster sest mis, 726 Et li burgois lont assis. 727 Mult par lassaillent durement, 728 Et il se defent asprement ; 729 Les quareus de la tour enrue ; 730 Mulz en mahaigne, plus en tue. 731 Li senescaus cheual demande, 732 A touz ses cheualers comande 544 ^ al siste le poing trencha ; 546 Es vusy le cri mult criminel, 547 Prist ses vallez e sa moiller, 548 Si sen entra en un muster ; 549 Per ma les us, pur la pour, 550 Puis monterent sus en le tur. 553 Kar cil tres bien se defendirent ; 556 (Dan Sygar) Veit cum les pieres vait ruant. 554 Blescied i erent eels kis assailirent. Variants of Ms. P 708 lieue)leua. 710 vindrent) venent. currant) cor- ant. 7 1 2 Pus unt lus clos e ferme. 714 burg) burc. lasseent) le seent. 716 bien) mult ben. 717 meur) mur. piere) pere. 718 ignielement) uiuement. 719 au) al. 721 is followed by — E a sa table aueit mange. 722 tue) tues. 724 Omitted. 727 par) omitted. 728 asprement) mult asprement. 729 quareus) karuels. en rue) lur rue. 730 Mulz) Mut. mahaigne) maime. 731 senescaus) riches home, cheual) cheuals. ^ After this, according to G, Havelok starts toward his ostel with his wife. 70 In P the description is fuller than in G. 71 Gaimar suggests this act of rolling stones later in his brief line where he states that Sygar saw them falling as he rode up (556). In F the author de- scribes the news brought to Sygar, and his departure, but he adds no new details. A COMPARISON OF THE FRENCH VERSIONS 87 THE LAI D HAVELOC 733 Q£ o^ ^i augent a la meslee 734 Qs ^^ ^^ ^^^^ ^^^ leuee. 735 Tut primerain vet au mouster, 736 Et vist celui si bien aider 737 Qh^I ^^s fet tuz trere arere, 738 Chescun se doute qil nes fiere. 739 Li senescaus ala auant ; 740 Vist Haueloc et creu et grant, 72 741 (Et a sa table auoit mange, 73 742 Ensemble od lui out este), 743 Gent cors et bele feture, 744 Lungs braz et grant furcheure. 745 Ententiuement lesgarda, 746 De son seignur li remembra, 747 Del roi GunteVy qil tant ama ; 748 Anguissousement suspira, 749 Cil le resembloit de visage 750 Et de grandeur et de corsage. 751 II ad fet remaneir lassaut Et defent qe nuls ni aut ; Le valet ad a reson mis : "Negettez mes," fet il, **amis: ** Triues te doun, parole a moi ; **Lachaison me di et pur quoi74 ** Tu as mes homes issi morz. **As quieus de vus en est li torz ? " GAIMAR 752 753 754 755 756 757 758 555 Quant Dan Sigar vint puignant, 556 Veit cum les pieres vait ruant. 559 ^ig^i^ l^ "^ity 107 Bel vis a veit, e bele mains, 108 Cors eschevi, suef e plains. 559 . . . si lavisat; 560 Del rei Guntier dune li membrat: 564 Que quant le vit tel pitied en ot, 565 Qua mult grant paine pot parler. 563 A sun seignur resemblot, 566 Tut las alt ad fait cesser : 567 Peis e trues lui afia. Variants of Ms. P 733 meslee) melle. 735 primerain) premereins. au) al. 738 Chescun) Chescuns. nes) nel. fiere) fere. 741-742 Omitted. 743 cors) cors ot. feture) stature. 745 lesgarda) len esgarda. 750 corsage) corage. 751 lassaut) les essaut. 752 defent) defendi. 754 gettez) gette. 755 Triues) Trois. doun) doin. 756 di et) diez. 758 As quieus) E a quel, en est) ert. 72 The author of F used the same words, creu et grant, also in 178 and 285. 73 The insistence of the author of F on the table episode in this connection is odd. If the scene had been described in his source as he had already described it, why mention it again and for the third time ? 74 In F the conversation between Sygar and Havelok takes place on the spot ; the first question concerns the actual situation, the second, Havelok' s identity. In G the procedure is less natural (567 fF. ) : Peis e trues lui afia. En sa sale len amena, Lui e sa femme, e ses compaienz, Les dous vallez, dunt dis ainz. E quant furent aseurez, Li riches hom ad demandez, Ki il estait, etc. 88 THE OLD FRENCH LAI D'HAVELOC THE LAI D HAVELOC 759 **Sire," fet il, '*ieo le vus dirrai, 760 ** Que ia dun mot nen mentirai. 761 '* Quant nus del manger turnames 762 ** Oreinz et al ostel alames, 763 **A leisir de vostre meson 764 ** Me pursuirent li garcon ; 765 " Ma femme me voudrent tolir, 766 ** Et deuant moi od lui gisir, 767 **Vne de lur haches saisi, 768 **Et moi et li en defendi. 769 **Verite est qe ieos occis, 770 **Mes sur moi defendant le fis." -]-] I LES SENESCHAUS quant il oit 772 Le surfet de ceus, li respondit, 773 "Amis," fet il, **venez auant, 774 **Si ne dotez tant ne quant ; 775 "Gardez qe pas ne vus celez, 776 **Dites moi dont estes nez.^^ 7S GAIMAR 777 ** 778 - 779 " 780 " 781 ** 782 " 783 ** 784 " 785 " 786 *« 787 - 788 " Sire y'^ fet il, **de c est pais ; Ceo me conta vns mis amis. Vn riche home, qi Grim out a non Qui me nurrit en sa maison. Puis qe la regne fut conquis Et mis pieres fut occis. Ensemble od moi et od ma mere Menfui puis la mort mon piere ; Mult enporta or et argent. 76 Par mier errames lungement, De outiaghes fumes assailli. Ma mere occistrent et ieo garri. 571 E quant fiirent aseurez, 572 Li riches hom ad demandez, 573 Ki il est ait, e com ad nun, 577 ** Sire,'' fet il, *'ne sai ki sui . 578 En cest pais quid ke nexfui, 579 Un mariner, ki Grim out nun. 580 Men men at petit valetun. 582 Com venimes en halt mer, 583 De uthlages sumes asailliz, 585 Ma mere i ert, si fu oscise ; 760 Que . . . nen) De ren ne vus. 761 nus del) de nostre. 762 Oreinz) Or ainz. et) omitted. 763 A leisir) Al issir. 764 Me) nus. 765 voudrent) voleient. 768 en) omitted. 769 ieos) io les. 771 Seneschaus) seneschal, quant il oit) li respundit. Variants of Ms. 772 775 775- 777- 784 785 Quant le sur fet de ceus oi. ne vus) nel me. ■776 Transposed. ■781 Co me cunt uns meus amis Ki io fu ne en cest pais Vns riches home de cest pais Grim ot a nun qui me norris. Menfui) Senfui. Muh) Mut. 75 Gaimar adds questions as to Argentille and his companions. 76 The author o{ F repeats the same details regarding Grim's station that have been noted before. He adds now that gold and silver were taken with Havelok. A COMPARISON OF THE FRENCH VERSIONS 89 THE LAI D HAVELOC 789 '^Et li prodoms sen eschapa J go ''^i me nurrit et mult mamaJ'J 791 ** Quant nostre nief fut ariuee 792 **En vne sauuage contree, 793 ** Li prodoms mansion ileua, 794 **Tut primerement si herberga ; 795 **Assez nus troua a manger 796 ''Par vendre siel et par pescher. 7^ 797 **Puis iad tant de gent herberge, 798 **Qe ville i est et marche, 799 **Pur ceo qe Grim lapelloit lorn, 800 ** Grimesby ad la ville a non. 801 ''^uant ieo fui grant ieo men parti,n 802 **En la maison le roi Aelsi 803 **Fui souz le keus en la quisine ; 804 '* II me dona ceste meschine; 805 ''Sa parente ert; ne sai purquei 806 '*Il assembla et lui et mei.^° 807 **Ieo lenmenai fors de la terrc. 808 **Ore sui venu mes amis quere ; 809 **Ne sai ou pusse nul trouer. GAIMAR 586 Jo guariy ne sai en quele guise; 587 E li prodom en eschapa y 588 Ki me nuri, e mult mama: 444 380 59 789 prodoms) prodome. sen) en. 793 mansion) maisun. ileua) idresca. 794 primerement) premereins. si) se. 796 siel) eel. 797 de) omitted, herberge) herbergez. 798 marche) creu marchez. •'Peison eumes a manger; ■'Grim vendi sel, sifu pes ch ere. (Grim and his wife) ^uantfurent mort, si men turnaiy 592 Un rei servi u jo alai; 593 E dous vadlez furent od mei 594 Tant cum jo fui od eel rei; 597 Si cum lui plut la me don ad y 596 E ceste dame iert sa parente. 598 E ensemble nus espusad. 599 Ci sui venud en cest pais. 606 II (Alger) me load, e sa muillier, 607 Ci a venir, mes amis querre 609 Mes jo ne sa un sui nomer. Variants of Ms. P 799 lom) horn. 801 ieo men) de lui. 803 le) les. 807 fors) hors. 808 venu) venus. 809 pusse nul) en pusse. 77 These passages, F 786-790 and G 582-588, are remarkably alike. 78 In F the details mentioned in 123-142 are again repeated. This descrip- tion is based on G 437-452, with additional data as to the country where Grim landed and the growth of the town. Gaimar enumerates fishes and food, which are omitted in F. 79 The difference in the time of leaving Grimsby is striking. It is, on the other hand, noticeable that the line quant ieo fui grant, ieo men parti corresponds in form to quant furent mort, si men turnai (591 G). 80 Gaimar mentions the two valets here. The author of F had brought that item into his account before, and so omits it. On the other hand, he is more detailed as to Havelok's position at court. 90 THE OLD FRENCH LAI D'HAVELOC THE LAI D HAVELOC P un sul 8 1 o * * Car ieo ne sat nul nomer. * 8 1 1 LI SENESC AUS li respondi 812 **Beaus amis, ton non me di.* 813 **HaueloCy sire, sui nomez, P apellez 814 **Et Coaran fui rappellez ^^ 815 ^*^ant en la curt le roi estate 816 "Et de sa quisine seruoie.'* 817 Li senescaus se purpensa, 8 1 8 En son corage se remembra 8 1 9 Qe si out non li fiz le roi 820 Qe Grim en out mene od soi. 821 Purpoi qil nel ad coneu; 822 Mes nepurquant en doute fu.^^ 823 Par triues lad asseure. 824 Et el chastel lad amene,^'^ 825 Sa femme et ses compaignons. 826 II les appelle ses prisons. 827 Mult les fist bien seruir, 828 La nuit en sa chambre gisir. P furent choche 829 Quant li enfant cuche, 830 Vn son priue iad mande 831 Pur saueir quant cil dormira 832 Si flambe de lui istra GAIMAR 610 Ne ne sai com les puise trover. ' ' 61 1 Dist li prodom : **Cum as tu nun?*' D L Aveloc oi 616 **Sai ben que Haveloc eut nun D L Cuarant 614 ** Si mapelerent Cuherant: 613 * * Mes cum jo sui en la curt grant, 621 SYGAR sestut, si escultat: 622 Del fiz le rei bien // mem brat. 624 Le fiz Gunter eel non aveit. 571 E quant furent aseureZy 567 Peis e trues lui afia, i) E en 568 En sa sale len amena, 569 Lui e sa femme, e ses compaienz. 629 La nuit le fist tres bien guaiter, 625 Si li membrat de un altre vice. 810 nul) un sul. 812 Beaus) Beus duz. 814 rappellez) apellez. 816 seruoie) le serueie. 817 senescaus) riches home. 818 se)li. Variants of Ms. P 824 Cest chastel od lui menee. 825 ses) ses dous. 827 bien) ben le ior. 829 cuche) furent choche. 830 mande) enveie. 832 Si) Si la. 81 The statement regarding Havelok's name is different in the two versions. In G he does not know which name to give, and explains where each was used, and refers to Alger for the information about his boyhood ; in Z' he gives the name Havelok but says he was called Cuherant at court. ^2 There is much similarity in these two passages though F is often less concise. 83 In G Sygar had taken Havelok home before questioning him. In F the explanations had been made on the spot ; hence the difference in the two lines where Gaimar's sale is replaced by el chastel. The author o{ F also adds a few lines to tell of the kindly treatment of Havelok. A COMPARISON OF THE FRENCH VERSIONS 91 THE LAI D HAVELOC Ms. P. La norrisce quil ot norriz Souentes fez le regei Ken sun dormant li aueneit Meis cil feu pas ne lui nuseit.^4 833 Car ceo auenoit au fiz le roi 834 Qe Grim out mene od soi. 835 Haueloc fut mult lasy 836 E?idormi sest igtiiel pas ; 837 Meisme lure quil dormit 838 De sa bouche le feu is sit. 839 Le chamberlenc out grant poour; 840 Conter le vait a son seignur; 841 Et il en ad Dieu mercie, 842 Qe le dreit heir ad recouere.^5 843 844 845 846 847 848 849 850 851 852 853 854 Ses chapeleins fet demander ; Ses briefs escriure et enseeler ; Par ses messages les manda, Et pur ses amis enuoia. Pur ses homes f pur ses parenz. Mult i assembla granz genz, Tuz ceus qi el pais estoient Qui le roi Hodulf haoient. Par matin fet les baigns temprer' Et celui baig?ier et lauer ; De riches draz lad reuestu, Et sa femme, qe od lui fu ; ^7 GAIMAR D L sot 626 Kil vit jadis par la nurice. 627 De la flambe ki ert issant 628 De sa buche, quant ert dormant. 300 Ke CO mavient en dormant. 298 De la flambe nient ne me sent. 631 Pur CO kil ert ferment lasse 634 Sil sendormi, nuls nel demant. 635 Ignelpas com il dormiy 636 De sa buche la flambe is si. 637 E li servant ki lunt guaite, 638 A lur seignur lunt tost nuncie. 7 2 1 Dieu seit loez, 722 Ore di mon dreit seignur trouez 641 Dune sot il bien que veirs esteit 642 Co que de lui pensez aveit. 647 5/ mandat pur ses chevaliers, 648 Pur geldons e pur peoniers. 646 Dun pur ses humes enveiad. 650 Quant il en ot mult assemblez. 652 Baigner le feit e conreier. 653 De novels dras lad feit vestir : Variants of Ms. P 833-834 P substitutes the lines in- serted in the text. 835 mult) durement. 836 igniel) ignele. 843 chapeleins) escri veins. 844 enseeler) enseler. 845 man da) charga. 846 Et) omitted, enuoia) les enveia. 847 Pur, pur) E, e. 848 Mult i) Al demein. 849 estoient)maneient. haoient)aeient. 85 I baigns) bainz. ^4 These lines of Ms. P, wholly lacking in Ms. //, are close to those in G. ^5 In Fy after Sygar was told of the flame, he rejoiced, sent for his clerks, and through them summoned his people. In G Sygar went to verify the report, and recognized that Haveloik was the heir of Denmark. He too summoned his people. ^6 These descriptions are similar but the author of F amplifies several details. ^7 The author of Fy here as elsewhere, is more careful to mention Argentille than is Gaimar. 92 THE OLD FRENCH LAI D'HAVELOC THE LAI D HAVELOC 855 En la sale les ad menez. 654 856 Haueloc fut mult efFreez 657 655 857 De la grant gent qil veoit: 656 858 Haueloc mult se cremoit. 657 859 Pur les homes qil out occis. 659 860 Qe ceo fust vs de eel pais 861 Qe lorn le deust issi seruir, 862 Baigner, lauer, et reuestir, 863 Et puis iuger pur Ic mesfet, 864 Et auant amener au plet. 865 Nest merueille sil se dota :^ 866 Fne grant hache recouera 867 (El paleis pendit par vn croc) ^ 868 As ii. poigns lad pris Haueloc, 869 Vigerousement se voudra defendre 870 Sil le voelent iuger a pendre 871 LI SENESCHAUS se regarda, 872 Vers lui se trest, si lacola ; 873 * * Sire, * ' fet il, « * neiez effrei ; 874 **Cele hache rendez a mei.^ 875 **Neiez garde, ieo le vus di, 876 **Ma leaute vus en affi.'*^ ^jj II li ad la hache rendue ; 878 Et cil lad au croc pendue.91 GAIMAR 654 En la sale le feit venir. Pour out grant ke cele gent Com en la sale est entrez, V vist tant homes asemblez. Pour out grant ke cele gent Pur les cine homes hut tuez. 658 Ne li fascent mal jugement : 661 Pur une hache volt aler, 662 Ke iloc teneit un bacheler, 663 Saisir la volt pur set defendre. 664 Sigar le vait, si lad fet prendre. 665 Com il le tindrent de tuz leez, 666 Sigar li dist : **Ne vus dotez: 664 L Mes la hache de vus metez, 667 **Naiez guarde, li mien ami, 668 ** Bien le vus jure, sil vus aji, 669 * * Kore vus aim plus ke ne fis hier,9* Variants of Ms. P 857 De) Pur. 868 poigns) poinz. lad pris) le tint. 858 Sauez ke li uallet cremeit. 869 Vigerousement) Viuement. 864 au) al, and elsewhere. 870 a) ou. 866 recouera) retroua. 871 se) le. 867 par) a. 875 Neiez) Ni naiez. 88 The cause of Havelok's fear is explained more fully in F. The idea, how- ever, is the same as that in G. 89 Cf. this line with 703 of F, which occurs in the passage about the fight with Sy gar's men. 703 will be seen to correspond to 662 of G, whereas 867 of i^ corresponds to G 540, ken une meson trova pendant, used in the descrip- tion of the fight. This is a coincidence worthy of note. 90 The details are the same in the two versions, but Sygar's actions in F are more gentle and affectionate. 91 The acquiescence of Havelok and the returning of the hache to the hook is noted in F. Gaimar leaves the former to be understood. 92 G alone contains these lines which have already been referred to as con- taining an obscure allusion to a meal which has not been described. In F the meal is discussed and these lines are omitted or replaced by the lines about the hache. The omission is interesting. A COMPARISON OF THE FRENCH VERSIONS 93 THE LAI D HAVELOC P A une part ser le fit 879 A vne part le fet seeir, 880 P Sa femme juste li sasist,93 880 Qe bien le poent tuz veeir; 881 P Sun chamberlenc ad apelle, 882 P Le corn le rei ad demande, 881 De son tresor/>/ apporter 882 Le corn qe nul ne poet soner,94 883 Si dreit heir nest de lignage 884 Sur les Danois par heritage,96 883 P Co dit cil quil assaierunt 884 P Sauer si soner le porrunt, 885 Sauoir si soner le porroit ; 886 Dist lur quil essaieroit. 887 Cil qui porra soner le cor 888 // lui dorr a son anel dor. 97 889 Nout en la sale, cheualer, 890 Sergant, valet, nesquier 891 Qa sa bouche nel mist ; 892 Onques nuls soner nel fist. 893 Le seneschal ad le corn pris, 894 Haueloc lad en la main mis ; 895 **Amis," fet il, **car essaiez 896 **Si le corn soner porrez." 98 897 **Parfoi!" fet il, **sire, nesai;99 898 ** Onques mes corn ne maniai, 899 **Ieo nen uoudroie estre gabez ; GAIMAR 670 ** Quant vus asis a mon manger," 671 Puis si la sis t delez sei ; 672 Aporter feit le corn le rei. 686 A un chevaler le feit liverer;95 687 Z) Z Si lui ad dit tut en riant, 688 Quil sune si quil seit cornant. 687 **Kil sonerat kil seit cornant, 689 Jo li durrai un hon aneU"* 696 Ore vont corner le mainel, 697 Li chevaler e li sergant : 699 Unc pur nuls dels ne volt soner 700 Done lont bailie al bacheler, 702 Ki Avelocs out rion. D cor ne sonat 704 E dist, ke unkes ne cornat. 879-886 A une part ser the text. 889 II ni ot un sul cheualer. 891 mist) messist. 892 nuls) nul deus. Variants of Ms. P etc., as in 893 senescal) seneschaus. 894 la main) les meins. 896 Se ia soner le poez. 897 Par . . . il) Cil li respunt. 93 Again the author of F gives Argentille prominence, where G has no men- tion of her. 94 882-884 repeat with change of tense the exact words of 48— 50 of F. 95 G (673, 682—686) contains the description of the horn and the guardian- ship of it, which was given in F. 96 Ms. P and Mss. D and L are somewhat more alike than are Mss. H and M with respect to these two lines. 97 Gaimar devotes six lines to the magic properties of the ring. 98 The words of Sygar are given only in F. 99 G has Quant cil le tint, sil esgardat. 94 THE OLD FRENCH LAI D'HAVELOC THE LAI D HAVELOC Mes puis qe vus le me comandez, A ma bouche le corn mettrai, ^°° 900 901 902 **Et si ieo puis, ieo le sonerai. 903 Haueloc est leue en piez, 904 Et del corner apparaillez, 905 Le corn benesquit et seigna; 905 P Le corn sona par tel heir Haute men t et bien le sona^°^ P Ke mult le pout hum de loinz oir, A grant merueille le tenoient 908 Tuit cil qen la sale estoient 909 Li senescaus les appella, 910 A tuz ensemble le mustra ; ** Seignurs, purceo vus ai mandez, **^ Dieu nus ad reuisitez. **Veex ci nostre dreit heir ; ** Bien en deuom grant ioie aueir. ' ' Tut primerain se desafubla, Pardeuant lui sagenuilla; Sis homs deuint, si li iura 918 Qe leaument le seruira. 919 Li autre sont apres ale. 906 906 907 911 912 913 914 915 916 917 920 Chescuns de bone volente ; 92 1 Tuit si home sont deuenu 922 Puis quil li eurent receUy'^°'^ 923 La nouele fut recontee : 924 Ne pout estre lunges celee. 710 **J vostre buche le metez.*^ 712 **De mai serra ja aseie.'* 714 A sa buche lad asaie. 713 Done prist le corn, si lad seigne ; 716 Le corn tant gentement sonat, 717 Ke unc ne fu ainz oi son per ; 7 1 9 Sygar lentent ; sailli en piez, 720 Entre ses bras lad enbracez. 721 Puis sescriat : ^^Deu seit loez, 722 *'Ore ai mon dreit seignur trovex: 723 **Ore ai celui ke desirai." 729 II meismes sagenullay 730 De fai tenir laseura. 727 Tuz ses homes ad done mande : 728 Lores li firent felte. 733 Tuz sunt ses homes devenuz, 734 E a seignur lunt receuz. Variants of Ms. P 900 le) omitted. 901 corn) corin. 902 ieo) iol. ieo le) sil. 903 est) sest. piez) pez. 904-906 De corner sest aparilez Le corn etc. as in the text. 908 sale) mesun. 910 le) lur. 914 Bien) Mult, deuom) deuez. 915 primerain) premerement. desa- fubla) desfubla. 917 Sis) Ses. 921-922 Transposed in P. 100 Havelok*s remarks in F repeat in the first person the directions given by Sygar in the second person, according to G. 899 belongs to F alone. 898— 901 m F represent ten lines of G where more conversation is given. ^°^ Gaimar continues to describe the sound. Ms. P contains a descriptive clause as does G (717), although the two lines differ. ^°* The general description of the homage is the same. In F the order is different ; Sygar goes first and then the others ; in G all did him homage and Sygar himself knelt and pledged his faith. A COMPARISON OF THE FRENCH VERSIONS 95 THE LAI d'hAVELOC GAIMAR 925 De totes parz i accurroient, 926 Et riche et poure, qui loeient, 927 De lui firent lur auowe,^°3 928 A cheualer lont adubbe.^°4 929 Tant li aida le senescal, 930 Qui prodome fut et leal, 93 I Qe merveillous ost assembla.^°5 735 Quant co unt fet, asemblent gent; 736 En quatre jurs en ont maint cent, 737 E al quint jor, des chevalers 738 Ourent il bien trente millers. 932 Au roi Hodulf par brief manda 739 Li reis Edulf done deffierent : 933 Qe la terre li deliverast, {D L Odulf) 934. Hastiuement si sen alast. 935 LI ROIS Hodulf, quant ceo oi, 936 Mult sen gaba et escharni;^°^ 937 Ceo dist qa lui combatera. 938 De totes parz gent auna, 939 Et li valez en reont assez. 940 Au iour qentre eus fut nomez. Variants of Ms. P 925 accurroient) acoreient. 930 leal) bon vassal. 926 qui loeient) quil oieient. 938 auna) assembla. 927 auowe) auoe. 939 reont) runt. ^°3 The author of F alone mentions the concourse of all kinds of people to do homage to Havelok, after the news of the new king has become known. ^°4 Only in F do we find that Havelok was knighted. ^°5 G is more exact in his information about the mustering of an army whereas the author of F dwells on the challenging of Hodulf 106 Gaimar does not note Hodulf s reception of the challenge or his assemb- ling of an army, though that Hodulf collected one is evident. He appointed a day for the battle. Havelok fought Hodulf in single combat ; Gaimar, however, states merely : En un plein sentrecontrerent. Asez i out granz colps feruz : Li reis Edulf fu dunke vencuz. Car Haveloc si se contint, II sal en oscist plus de vint. Dous princes aveit el pais, Ki ainz erent ses enemis, E od Edulf serent tenuz : Ore sunt a sa merci venuz. (740-748) In the passage in F the circumstances given by G about the victory, the menue gent and Havelok' s forgiveness of them are kept. In addition the menue gent are utilized to introduce a single combat replacing the battle in G. The two princes are left out by the author of F. 96 THE OLD FRENCH LAI D'HAVELOC THE LAI D HAVELOC 941 Que li dui ost sassembleroient 942 Et ensemble se combateroient, 943 Haueloc vist la gent menue 944 Qen saide estoit venue ; 945 Ne voelt quil soient occis : 946 Au roi Hodulf, par ses amis, 947 Manda qa lui se combatist 948 Cors contre cors, et, si le ven- quist, 949 Les genz a lui touz se venissent, 950 Et a seignur le tenissent : 951 *' Ne sai purquei se combateroient 952 **Qui nule culpe nen auoient." 953 Li rois nel deigna refuser, 954 Tote sa gent fist desarmer, 955 Et cil la sue de lautre part ; 956 Mult durement li sembla tart 957 Qyil soient ensemble venu, 958 Et quil eust gaigne ou perdu. 959 Ensemble vindrent li baron, 960 Requistrent soi come leon. 961 Haueloc fut de grant vertu ; 962 Le roi Hodulf ad si feru 963 Dune hache qil apporta, 964 Quil labatit, puis ne leua, 965 Iloec loccist deuant sa gent, 966 Qe touz li crient hautement, 967 **Sire, merci, qe ni moroms, 968 **Car volenters te seruiroms.'* 969 Cil se sont a lui tourne, 970 Et il lur ad tut par done. 971 Apres cest fet ad receu 972 Le regne qa son piere fu. 973 Par la terre bone pees mist, 974 Et des felons iustise prist. GAIMAR 749 Del pais la menue gent, 750 Vindrent a merci ensement ; 751 E Haveloc lur fist pardons y 752 Par le conseil de ses barons. 756 De lui firent seignur e reis. Variants of Ms. P 941-942 Quant les os furent assem- blez E de bataille conreiez. 944 saide) saie. 950 a) cum. tenissent) seruisent. 951 se combateroient ) cil en moreient . 952 culpe) cupes. 956 sembla) est. 957 soient) fussent. 958 gaigne) pris. 959 vindrent) venent. 960 Requistrent) Requerent. 967 qe ni moroms) nus ne morom. 969 tourne) comande. After 972 Li Daneis en firent lur Rei Tuz ses uesins suz mist a sei. 973 Par la) En sa. 974 prist) fist. A COMPARISON OF THE FRENCH VERSIONS 97 THE LAI D HAVELOC 975 Sa femme creut et lamoit, 976 Et ele mult bien le seruoit. 977 Mult fut eincois desesperee, 978 Mes ore lad Dieus reconfortee, 979 Quant Haueloc est rois pussanz. 980 Le regne tint plus de iiii. anz,^°7 981 Merueillous tresor i auna. 982 Argentine li comanda 983 Quil passast en Engleterre 984 Pur son heritage conquerre, 985 Dont son oncle lout engettee, 986 A grant tort desheritee. 987 Li rois li dist quil fera 988 Ceo qele li comandera. 989 Sa nauie fet a turner, 990 Ses genz et ses ostz mander ; 99 1 En mier se met quant orre a, 992 Et la reyne od lui mena. 993 Quatre vinz [nefs] et quatre cenz 994 Out Haueloc pleines de genz.^°^ 995 Tant ont nage et sigle, 996 Qen Carleflure est ariue. GAIMAR 759 Apres samond tute sa navire^ 760 De son realme tute lempire. 494 Lur sigles drescent al vent. 495 Tant unt nage e governe, 496 Ken Denemarche sunt arive. Variants of Ms, 975—976 Sa femme ama e mult cheri 990 E le lout ben deserui. 977 eincois) anceis. 980 regne) realme. iiii.)treiz, 981 Merueillous) merueile. 982 comanda) conseilla. 985 son oncle) ses uncles, lout en- gettee) aueit iete. 988 Quan quele li conseilera. 990 Ses . . . ostz) A sagentesun ost. 996 991 993 994 994 is followed in P by Quant sun cire ot apareille Ni ad pus gaires atarge. orre) oree. nefs) in P only. Out) Od. followed in P by Armes portent e garisun, Vin e forment, char e pesun. Carleflure) Carreflod. est) sunt. ^°7 In F the peace in Denmark is described, and reference is made to Argen- tine (975—978). Havelok reigned four years, then at Argentine's instigation went to England. G describes the oaths of fealty which were taken after the battle, and in 757-758 mentions a feast and the general rejoicing, si cum nus dit la verai estoire. Havelok then leaves immediately for England. No reference is made to Argentille. The feast of Gaimar's story has given rise to much com- ment, as one of the principal differences in the two versions. i°2 F is more detailed in this description of the return to England. The author gives the number of boats, mentions Argentille, and names the landing place in England. 995 corresponds closely to G (495). The line occurred in F before. Cf 122. 98 THE OLD FRENCH LAI D'HAVELOC 997 998 999 1000 1 00 1 1002 1003 1004 1005 1006 1007 1008 1009 lOIO 101 1 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 997 THE LAI D HAVELOC Sur le hauene se herbergerent. Par le pais viande quierent. Puis enuoia li nobles rois Par le consail de ses Danois, A Alsi, quil li rendist La terre qe tint Ekenbright, Qa sa niece fut donee, Dont il lout desheritee ; Ety si rendre nel voleity Mande quil le purchaceroit. AU ROI uindrent li messager. Mult le trouerent fort et fier Quant il li ont ceo conte, Et il en ont ris et gabe/°9 Par orgoil lur respondi, **Merueille," fet il, **ai oi **De Coaran eel mien quistron, "Que ieo nurri en ma maison, **Qe me vient terre demander. ** Mes keus ferai a lui luster *' Od trepez et od chaudrons, "Od paeles et od ploms.'* Li messager sen sont turne ; A lur seignour ont conte Le respons qe le roi lur fist, Et del terme qe le roi lur mist. Dedenz li iour qil eurent pris, Alsi manda ses amis Et touz ceus quil pout aueir ; Nul ni lessa remaneir. A Theford les ostz assemblerent E del ferir se conreierent. Rois Alsi primes sarma ; Sur vn cheual ferant monta ; GAIMAR 762 Li reis Edelsi done deffia. 764 Sil ne li rendy le drait sarnie 763 Co li manda y kil le defie. 99/ hauene) marine, se) omitted. 1002 Ekenbright) Achebrit. 1003 donee) iure. 1006 Mande) omitted. 1008 fort) dur. fier) fer. 1009 Lur message li unt cunte. 10 1 1 Par) Par grant, lur) les. Variants of Ms. P I o 1 7 chaudrons ) caudruns. 1019 turne) alez. 1 02 1 le roi) omitted. 1021 followed by Quant les douz oz sesemblerent E ensemble se cumbaterent. 1027 Theford) Tofort. ^°9 Alsi's jests here belong only to the author of F. It will be remembered, however, that Gaimar reported his jests earlier in the poem where the author of F omitted them. Cf. G 323 IF. A COMPARISON OF THE FRENCH VERSIONS 99 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1 04 1 1042 •043 1044 1045 1046 1047 1048 1049 1050 1051 1052 THE LAI D HAVELOC Ses enemis vet surueeir, Combien de gent poent aueir Quant il ad les Danois veuz As enseignes et as escuz, Ne li remembra des caudrons, Ne des paeles ne des ploms, Dont il les auoit manacez ; Arere sen est repairez, Sa gent enseigne qil feront. Et coment se combateront.'^^'^ Entre eus fut dure la meslee Dissi qe vint a la vespree^ Quil ne poeient plus suiFrir ; La neire nuit les fist partir. Mult tout des Danois maumis, Et des autre s assez occis "'^ Haueloc fut irascuz Pur ses homes qil out perduz ; Od ses Danois sen fust alez, Et a sa nauie retournez. Si la reine li suiFrisist : Mes vn engin ele laprist "^ 053 Dont il veincroit son enemi. 054 Remist le roi, si la crei ; 767 {^Combatirent sei en un plain,) 768 Ty€i Y£\.2i\\r\. treskal serain. 7 7 1 ^ant naire nuit les desevera, j6() Mult i out homes afolez 770 Dambedous parz, e mort rueiz. 773 Mes par conseil de la reine, 774 Ki enseignat une mes cine D L omit Par ki 775 Par ki remist le mal en la bataille. Variants of Ms. P 1032 poent) il put. 1038 Arere sen est) Ainz est arere. 1 043- 1 044 follows 1 04 5- 1 046. 1043 plus) mes. 1044 La neire) quant la. 1045— 1046 maumis) oscis. Trans- posed in P. 1047 fut) fu mult. I o 5 1 suffrisist ) suiFrit. 1052 ele laprist) quili pramist. 1053 veincroit) veincerei. ^^° Gaimar says only : Li reis Edelsi li remandat, Ke contre lui se combaterat. Combatirent sei en un plain. (765—767) The author of F adds the description of the mustering of an army, Nul ni lessa remaneir, the plan of the battle at Theford, the feeling of Alsi at the sight of the Danes, and his forgetting his jests. "^ These passages, 1 042-1 046 of F and 768-770 of G, are very similar. "2 The action of the queen is even more emphatic in Fy since Havelok, furious at the loss o^ his men, is about to return to Denmark when she saves him. Lines 1052— 1 100 are, with the exception of an occasional detail elaborated by the author of Fy almost word for word the lines of G. There is not a variation of importance, and the number of identical lines in the two poems is astonishing. oo THE OLD FRENCH LAI D'HAVELOC THE LAI D HAVELOC 1055 Tote nuitfist grant pens trencher , 1056 Et de ii. parz bien aguisser ; 1057 Les homes morz i enficherent, 1058 Et entre les vifs les drescerent; 1059 ^^^^ e Scheie s en ont rengees, 1060 Les baches sur les cols leuees. 1 06 1 AJJ MATINy quant il aiourna, P saparila 1062 Rois Alsi primer sarma ; 1063 Si firent tuit si cheualer, 1 064 Pur bataille comencer ; 1 06 5 Mes quant il virent ceus delay 1066 Tote la char lur her tea, \Q)(i'] Mult fut hidouse la compaigne 1068 Des morz qil virent en la plaine. 1069 Contre vn home qil auoient, 1070 De lautre part vii. estoient. 1 07 1 Au roi dient si conseiller 1072 ^ bataille ni ad mester ; 1073 Li Danois sont de genz creuz, 1074 Et il ad multz des soens perduz ; 1075 A la dame rende son droit , 1076 Et face pees einz qe pis soit. 1077 P 1a reis ne pout par el passer 1077 Au roi les toe t tut grant er, 1078 P Le plet li estut granter, 1078 Car il ne poet par el passer, 1079 Par le cons ail de ses priuez, 1 080 Au roi Danois sest acordez ; 1 08 1 Par fiance lasseura, 1082 Et saufs ostages li dona, 1083 Tote sa terre li rendit 1084 Que Ekenbright tint tant come il vesquit. P De Hoillant tresque a Coles- estre. 777 Tute nuit fist en terre fie her pels, 778 (Plus gros e granz ke tonels) 779 Les morz homes en sus ficherent, 780 E tute nuit sus les drescerent. 7 8 I Dous escheles en firent granz, 782 Ke veirement estait semblant 783 Kil fuissent combatanz e vifs : 784 Le jor devant erent oscis. 789 Lendemain se reparillerent ; 790 De combatre mult saficherent. 785 Home ki de loinz les esguardout, 794 Tut la char len h eric a: 786 Tute la char len hericout. 788 Hydus semblent morz desconfes. 787 Ambure de loinz e depres, 795 Car en contre uns horn kil aveient, 796 Del altre part set en vaient. 797 Arere en vont al ret nuncier, 798 Li combatre ni ad mester : 799 Rende a la dame son dreit ; 800 E fasce pets ainz ke pis seit. 802 Done li estut co granter ; D L passer 801 Li reis ne pout par el aler, 803 Car si baron li ont loe. 804 Rendu li fut tut li regne. Variants of Ms, P 1055 grant) omitted. 1056 bien) omitted. 1057 enficherent) fichetent. 1060 cols) cous. leuees) drecez. 1062 primer sarma) saparila. 1064 Pur) Pur la. 1068 plaine) champaigne. 1069 Contre) En cuntre. 1070 estoient) enveient. 1077-1078 as above in the text. 1082 saufs) ses. dona) bailla. 1084 tant come) quant. A COMPARISON OF THE FRENCH VERSIONS lOI THE LAI D HAVELOC 1085 De Holande desqen Gloucestre 1 086 Furent Danois seignur et mestre ; 1087 Mes Haueloc sa feste tint 1088 A la cite quant il vint ; 1089 Des barons recent les homages y 1 090 Si lur rendit lur heritages. 1 09 1 Enpres cest fet rots Aelsis Ne vesquit mesqe quinze dis ; II nout nul heir si droiturel Come Haueloc et sa muiller. Li baron les ont receuz Et citez et chasteuz renduz. Haueloc tint en sa baillie 1092 1093 1094 1095 1096 1097 1098 Nicole et tote Lindesie ; 1099 XX. anz regnay si en fut roiSy 1 100 Assez conquist par ses Danois ; 805 Des Hoiland treska Colecestre. 806 Rei Haveloc la tin sa feste ; 807 Les homages de ses barons y 808 Recuz par tut ses regions. 809 Puis apres co, ke quinz dis 810 Ne vesqui li reis Edelsis. 811 // nout nul eir si dreiturel 812 Com Haveloc e sa muiller: 814 Li barnage tresben otreienty 815 816 817 818 Ke Haveloc, e sa amis. Ait la terre rei Edelsis. Sue ert Nicole e Lindeseiey J a si ot il: vint anzfud reis: Mult cunquist par les Daneis. Variants of Ms. P 1085 De Hoillant tresque a Colesestre. 1093 nout) nolt. droiturel) dreiturer. 1 09 1 Enpres) Apres. 1099 si en) e si. 1092 mesqe) mie ke. 1 100 ses) ces. The first impression received from the preceding compar- ison is that of surprise that so many lines reflect the verses of Gaimar almost word for word, and that the parts showing a real change in the details narrated by him are comparatively few. Reviewing these changes, one sees that they consist of the fol- lowing points: (i) In the Lay Gaimar's mariner has become a baron ; (2) Havelok leaves Grimsby during Grim's lifetime, and not after his death as in G; (3) in i^ the king appoints Alsi as guardian for Argentille before he dies, whereas in G Adelbricht dies and the queen goes to her brother; (4) the author of F says of Havelok, Pur la franchise q'il out Entre eus le tenoient pur sot ; Gaimar, on the other hand, speaks of his generosity as making him beloved at court; {5) in F Kelloc's husband in person conducts Havelok to Denmark, in G they commend him to a merchant whom they know; (6) in the two descriptions in the Lay where the "hache" is mentioned the lines in which Gaimar 1 02 THE OLD FRENCH LAI D'HAVELOC described the place from which it was taken are interchanged, 702-704 and 856-867; (7) when Sygar in F addresses Have- lok after the tower episode, he asks first about the fight, then as to Havelok's identity, and then takes Havelok and Argen- tine home with him : in G he gives a truce, takes them home, and then asks Havelok who he is; (8) the place assigned to Havelok before the horn is brought in is different in the two versions ; (9) a simple variation occurs in the order in which all do reverence to Havelok; (10) the single combat replaces the battle mentioned by Gaimar {F 935-965). Points I, 2, 3, and 10 are important modifications of the de- tails found in G, and each modification is followed by a whole scene lacking in G: i, by the description of Grim's guardian- ship, the castle, and embarcation ; 2, by the elaborate account of Grim's interview with Havelok, the instructions given the departing youth, and the gift of clothes; 3, by the scene in which Argentine's guardian is appointed, including his oath which leads eventually to the scene with the barons; and 10, by the lines telling of the single combat, i has been conceded by all to be a change made by the author of the Lay, who be- trayed himself twice by inconsistencies, 118 and 135. 10 may also be considered his invention, since he probably wished to introduce a knightly combat. The remaining points 4,5,6,7,8, represent very slight differences. They give a variation in tone to the description by stressing the prominence of Argentille (9), by noting the more courtly deference paid to Havelok by those surrounding him (5, 10), by emphasizing the inability of the vulgar to comprehend Havelok (4), or by making the narrative more complete and clear (6,7). These minor changes all appear to indicate the attempt of the author of F to impart clearness and courtly coloring to his work, and the same intention seems to explain the two more sweeping changes i and 10. Point 2 also occurs in a scene which is probably the author's own, — the elaborate account of Grim's interview with Havelok, the in- structions to him, and his gift. The question is only whether in his source he found that Grim was living when Havelok left Grimsby, or whether he changed that detail also in order to introduce the scene of farewell. Again 3, as we have seen in the discussion of the English Lay, seemed to be a change from the more primitive details given by Gaimar. In that same cornparison it seemed also that 2 was less primitive in form A COMPARISON OF THE FRENCH VERSIONS 103 than the circumstances of Gaimar's account. Is it not probable, then, that the author of the Lay made this change in order to provide Argentille as well as Havelok with a guardian, and that he introduced the oath and the scene of the barons to increase the courtly atmosphere of the poem, and to give his audience, according to his point of view, a plausible excuse for the con- duct of the barons in allowing their queen to be tfeated in such a disgraceful manner? Is there not here evident the same spirit that inspired the line of the Lay describing the valiant defence by Havelok's party in the scene of the pirates* attack? In Chapter III, I suggested the intermediary position which the Lay seemed to hold with regard to Gaimar and the English Lay in these very scenes. It seems probable that Gaimar's details were changed. Is it therefore not likely that these points, 2 and 3, the guardianship of Argentille and the depart- ure from Grimsby, were modified for the first time by the author of the Lay who, as we know, made other modifications and changes, the spirit of which corresponds exactly to that of these two variations? Few actual changes of Gaimar*s narrative are to be seen in the Lay ; the versions accordingly differ chiefly because of the parts omitted or added by the author of the latter. The omis- sions cited by Petrie have already been discussed and found to be unimportant. As additions to Gaimar, besides such details as the visit to the hermit, Petrie mentioned the full account of the tower episode referred to only vaguely by Gaimar, and the meal described in the Lay and alluded to very obscurely by Gaimar. Kupferschmidt, however, shows that these descrip- tions do not differ in substance, and, moreover, that the in- cidents of the story of the Lay are all found in Gaimar's account, although in more condensed form. The description of the meal at Sygar's offers no noteworthy details not found in Gaimar's description of the fight, or elsewhere in his nar- rative, but they give a clear, connected scene where Gaimar gave only an obscure reference to a meal, when Sygar said to Havelok: Kore vus aim plus ke ne fis hier. Quant vus asis a mon manger. (669-670) That this is a conscious expansion on the part of the writer of the Lay seems attested by the unnecessary stress which he I04 THE OLD FRENCH LAI D'HAVELOC gives the detail, by bringing in an allusion to it again, when he describes Sygar*s seeing Havelok on the tower: Vist Haveloc et creu et grant, (Et a sa table auoit mange. Ensemble od lui out este). (740—742) The principal additions to Gaimar*s version include the fol- lowing scenes : the description of Baron Grim, the castle and embarcation (51-69, 74-82, 84-109); Grim's parting from Havelok (157-192); the appointment of the guardian of Argentine (212-230); the long scene of the king and the barons (279-360, 364-376); the chamberlain and the hermit (485-536). Other additions consist of short passages, or at times merely a line, inserted for the purpose already referred to — to add clearness to the description or to bring out some detail more in keeping with the courtly ideal of the author. Several such additions are interesting, as, for example, the allu- sions to the Bretons (21,40, 258-260); references to Grimsby and the founding of the town (125-130, 137-142, 791-800); and suggestions of Argentine's prominence (639-643, 975-988, 1047-1053). As I have said, the differences between Gaimar and the French Lay have been much emphasized, and the occurrence of some identical lines in the two works noted, but the signifi- cance of these identical lines and similar passages has, it seems to me, been underestimated, and they have never been investi- gated with the same thoroughness as the differences. Com- parison shows how many such lines and passages there are. The character of these similar passages and their distribution are also to be noted. It is to be observed first of all that the passages are of two kinds: (i) those which represent Gaimar's details with different wording; (2) those which render Gaimar's details with slight change in wording or with no change at all. To the first class belong the following : 25-39541-50, 69,71-73, 116, 117, 131, 134-136, 143-146, 150-156, 194-196, 199, 200, 203-210, 231-234, 241, 242, 244, 251, 252, 264-272, 361, 362, 377-379> 389-393* 397> 398, 409, 410, 413, 420, 422, 428, 435, 436, 442, 445, 457, 458, 466, 468, 475, 537, 547-555> 559> 560, 567-573> 576-579* 584-586, 588, 589, 593, 595-605, 608-614, 621, 622, 625-632, 637, 643-649, 653-667, 695, 698, 702, 704, A COMPARISON OF THE FRENCH VERSIONS 105 706, 709, 711, 714, 715, 717, 718-745. 747> 748, 750-754, IS^-llS^ 778-786, 791-795, 802-807, 811, 812, 816-818, 820-822, 826-834, 841-850, 856, 858, 868-873, 876-878, 881-887, 889-891, 893, 903-912, 914, 915, 917-920, 931- 933, 989, 999-1004, 1042, 1046, 1053, 1054, 1060-1064, 1068, 1080 1082, 1084, 1086, 1097. To the second class belong the following : no. III, 115 and two lines following in Ms. P, 11 8-1 20, 122, 132, 2^3^ 380, 384, 399, 400, 401, 403-408, 411, 412, 414-419, 421, 423-425> 427, 429> 430, 433, 434, 437, 438, 440-444, 446, 449, 450, 454-456, 459-465, 467, 469, 471- 473, 482, 483, 539, 540-543, 552, SS^^ 574, 575, 583, 587, 590-592, 606, 607, 631, 638, 668, 66^,6^6, 697,701, 705, 707, 708, 710, 712, 713, 716, 746, 751, 755, 776, 777, 779, 787-790, 796, 808-810, 813-815, 819, 823-825, P, 832 and the line following, 835-838, 840, 852, 853, 855, 857, 859, 866, 874, 875, 888, 892, 901, 913, 916, 921, 970, 1005, 1044, 1052, 1055, 1057-1059, 1065-1067, 1069, 1070-1072, 1075-1079, 1083, 1085, 1087, 1089, 1091-1095, 1098 -1099, 1 100. Total 170 lines. This number of almost identical lines seems too large to be explained by assuming a common source for Gaimar and the French Lay. It does not seem credible that so many lines could have been taken from a third work by the two authors and have been so faithfully preserved in their poems that there is scarcely more difference at times between the two versions than exists between different manuscripts of the same version. Moreover, if we regard the arrangement and character of the lines which are similar, we note an interesting fact. The lines occur throughout the poems sometimes singly and at other times in groups. The first large group forms a passage con- taining lines 397-473, describing Argentine's vision; the second comprises many lines between 1052 and iioo, describing the trick devised by Argentille to win the battle in England. These passages seem to be transferred almost bodily from Gaimar's text. There is scarcely an addition or a change. This fact would indicate that, where the author of the Lay wished to follow Gaimar, he did not hesitate to take the whole passage. Such a passage speaks more strongly for Gaimar as source of the Lay than do isolated lines occasionally found in the poem. It would be far harder to transfer whole passages from a third [o6 THE OLD FRENCH LAI D'HAVELOC work and keep such close resemblance between the two derived poems than to do so in the case of occasional lines. These two longer descriptions — that of the vision and that of the device for winning a battle — the writer evidently did not care to omit, and, since no modification would adapt them better to the spirit of his Lay, he simply used Gaimar*s text. As a further indication that he followed Gaimar's version, compare the three descriptions of the pirates* attack, which occur in the Lay. (ii 1-120) (A) Car outlages les encontrerent. Qui hautement les escrierent. Mult durement les assaillirent, Et cil forment se defendirent ; Mes il eurent poi desforz. La nef unt robe e mal mise, E la Reine i fu ossise, Li outlaghe les ont touz morz. Ni remist nul petit ne grant Fors Grim, quert lur conoissanty GAIMAR (427-436) De utlagles fumes encontrex: En mer furent trestuz ruez Nos chevalers, e nostre gent, E la reine ensement. Unc ne guari horn, fors mun pere, Ne nule femme, fors ma mere. Mis pere est ait lur conussant ; Pur CO guarirent li enfant, E jo, e vus, e mi dui frere. Pur la priere de mun pere. (603-609) (B) Grim nostre piere senfuit. Pur toi garrir terre guerpit. Ta miere fut en mier perie ; Car nostre nef fut assaillie De outlaghes, qui nus saisirent. Li plus de nostre gent i perirent. Nous eschapames de la mort. (786-790) (C) Par mier errames lungement, De outlaghes fumes assailli. Ma mere occistrent e ieo garri, Et li prodoms sen eschapa Qui me nurrit et mult mama. (582-588) Com venimes en halt mer, De uthlages sumes asailliz. Ma mere i ert, si fu oscise ; Jo guari, ne sai en quele guise ; E li prodom en eschapa, Ki me nuri, e mult mama. It is evident that A was based on Gaimar's version, and that Kelloc's speech in the first person was changed to the third person to serve as a description of the events in Denmark for the first part of the Lay. At times the lines are almost identi- cal, even to the extent of introducing the preposterous state- A COMPARISON OF THE FRENCH VERSIONS 107 ment that Grim, a baron in the Lay, and no longer a poor fisherman, was spared because he was the "conoissant" of the pirates. Additional lines show the valor of Havelok's party. Having expanded Kelloc's speech for the earlier part of his narrative, the author of the Lay puts into Kelloc's mouth merely a short resume of that speech when she tells Havelok of his birth later on. But Havelok's account of the voyage, as he told it to Sygar, is again Gaimar's version word for word. As another instance of the careful reworking of Gaimar's details, take Havelok's departure from Grimsby for Denmark. In Gaimar's story, Kelloc and her husband Alger advise Have- lok to go to Denmark, and tell him to wait in Grimsby for a favorable wind. They give him clothes, provision the ships, and make the bargain with the merchant. They remain in Grimsby, however, promising to go to Denmark if Havelok sends for them after regaining his kingdom. Havelok thanks them and says that he will reward them. In the Lay, Kelloc's nameless husband, a merchant, himself conducts Havelok. They start for Denmark without delay, and Argentille thanks Kelloc. On arriving in Denmark, they disembark and the merchant husband of Kelloc gives Havelok and his wife clothes, and in a long speech beginning " Biaus filz " tells him what to do. In Gaimar's account Havelok does not even know Sygar's name and probably lines have been left out which tell how Havelok chanced to enter his house at meal time. The author of the Lay tries to make all of this more distinct by having in- structions given to Havelok by the merchant. He is to go to Sygar's : En son chastel va herberger, Et a sa table va manger. Par charite quier le conrei ; Ta femme meine ensemble od tei, Assez tost te demanderont. Par la beaute qen lui verront. Qui tu es, et de quiele contree, Et qi tiele femme tad donee. Havelok leaves the merchant and they go wandering until they come to the city of Sygar. They go and ask him for food and shelter. Argentille sits beside Havelok and attracts much attention at the meal. There are no details invented in the scene. Gaimar mentioned a meal that had taken place io8 THE OLD FRENCH LAI D^HAVELOC before the episode of the throwing of the stones, and he had also mentioned the beauty of Argentille in this connection. Later in the Lay the author uses Gaimar's own words in referring to the meal : Et a sa table avoit mange. Ensemble od lui out este. (741—742) I have selected this passage because it is extremely sugges- tive. It seems to correspond in a general way to the scene in which Havelok leaves Grimsby the first time, since the coun- sels, clothes, and speeches appear in both descriptions. Does it not look as if our author had duplicated here his former episode — an episode which he appears to have invented? Sometimes a single line seems to suggest the same process of readjustment to the Lay. Take for example the two lines which Gaimar uses to denote the place from which Havelok took the " hache " on two occasions. Prent une hache mult trenchant. Ken une meison trova pendant. (539— 54°) These lines of Gaimar's refer to the weapon used at the time of the fight between Havelok and the men who attacked him. In the Lay the author says : Quant Haueloc ad recoueree Une hache trenchante et dure Ne sai par quele aventure Un de ceus la tint et porta. (700—703) It is curious therefore to see that the next time, when the "hache" is used in the house of Sygar at the moment when Havelok thinks he is to be attacked, he finds it in the house: Une grant hache recouera (El paleis pendit par vn croc), (866-867) and it is carefully returned to the same place : II li ad la hache rendue ; Et cil lad au croc pendue. (877-878) In the same scene in Gaimar we read: Pur une hache volt aler, Ke iloc teneit un bacheler. (661-662) The two lines are simply interchanged. A COMPARISON OF THE FRENCH VERSIONS 109 Would it be going too far to see a connection between the lines used by Gaimar in speaking of the departure from Grims- by and those of the Lay? Havelok says to Argentille in G: Diloc turnai quant jo vine ci (308) and to Kelloc: Ci laissai Mun parente quant men turnai. (365—366) And again Gaimar makes Havelok say : Quant furent mort, si men turnai. (591) In the Lay where the circumstances are changed and Grim is represented as still alive, Havelok says : La les lassai quant ieo vine ei, (540) and Quant ieo fui grant ieo men parti. (801) Reviewing these results, I am convinced that the basis of the French Lay is Gaimar's account of the Havelok tale in the Estorie, since no other source could explain the great number of similar passages and the overwhelming number of almost identical lines. The detailed investigation of a few similar passages not only corroborates the testimony given by the many similar lines but suggests a very interesting study — the consideration of the author's ideal of a lay and his method of transferring and transforming Gaimar's material in order to bring it into conformity with that ideal. The Composition of the L,ai d' Haveloc In considering the spirit in which the author of the Lai d'Haveloc adapted the material found in Gaimar's Estorie to a poem with a prologue, an epilogue, and allusions to the Bretons and to an earlier lay which they had made on the subject, we are led to ask what were the ideals of his time and what the literary influences to which he might have responded. But first we should know, as precisely as possible, the date of the Lay. Heretofore we have given the general date men- tioned by Deutschbein and others, the second half of the twelfth century.' It has been shown in the present study that the Lay was not earlier than the second half of the twelfth century, since Gaimar^s Estorie was composed between 1147 and 1 151. Can we draw any conclusions from the poem itself that will show how late in the second half of the twelfth cen- tury the Lay was written ? In the second half of that century the short narrative poems called lays were flourishing in England as they were a little later in France, and as our author chose that form for his story we turn naturally to the consideration of these lays. The old- est and best of them were those of Marie de France, and we know how popular her collection of lays was from the number of manuscripts and translations of them that have come down to us,^ from the testimony of her rather jealous contemporary Denys Piramus,^ and from the imitations of her stories that abound in the later anonymous lays.'^ Although there has been much discussion as to the date of Marie's collection, Warnke, the authority on the subject, de- cided in favor of the year 1165,^ and his date is generally fol- lowed. Warnke thinks also that Marie probably introduced ^ See p. 6. =» Warnke, 2d ed., 1900, Halle, Die Lais der Marie de France y pp. xxxviii-xli. 3 Denys Piramus, La vie Seint Edmund le rei, ed. Ravenel, 1906, Philadel- phia, p. 58. 4 Foulet, Z/. /. rom. Phil.y 1905, XXIX, 19-56, 292-322. 5 Warnke, op. cit. p. xxxv. THE COMPOSITION OF THE LAI D'HAVELOC iii the lay into literature. He says:^ " Keins der anonymen Lais scheint alter zu sein als die Sammlung der Marie ... So ist trotz aller laut gewordenen Zweifel die Ansicht vielleicht doch nicht von der Hand zu weisen, dass Marie die Gattung der Lais in die Litteratur eingefiihrt hat. Nach dem Prolog, V. 28, gedachte die Dichterin wenigstens einen Weg einzu- schlagen, der vor ihr noch unbetreten war." The study of the anonymous lays by Foulet has confirmed this opinion.^ Marie, in the Prologue to her collection, explains what her lays are to be : ^ Des lais pensai qu'oiz aveie. Ne dutai pas, bien le saveie, que pur remembrance les firent des aventures qu'il oirent cil Id primes les comencierent e ki avant les enveierent. Plusurs en ai oiz confer, nes vueil laissier ne oblier. Rime en ai e fait ditie, soventes feiz en ai veillie. (33—42) In Guigemar she says : Les contes que jo sai verais, dunt li Bretun unt fait les lais, vos conterai assez briefment, (19—21) and in the epilogue of the same lay, she adds : De cest cunte qu'oi avez fu Guigemar li lais trovez, que hum dit en harpe e en rote ; bone en est a oir la note. (883-886) But it is in the prologue to Equitan that she explains most fully her idea of the Bretons and their lays. She says : Mult unt este noble barun cil de Bretaigne, li Bretun. Jadis suleient par pruesce, par curteisie e par noblesce des aventures que oeient, ki a plusurs genz aveneient, faire les lais pur remembrance, qu'um nes meist en ubliance. ^ Ibid.f p. xxxvii. 7 Foulet, op. cit. ^ Warnke, p. 4. All the citations from Marie are taken from this second edition of Warnke's. 112 THE OLD FRENCH LAI D'HAVELOC According to Mane's statements then, the ancient Bretons, when an interesting and notable adventure occurred in their midst, made a lay about it which they executed to harp or rote. Marie says that she is about to relate these adventures. This conception of a Breton lay is certainly a curious one, which would have needed a word of explanation from an author to make it understood by his contemporary public. Now if the expression Li Bretun en firent un lai, which she used, had been current in the literature of her time, Marie would not have found so much explanation necessary. On the other hand, if the Lai d'Haveloc had been written before 1165, the author would have had to make some general remarks on the Bretons, their customs, and their lays, in order to have made the phrase clear and effective. What then do we find in Havelok about the Bretons and their lays ? We note the following points : (i) The author himself nowhere speaks of his poem as a Breton lay. (2) He does not say that he himself ever heard this Breton lay which he mentions. (3) The statement he makes is merely the following: the Bretons made a lay about it. Purceo vus voil de lui center, Et sauenture remembrer ; Que vn lai en firent li Breton, Si lappellerent de son non Et Haueloc et Cuarant, (Prol. 19-23) and Li auncien par remembrance Firent vn lai de sa victoire. (EpiL 1 1 02-1 103) (4) He does not say that the lay was a musical one. We may regard this as implied, however, since we know what a Breton lay was from Marie's careful explanation. Without her words the line of Havelok would not be clear. (5) There are four other allusions to the Bretons or Bret- agne in the poem, but these passages do not refer to the Breton lays. The Bretons have had nothing to do with the story of Havelok as we know it from Gaimar, our author's source, and the more closely the line que un lai en firent li Breton is exam- ined, the more external and artificial does it seem. It strikes THE COMPOSITION OF THE LAI D'HAVELOC , 113 one as a mere formula introduced into a poem composed of prologue, epilogue, and "aventure." The conventionality of the phrase has long since been pointed out. Kittredge ^ said of Havelok, "It will not do to be too positive, however, that Havelok ever passed through Breton hands. The question is difficult and has never been adequately discussed. Madden, in his elaborate introduction,'" dismisses it with scant notice. Wright " remarks that the term ' Breton lay * had become almost proverbial, and adds that it is not at all likely that Havelok ever existed in a Breton version." Lot'^ expressed the suspicion that the term " Breton lay " was conventional, "des I'epoque de Marie de France"; and four years later '^ in replying to Brugger in the discussion of the meaning of the words Breton and Bretagne, he referred to his former article and added: " J'aurais du insister sur ce point ... Je trouve encore un exemple du caractere conventionnel de Texpres- sion lai breton dans Haveloc, Tout le monde sait que le Lai d'Haveloc le Danois a pour source une saga scandinave. Cela n'empeche qu'on ne trouve au debut : Un lai en firent li Breton {sic)y It is interesting to find that these scholars felt the lack of sincerity in these phrases although they did not recog- nize the Lay to be a mere reworking of Gaimar*s version. Warnke,''^ although he feels that the existing French Lay may be dependent on Gaimar, does not admit the convention- ality of the expression que un lai en firent li Breton but combats the arguments of Lot. He believes that among the Armori- cans who came to England in the train of William the Con- queror, there were certainly Armorican minstrels, who seized upon the native traditions to enrich their repertories, and that in this way a musical lay, Havelok^ came into existence. He says that in the prologue and epilogue the author speaks with such insistence of a Breton lay called Havelok that it is impos- sible to doubt the existence of such a lay. But Warnke's posi- tion is made less forcible by the fact that he acknowledges that 9 Amer. Journ. of Phil., 1886, VII, p. 184. ^° Madden, ed. pp. v-vi. " Wright, Chronicles of G. Gaimar, ed. for Caxton Society, app. p. 3. '^^Lot, Rom. XXIV, p. 527. ^3 Lot, Rom. XXVIII, p. 42, n. 3. ^4 Warnke, op. cit. pp. ix— xvii. 114 THE OLD FRENCH LAI D'HAVELOC perhaps this musical Breton lay had no direct connection with the narrative Lai d'Haveloc. His answers to Lot are by no means convincing. They will be referred to later in consider- ing the meaning of the word Breton. But whether the Bretons made a lay on Havelok or not, the statement in our Lay that they did so is significant. If our author introduced these words regarding the Bretons into a poem treating matters foreign to them, he must have done so with a purpose. He may have thought the phrase a necessary part of the lay, the literary form which he was adopting. His use of the unexplained conven- tional phrase shows, at least, that both he and his contempor- aries were already familiar not only with the expression itself, but also with the narrative lay — that is, a short metrical romance purporting to narrate the adventure on which the Bretons had made a musical lay. Therefore it seems reason- able to suppose that the Lai d'Haveloc is later than the Lais of Marie, where, as we have seen, the phrase is explained at length. The Lai d'Haveloc in that case is later than 1165. If, however, this conclusion is well founded, can it be that our author escaped Marie's influence? He too wished to tell a story or adventure in the form of a lay, and it is possible and even probable that he here and there followed Marie, since her Lais had become very popular. It is also possible that her success in writing Lais inspired, not his adoption of Gaimar's story, but the idea of treating the material taken from Gaimar according to a certain ideal, following the rules of a certain literary form. Is it possible, then, to find in Havelok the type of Guigemar or some other hero of Marie's Lais ? Traces of the influence of Marie will be sought (A) in the external form of the poem, (a) in the prologue, (b) in the epilogue, (c) in the geography, and (d) in the allusions to the Bretons ; (B) in the author's treatment of the story, (a) in the modifica- tions of actual events of the tale, (b) in the minor details of description or general changes of tone. THE COMPOSITION OF THE LAI D'HAVELOC »i5 (A) The External Form of Havelok (a) Prologue It is interesting to note that the first line of the Havelok prologue is identical with the opening words of Marie's Pro- logue to Guigemar'5 as found in Ms. P.'^ GUIGEMAR Ms. P 1 Volentters deuroit on oir 2 cose quist boine a retenir 3 ki de boine matere est faite 4 mout me poise se nest bien faite 5 Oez, seignur, que dit Marit, 6 qen en son tens nus ne soublit 7 Celui deivent les genz loer, 8 ki en bien fait de lui parler, 9 Mais quant oent en un pais 10 home ne femme de grant pris, 1 1 cil ki de sun bien unt envie I 2 sovent en dient vileinie, I 3 Sun pris li vuelent abaissier : 14 pur ceo comencent le mestier 1 5 ceo est lur dreiz de mesparler. 16 Li contes que sai est verais, 1 7 dunt li Bretun unt fait lor lais, 1 8 Si conterai assez briefment. 1 9 El chief de cest comencement 20 Sulunc la letre e I'escriture 2 I Vos conterai une aventure, 22 ki en Bretaigne la Menur 23 avint al tens ancienur. 1 Volenters deueroit lorn oir, 2 Et reconter et retenir, 3 Les nobles fez as anciens, 4 Et les prouesces et les biens,^7 5 Essamples prendre et remembrer 6 Pur les francs homes amender. 7 Vilainies et mesprisions, 8 Ceo deuereit estre li sermons 9 Dont lorn se deust chastier ; I o Car mult iad mauueis mester. I I Chescuns se garde come pur soi I 2 Lauenture dtm riche roi, 13 Et de plusurs autres barons, 1 4 Dont ieo vus nomerai les nons ; I 5 Assez briefment le vus dirrai,^^ 16 Lauenture vus conterai. 1 7 Haueloc fut cil roi nome, 1 8 Et Cuaran est appelle. 19 Purceo vus voil de lui conter, 20 Et sauenture remembrer ; 21 Que vn lai en firent li Breton, 22 Si lappellerent de son non 23 Et Haueloc et Cuarant. The two prologues resemble each other in the following par- ticulars : (i) their first lines are identical as is the rhyme-word retenir in the second. (2) Each begins in a different manner from Marie's other lays. Instead of giving information about ^5 This was pointed out by Warnke, p. 226, and suggested the closer exami- nation of the two prologues. ^6 Paris, Bib. nat., fran9. 2168, a Picard Ms. of the second half of the 13th century containing Guimar, Yonec, and Lanval. In this manuscript the prologue is incomplete and obscure, but, of course, the author may have had at his disposal a complete and clear manuscript of the same family. ^7 Common to other lays. Cf. Espervier, Rom. VII, p. 3, Car qui bien i voudroit entendre Maint bon essample i porroit prendre (9—10). ^8 Cf Espervier, Rom. VII, 1. 29. ii6 THE OLD FRENCH LAI D'HAVELOC the lay in question, or some remark on lays in general, each of these prologues opens with a moralizing observation. Marie soon becomes personal, names herself and then refers to her slanderers. The author of Havelok makes the banal statement that we should gladly hear and tell of good deeds to make men better, and of evil ones to warn the bad. The actions of Havelok and of the other "barons*' in the story will serve both ends. (3) The sense of these two passages of Havelok and Marie is different, but it looks as if the second part of the reflection in the prologue of Havelok had been suggested by the "vilainie" and "mestier" in lines 12 and 14 of Guigemar. (4) Lines 15, 16, 21 of Havelok correspond respectively with lines 18, 21, 17 of Guigemar. Marie, as we have seen, was explicit just where our author uses a phrase which must have been already current. She carefully explains to the public what he takes for granted is understood by his contemporaries. Evidently she could not have imitated Havelok. Hence these points of similarity between the two prologues point to Marie as the inspiration of the opening lines of the Lai d^ Haveloc. If the other epilogues and prologues of Marie's are compared with Havelok^ there may be found in the latter poem other echoes of her Lais. EQUITAN HAVELOK Jadis suleient par pruesce, Les nobles fez as anciens, par curteisie e par noblesce Et les prouesces et les biens, (3-4) des aventures que oeient, ki a plusurs genz aveneient, faire les lais pur remembrance, qu'um nes meist en ubliance. (3—8) Ki bien voldreit raisun entendre, ici purreit ensample prendre : Essamples prendre et remembrer ( 3 ' 3~3 ^ 4) ^^^ ^^^ francs homes amender. ( 5-6) Havelok (20) Et sauenture remembrer is the same in sub- stance as Equitan (5 and 7). Compare Havelok (14) Dont ieo vus nomerai les nons with Le Fraisne (255) Ici vus numeral sun nun, with Milun (22) Mes jeo ne sai numer sun nun, with Chaitivel (4-5) e la cite vus numerai u il fu nez, e cum ot nun, and (34) mes jeo ne sai numer lur nuns. Several other lines of the prologue are paralleled by lines of Marie's Lais\ 17, 18, 22, and 23 are expressed in the same manner as those in which she insists on the name of her hero THE COMPOSITION OF THE LAI D'HAVELOC 117 or of the lay. None of Mane's heroes has two names, but two of her lays have a double title. Thus the last two lines of the prologue of Havelok remind us of Eliduc : D'eles dous a li lais a nun Guildeluec ha Guilliadun. * Eliduc ' fu primes nomez, Mes ore est li nuns remuez, (21—24) and also of two lines of Chaitivel, * Le Chaitivel ' I'apele hum, e si i a plusurs de eels ki I'apelent *Les Quatre Doels.' (6-8) In the entire prologue of the Lai d'Haveloc, there is scarcely anything original. It is a skillful combination of ideas and expressions gleaned from the prologues and epilogues of Marie's Lais, with especial imitation of the prologue of Guige- mary which furnished the plan for the whole prologue. (b) The epilogue HAVELOK Mult fu de li grant parlance ; Li auncien par remembrance Firent vn lai de sa victoire, Qe touz iors en soit memoire. Ceo fut le lai de Coarant, Qui mult fut prouz et vaillant.^9 (i 101-H06) The sense of this passage corresponds closely with that of Eliduc : De I'aventure de cez treis li ancien Bretun curteis firent le lai pur remembrer, qu'um nel delist pas oblier. ( 1 1 8 i-i 1 84) In her epilogues, Marie twice refers to the popularity of the "aventure." It became known, and a lay was made concerning it by the Bretons. Compare with Havelok (iioi ff.) Laustic\ Cele aventure fu cuntee, ne pot estre lunges celee. Un lai en firent li Bretun, le Laustic I'apele hum, (157—160) ^9 In Ms. P of the Lay, lines 1 105— 1 106 are not found. ii8 THE OLD FRENCH LAI D^HAVELOC and Le Fraisne: Quant Taventure fu seiie coment ele esteit avenue, le lai del Fraisne en unt trove : pur la dame Punt si nume. (533—536) There are traces of the phraseology of other epilogues of Marie. The vfords firent un lai It auncien or // Bretun occur in nearly all of them. Compare Milun: De lur amur e de lur bien firent un lai li ancien. (531—532) The idea of commemorating the adventure and of preserving the memory of it is found in the Lais also : pur les paroles remembrer, Tristram, ki bien saveit harper, en aveit fet un nuvel lai. ( Chiev, 1 1 i-i i 3) In the epilogue of many of her poems, Marie again names the title/°and we find the same repetition in Havelok. In the pro- logue the author had stated that the lay was called both " Have- loc and Cuarant." In the epilogue, he says: Ceo fut le lai de Coarant, Qui mult fut prouz et vaillant. Marie's influence on the epilogue is therefore obvious. (c) Geographical details These details correspond in general to those of Gaimar's text although they are not so abundant. They are as follows, (i) Grim lands at Grimsby, as in Gaimar's account." The author of the Lay adds el North.^'' (2) Alsi's court is at Nicole. Grim has told Havelok to go to Engleterre to the court of a rich king,''^ and we find that Nicole is the seat of the court. 20 Cf. Guig.y Chait., Le Fraisne y Eq., Chiev. y Elid., Laiis.y D. Am.y and Bisc. 21 Gaimar shows by 604, 329, 307, 617 how the words used by Kelloc en cest pais are to be interpreted. "Ceo fut el North, a Grimesbi (125). «3i73. THE COMPOSITION OF THE LAI D»HAVELOC 119 This is not expressly stated in Gaimar, but it may be easily inferred from the Estorie. Sue (Edelsi's) ert Nicole e Lindeseie, (50) E Argentine fu norie A Nicole, e en Lindeseie. (93—94) The author of the Lay was probably desirous of naming a city as the assembling place of the court, just as Marie mentioned Arthur's gathering at Kardoeil,^^ or Marc's at Tintagel.^^ (3) The kingdoms of the two English kings are practically the same, although the terms of the description vary. LAY GAIMAR Un roi qert nome Alsi Li altres out nun Edelsie ; Tenoit en la terre en sa baillie. Sue ert Nicole e Lindeseie ; ^6 Nicole e tote Lindesie, Des Humbre desken Roteland Cele partie vers le north ; Ert le pais en son comant. (49—52) Et Rotelande et Stanford Out cil Alsi en heritage. (194-199) Rutland and Stamford mark the southern boundary of his kingdom, North Lindsey the northern, and these terms describe the same territory as does Gaimar's line: Des Humbre desken Roteland. (4) There is less similarity in the descriptions of the dominion of Ekenbright's. LAY GAIMAR Le roiaume vers les Surois Li altre ert reis de la contree Gouernoit vns autres rois ; Ki ore est Nortfolc apelee. (52-53) Ekenbright out cil rois a non. En Denemarche le regnez (201-203) Aveit quatre riches contez, E en Bretaigne aveit conquis Cair Coel od tut le pais De Colecestre tresken Holland Durout son realme en un tenant. (71-76) The expression vers les Surois is not clear.^^ Heyman says:^^ "If by Mes Surois' is meant the population of Surrey, which in the English translation of the' Lai d'Haveloc is considered doubtful, this would be much further South than is indicated 24 Lanvaly 5-18. 25 Chiev.y 39-40. 26Skeat, ed. 1902, p. xl, **The Northern part of Lincolnshire is called Lindsey." 27 Ward, op, cit. p. 450; and Gaimar's Estoriey II, 221. 28 Heyman, op. cit. p. 6 1 . I20 THE OLD FRENCH LAI D'HAVELOC by Gaimar." It seems possible, however, that vers les Surois means simply toward the people of Surrey or southward from Alsi's kingdom, and that the Surois was brought in to furnish a rhyme with rois. The conception of the author probably does not differ materially from that of Gaimar, since he stated that Alsi returned the kingdom to Argentille : Tote sa terre li rendit Que Ekenbright tint tant come il vesquit. De Holande desqen Gloucestre.29 (1083— 1085) (Ms. P Colesestre) (5) A few of Gaimar's names are omitted, as Cair Coel (74), Teford (80), the place of Adelbright's death, and Colecestre (81), that of his burial. (6) The Lay, on the other hand, contains two items of information not found in Gaimar, that the battle against Alsi was fought at Teford (1027), and that Havelok landed at Carleflure (996). These slight variations do not point to different sources of information, but are arbitrary changes made by the author of the Lay, who in other geographical information conformed to Gaimar's account. He simply transferred the Theford ^° men- tioned by Gaimar (80) to the description of the battle. Perhaps he wished to give the name of the place of combat, because Marie several times gave the locality of tournaments or con- tests.3^ He found no Danish cities mentioned by Gaimar, and did not invent a name for the places where Havelok and Hodulf fought. The one new name introduced by him is Carleflure,^^ the English place where Havelok landed (996). This may be due to Marie's custom of naming harbors in some of the lays : A Suhthamptune vait passer ; Cum il ainz pot se mist en mer, A Barbefluet est arivez ; dreit en Bretaigne en est alez. (^Miluriy 317—320) 29 These lines correspond to Gaimar's Rendu li fiit tut 11 regne, Des Hoiland treska Colecestre (804-805), but had the author of the Lay had a different territory in mind, he would probably have omitted or changed the passage. 3oTetford, not far from Horncastle, England, according to Skeat, ed. 1902, p. xliii, n. 3. 'i^ Lais: Milun(T,^^) El Munt Seint Michiel s'asemblerent ; Chait. (73-74) Que devant Nantes la cite ot un turneiement crie ; in EliduCy the city of Excestre is attacked (91). 32 Skeat, ed. 1902, xlii, n. 2 : ** Possibly Saltfleet, suggests Mr. Haigh. Such, at least, is the position required by the circumstances." THE COMPOSITION OF THE LAI D'HAVELOC 121 It will be noted that the changes which the author of the Lay makes in Gaimar's geography are very slight. They may be due to the influence of Marie^s Lais. (d) The allusions to the Bretons The narrative of Gaimar contains only one allusion to the Bretons ; the English Lay does not mention them ; the Lai d'Haveloc refers to them four times as follows: 1 Quant Arthur out finie sa guerre, Hodulf dona tote la terre Puis sen ala od ses Bretons ; (40) 2 Nicole et tote Lindesie, Et Rotelande et Stanford Out cil Alsi en heritage ; Mes il estoit Bret par lignage. (200) 3 Fetes la loignz enmener En Bretaigne, dela la mer, Et a vos parenz comander ; (313) 4 Cuaran lapelloient tuit ; Car ceo tenoient li Breton En lur language quistron. (259) Since Gaimar's story furnished the material for the Lay, it is evident that three of these passages must have been intro- duced arbitrarily by the author of the Lay. Gaimar gave a suggestion for 2, but the sense of Breton in the Lay and in Gaimar seems to be slightly different. Gaimar, who had dis- cussed the Danes and Angles in England, and had mentioned the Angles as responsible for the change of the name Bretaigne, wrote of the reign of Arthur's nephews, and added that the Danes hated them. He then proceeded to relate the story of two kings who reigned in England at that time, the one a Dane, the other a Breton — a Celtic inhabitant of England in Arthur's time. The author of the Lay does not say that Adelbright was a Dane. He does not refer to Constantine or give a clue to the period. He follows in general Gaimar's geographical in- formation concerning the extent of the two kingdoms, and adds in regard to that of Alsi : Mes il estoit Bret par lignage. 122 THE OLD FRENCH LAI D'HAVELOC The expression "par lignage'* strikes one as perhaps imply- ing a shade of meaning different from that of Gaimar's unqual- ified "Bret," because ofMes; it is possible, however, that the phrase applied to Adelbright in Gaimar's account suggested the words in the Lay. Adelbright was "del lignage as Daneis," and the phrase may have simply been transferred to Alsi in the Lay. Even if the phrase is to be explained in this way, the entire line still leaves an impression that makes us suspect that Gaimar's conception of" Bret," a Celtic inhabitant of Eng- land, is not quite that of the author of the Lay. We could think that Alsi was a Celt living in England in Arthur's time unless told that he was a Dane or an Angle. It is therefore with surprise that we note the word Mes, which apparently denotes that there is something unusual in the statement fol- lowing. What is then the idea of " Breton " conveyed in the Lay by the author, who may have been a Norman, an Anglo- Norman, or a Frenchman who was living in England in the twelfth century ? ^3 The question of the meaning of " Breton " in the twelfth century is one which has given rise to heated discussion. It is perhaps best to recall briefly the main lines of the contro- versy before applying the results of it to Have/ok, As Warnke points out in his preface to the Lais of Marie de France, stu- dents of Old French literature gave the words Breton and Bretagne the natural interpretation of contemporary Armori- can, without further thought, until Wolf,^^ Gaston Paris,^^ and Bedier^^ questioned whether the words did not also denote the inhabitants of England who were Celts, especially the Welsh. There came to be three groups of scholars, those who inter- preted the words as Armorican and Armorica, those who in- terpreted them as Welsh, and those who combined the two theories holding that the words were applied to both the Armorican and the Welsh. Zimmer,37 in reviewing the article in wl^ich Gaston Paris 33 Suchier says of the author of Have/ok (op. cit. p. i 24) : ** Er dichtete viel- leicht in England, jedoch in reinem Franzosisch, war also vom Festland gebiirtig." 34 Wolf, Ueber die LaiSy Sequenzen und Leichcy Heidelberg, 1841, p. 251. 35 Gaston Paris, Romania y VIII, 36; Histoire litt. de la France y XXX, p. 3; La Litterature normande avant P annexion y Paris, 1889, p. 14. 36Bedier, Les Lais de Marie de France y Revue de deux mondesy 18 Oct., 1891, 835-864. 37Zimmer, Gbttingische gelehrte Anzeigeny 1890, I, 794 ff. THE COMPOSITION OF THE LAI D'HAVELOC 123 expressed his views, after a study of many of Marie's Lays, asserted that it was impossible to derive the "matiere de Bre- tagne" from either the Cymri or the Welsh. But it became necessary for him to show how those lays which had their scenes laid in England could have come from the Armoricans, whom he considered the only Celts concerned with the trans- mission of Breton material. He showed that Brittany and Normandy had been closely united politically from the tenth century, and pointed out that near the boundary line of the two provinces there existed a speech district where Norman and Breton were both spoken. He also noted that many Bretons, among them some of these bilinguistic Bretons, had gone to England with the Normans. They were con- versant with French before going to England and had trans- ferred their native legends to the Normans and Anglo-Normans on the continent and in England. Zimmer also indicated how they confused the geography of the Breton material by trans- ferring the scenes of their own legends to England. Lot 3^ disagreed with Zimmer about many of the Lays. He examined them again, noting the context, and decided that a large number of them could not be adapted to Zimmer's theory. They were in part Welsh. Bedier39 more recently explained how these Armoricans came into possession of the Welsh legends, and stated that they recognized the relationship existing between these tales and their own, and combined them so that through their work the " matiere de Bretagne " is the product of the fusion of the Armorican and Welsh legends. The second stage of the discussion was reached when Brug- ger^° defined "Breton" with more precision. According to him there were in the twelfth century two uses of the word : (i) the ordinary meaning of a contemporary Armorican; (2) the archaic sense of a Breton of Arthur's time. Brugger does not especially discuss the archaic sense of the word in the rest of his article which is devoted to proving that " Breton" in the twelfth century documents examined meant a contemporary 38 Lot, Etudes sur la provenance du cycle arthurieny Romania y XXIV, 497- 528, XXV, 1-33. 39 Bedier, Le roman de Tristan par Thomas , Soc. d. anc. textesfr.y II, 127. 40 Brugger, Vber die Bedeutung von Bretagne y Breton, Zeitschrift fur fran- xosische Sprache und Litteratur, 1898, XX, 79-162. 124 THE OLD FRENCH LAI D'HAVELOC Armorican. However, he put the whole controversy on a diflFerent basis by introducing a time element. It was no longer enough to distinguish between a Welshman and an Armorican ; one had to decide between a twelfth century man and a Breton of Arthur's time. Lof*' was quick to see the bearing of this on Brugger's theory of the interpretation of the word as Armorican, and, applying the new definitions to the Lays, he contended that the word Bretons in this literature certainly had the vague, retrospective sense of Bretons of Arthur's period. He now proclaimed with confidence the conventionality of the term " breton lai," which he had suggested before. Warnke,^^ in his introduction to the lays of Marie de France, opposed Lot. His arguments do not seem to me convincing. 41 Lot, Nouveaux Essais sur la provenance du cycle arthurieny Romania^ 1899, XXVIII, 1-48. 42 Warnke, op. cit. pp. ix-xi. To substantiate his theory. Lot gave the fol- lowing reasons (pp. 41—42, n. 4). (i) The words ** Bretons en firent un lai " is always in a past tense. (2) In passages of Eliduc and Milun the Bretons are specified as **li ancien.*' (3) Marie's words confirm this view. Or sont dites e racontees, de latin en romanz trovees ; Bretons tn firent lais plusors. Si con dient nos ancessors. Warnke (pp. ix— x) objects to the first reason on the ground that only a past tense could be used whether the Bretons were contemporaries or ancient musicians. In cases where a present tense could be used, it was employed. Cf. Guingamor (678), Einsi I'apelent li Bretun, and Tydorel (489) : Cest conte tienent a verai li Breton qui en firent le lai. These Bretons must be contemporaries. In the second place Tydorel offers an instance of Lai nouvel, thus showing that no importance can be attached to these expressions which the author used arbitrarily to enhance the value of his poem. Thirdly Warnke says the lines quoted by Lot are not Marie's but come from Tyolet (33 ff. ), and like other lines of the prologue do not contribute testimony of weight since the author of Tyolet probably did not use the **harfenlied " Tyolet for his poem. Foulet (Marie de France et les lais bretonsy p. 320) criticizes Warnke's first two points. He says that **einsi I'apelent li Bretun" may well refer to a written source and denote a time **tres recule dans le passe." Com- pare the last line of the Tyolet quotation. He also states that no information regarding the Bretons can be drawn from Tydorel since it is a late lay, much more recent than those of Marie. In his third objection, Foulet says that Warnke was right in giving Tyolet as the source of the quotation, and in denying that the words are a legitimate source of information about the Bretons. On the other hand, it seems to me that a kind of testimony in favor of Lot's theory does emanate from these lines of Tyolet. Foulet (pp. 48-52) shows THE COMPOSITION OF THE LAI D'HAVELOC 125 He took into consideration those lays which had heretofore been omitted or merely mentioned in the discussion. The one which he had most difficulty in adapting to his theory was the Lai d'Haveloc, because he had to deal there with a Scandina- vian legend entirely foreign to the Bretons, and one which originated in the north of England. How could contemporary Armoricans have transmitted that legend to the Anglo-Norman author of the Lay.? Warnk:e43 had recourse to what seems to be for Havelok a forced and unnatural process. The steps of it are as follows : Armorica and Normandy were closely united politically and geographically, and Armoricans shared in the conquest of England. In 1066 they fought at Senlac with William the Conqueror. They were rewarded with fiefs after the victory and settled in many parts of England, especially in Yorkshire, and in that part of northern England which was the locality of the Havelok legend. Among these Armoricans there were doubtless minstrels who must have come in contact with the Havelok story. It is undeniable that other minstrels came over in the next century. These would have come in contact with the earlier ones, who had gone to Yorkshire. The older generations of minstrels would have transmitted to these successive generations of twelfth century minstrels English legends with which to enrich their repertories. In this way the story of Havelok would have come into the pos- session of these twelfth century Armoricans. The whole explanation is far-fetched. It is complicated and consists of hypothesis after hypothesis offered in an effort to save the Armorican theory by showing that it applies to Have- that Tyolet is a compilation of an episode of Tristan, modified by Marie's Lanval, with parts of Chretien's Graal. The prologue is a servile imitation of Marie's prologue including a mistaken interpretation of her words. The author does not state that he has heard the lay Tyolety or that he knew the Bretons made one. He says he has it at second hand *'si com dient nos ancessors." The ancessors in this case were invented. With this author, at least, the phrase was an entirely conventional one. 43 Warn ke, op. cit. pp. xv— xvii. 126 THE OLD FRENCH LAI D'HAVELOC lok too. To show that Armoricans transmitted the material in the twelfth century, it was necessary to suppose first that the Armorican minstrels who penetrated to the part of England where the Havelok legend was found came into contact with the legend at the time of the conquest. But these Armoricans were not twelfth century men, so that we must conceive of their passing the legend on to generation after generation of newly arrived Armorican minstrels who finally gave it to the Anglo-Normans who arrived in the twelfth century. But a vital part of Warnke's conception of the lays is the relation between the narrative lay and the musical lay of the Bretons on the same subject. Warnke therefore added the statement that Havelok was also in all probability the subject of a musi- cal lay. He said:^'^ "So wird es denn in der That, wie der Dichter des franzosischen Gedichtes behauptet, ein bretonisches Harfenlied von Havelok dem Danen gegeben haben, wenn auch die franzosischen Darstellungen nicht direkt auf dieses Harfenlied zuriickgehen sollten." So the musical lay which, "probably existed" has perhaps no connection with our lay. To say that there was such a lay simply because it is stated in this poem that the Bretons made a lay on Havelok, seems to be forcing a point, and to add that this musical lay had perhaps no connection with the narrative lay weakens the whole argu- ment. Warnke had, as we can see, a hard position to maintain in the case of the Lai d' Haveloc^ because this poem is not com- patible with his theory of the Breton lays. But if Warnke has to yield to Lot and agree that in the case of Havelok, at least, the expression Li Breton en jirent un lai is purely conventional, his whole theory is in danger. If the words are artificial in one lay, doubt creeps in as to the sincerity of them in other lays. Is the testimony given by the allusions to the Bretons in Have- lok in favor of Lot's theory of the vaguely remote Bretons of Arthur's kingdom, or does it confirm Warnke's interpretation of contemporary Armorican ? One may judge from the follow- ing analysis. Reviewing the allusions to the Bretons which are contained in the Lay, we see that the sense of 2, that Edelsi was Bret par lignage, becomes clear from 3. The counsellors advise that Argentine be sent from England to " Bretaigne dela la mer." 44 Warnke, p. xvi. THE COMPOSITION OF THE LAI D'HAVELOC 127 This must be Armorica, and as her relatives are there, Edelsi was probably also an Armorican. Bret is used here in the ordinary sense of contemporary Armorican. Point I does not lend itself to this meaning. Arthur went away from Denmark od ses Bretons — the old Arthurian Bretons of a distant past. Hence we have here the second meaning. It might be possible to interpret 4 as Armorican, because, since the first line reads Cuaran V appelloient tuit, one might expect that tuit meant those surrounding Edelsi, the Armori- can. On the other hand, the references may be to the vague old Bretons, who made a lay on the subject, and whom we know as li auncien^^ It might be imagined that the name Cuaran would give us a valuable clue. Curiously enough no word Cuaran correspond- ing to quistrun has been found. Skeaf^^ says: "This surname is Celtic; and Anlaf Curan signifies 'Anlaf with the brogue'; from the Irish and Gael, cuaran^ explained by Macleod as 'a sock, a brogue of untanned leather or skin, commonly worn with the hairy side outwards,' cf. Welsh, cwran^ a buckskin. The surname is easily explained from Anlaf's connexion with Ireland. . . . This epithet is important, as it is the very one ap- plied to Havelok in the French versions of the story. Gaimar spells it Cuheran, and adds (1. 105) — ' Cil Cuheran estait quis- trun,' i.e. This Cuheran was a scullion, or kitchen-servant; precisely as in our poem, 11. 903-970. The author of the other French version . . . somewhat mistakes the matter, imagining that Curan had the meaning of ' scullion,' which is not the case. . . . This is, of course, a slip; but the Celtic origin of the name is nevertheless perceived. It does not, however, occur in the English version." A comparison with Marie's Lays, however, explains the pas- sage, and shows that there was probably no misunderstanding on the part of the author. Gaimar's lines and the hero's two names gave an opportunity to explain the names in a passage doubtless inspired by the following lines of Marie's. Bisclavret a nun en Bretan, Garulf Tapelent li Norman. (^Bisclavret y 3-4) 45 Havelok y 1 102. 46Skeat, ed. 1902, p. xxxvii. 128 THE OLD FRENCH LAI D'HAVELOC Une aventure vus dirai, dunt li Bretun firent un lai. Laiistic a nun, ceo m'est vis. Si Tapelent en lur pais; Ceo est russignol en Franceis E nihtegale en dreit Engleis. {Laustic, 1—6) * Gotelef* I'apelent Engleis, * Chievrefueir le nument Franceis. ( Chiev. 1 1 5-1 16) En Bretanz I'apelent Lanval. (^Lanvaly 4) We have evidently in this fourth allusion, another fiction concerning the Bretons/^ If our author could insert in his poem a passage such as this, the probabilities are that the language to which he makes allu- sion was an old one and not the contemporary idiom of the Welsh or Armoricans. Furthermore the fact that he has used indiscriminately the different meanings of Breton shows that he felt no distinction, and that the phrases were for him purely artificial. It is interesting to note that these conventional references to the Bretons in the Lay have in turn influenced opinions con- cerning the saga. Deutschbein,^^ who believes in a derivation of the saga independent of the Bretons, deduces from the greater Breton coloring which characterizes the Lay, that the Bretons (Armoricans) " Bretonized " the material slightly as they handed it on to the Anglo-Normans.^^ But it is now clear that, whatever the relation between the Bretons and the saga of Havelok^ no evidence concerning them is to be drawn from the Lay. The " Bretonizing " in the Lai d'Haveloc is the effort of one man to give Breton coloring to a finished literary product, in order to make it conform to a cer- tain ideal which he had in mind — the type of lay which Marie de France composed. He introduced the allusions arbitrarily, 47 The author of Havelok has imitated Marie in a mannerism of hers which is, in turn, probably due to the influence of Wace over her. Cf. English Words in the Lais of Marie de France y Modern Language Notes y 1905, XX, 10 9- ill, by L. Foulet. 48Warnke (p. xv) says that Arthurian Bretons would not have composed a lay on their enemies. Deutschbein (p. 140) says that the Armoricans had noth- ing to do with the legend or they would not have made Edelsi their compatriot. 49 Deutschbein, op. cit. p. 148. THE COMPOSITION OF THE LAI D'HAVELOC 129 and they cannot be considered as giving trustworthy informa- tion regarding either the Bretons or their lays. The only point of positive testimony, then, to be drawn from the study of Havelok in this connection, is that the ex- pression Li Breton en firent un lai is in one case, at least, as conventional as Lot suggested that it was. And if the words were used in Havelok without question, is it not prob- able that they had no more definite meaning in other works of the period? I realize, of course, how easy it is to find what one is seek- ing in the way of literary connections and I do not claim that any one of the following suggestions has great specific value. But a series of such resemblances assuredly appears signifi- cant, and especially so since we have already seen the author*s imitation of the prologues and epilogues of Marie and his slavish use of Gaimar. (B) Marie's Influence on the Author's Treatment OF THE Story In the comparison made between the French Lay and Gaimar, numerous variations in events and in details of the story of Havelok were observed. To weigh these differences and determine the author's method of composing a Breton lay from his borrowed material will be the next step in the study of the poem. It will be necessary also to keep the Lais of Marie in mind, and to note any suggestions of her influence, should such appear, since we know from the investigation of the external form, of the Lay that the author had these poems in mind as he wrote. (a) Modifications of important events of the story (i) Havelok's departure from Grimsby (157-192). Com- pare with this : (a) Milun : La dame ki lur fiz nurri, (tant ot este ensemble od li qu'il esteit venuz en ee), a chevalier I'a adube. Mult i aveit gent dameisel. (289-293) I30 THE OLD FRENCH LAI D*HAVELOC She then tells Milun of his father; he is told to show his prowess: "fors de la terre e del pais." His leave-taking is de- scribed in the following words : II ne demure fors le seir ; el demain aveit pris cungie. La dame Pa mult chastie e de bien faire amoneste asez li a aveir done. (312-316) * Compare also : (b) Guigemar, Marie gives Guigemar's name, saying that he is handsome and beloved of father and mother. He is sent to court to serve the king. Quant il le pout partir de sei, si I'enveia servir le rei. Li vadlez fu sages e pruz ; Mult se faiseit amer de tuz. Quant fu venuz termes e tens que il aveit eage e sens, li reis I'adube richement ; armes li dune a sun talent. Guigemar se part de la curt ; En Flandres vait pur sun pris querre. (41—51) The details of these scenes are changed to suit the circum- stances in which Havelok finds himself Since degradation is to be his lot at court the knighting of Havelok is postponed, but the outlines of the parting are the same — a short descrip- tion of the youth, the instructions given him, the gifts pre- sented, and his departure.^" Grim's instructions are practical, bearing on his life at court. A tote gent te fai amer (180) may be due to the influence of Gaimar (44). It may, however, be derived from Gaimar's description (139-140), from which pas- sage nearly all the other details of Grim*s counsel are drawn. The equipment of the knight, adapted to the situation, becomes a new suit of clothes. . (2) The scene between the barons and Alsi. Gaimar states merely that Edelsi covets the kingdom and marries his niece to his " quistrun " Cuheran. The question arises : how could he do such a thing with impunity ? The 50 Cf. also description in Milun, Mult i aveit gent dameisel (293). Also Guigemar (38-40), el reialme nen out plus bel. A merveille Tamot sa mere, c mult esteit bien de sun pere. THE COMPOSITION OF THE LAI D'HAVELOC 131 author of the Lay explains this by the invention of a scene between Alsi and his barons, a scene which was probably sug- gested by one of the Lais of Marie. She, in fact, makes fre- quent allusions to the barons assembled at court.^' In Lanval the barons play a prominent part in the trial scene. Although the character of this scene of Lanval differs totally from that of Havelok, yet the general idea of the court scene is carried out in Havelok, In Havelok the barons assemble to make a demand of the king; he asks for time to consult his advisers and another day is set for the meeting: Terme lur mist et ior noma, A repairer les comanda Quant il se serra conseillez. (299—301) His counsellors offer a suggestion, but he tells them his pur- pose and announces that he will brook no resistance. He comes with them to the assembly hall where he has had guards stationed, "au ior qil out a ceus nome." He then makes a speech to the nobies who murmur but are restrained by the presence of the guards. In the lay of Lanval the king sends for his men to judge Lanval, who has said that his lady is fairer than the queen (384). They meet and decide, as follows: que Lanval deit aveir un jur, mes pleges truisse a sun seignur qu'il atendra sun jugement e revendra en sun present ; si sera la curz enforciee, kar dune n'i ot fors sa maisniee. (391—396) The barons return, announce their decision (397-398); on the appointed day they reassemble : Al jur que cil orent nume, li barun furent asemble. Li reis e la reine i fu. (417—419) They proceed to judge Lanval : Li reis demande le recort sulunc le cleim e le respuns : II sunt al jugement ale. (426—429) 51 Bisc. 186, Chiev. 40, and Lanv, 132 THE OLD FRENCH LAI D'HAVELOC Some wish to condemn Lanval to please the king, but they decide that he must present his lady that they may judge. A deputation announces this decree to Lanval, who says he can- not comply with the command. The king urges his judges to decide. They are interrupted, urged again by the king, inter- rupted a second time, and, after a third summons from the king, they give the verdict, acquitting Lanval.^^ In spite of the divergence of the subject matter, there is a correspondence between the general outlines of these two scenes. Possibly another of Marie's Lais,Le Fraisne, may have suggested something to the author of Havelok concerning the assembly of his barons. They insist that Alsi marry Argentille, who is now of age, to a suitable person, as he had promised to do. In Le Fraisne, the subjects demand that Garun should marry. Lungement ot od lui este, tant que li chevalier fiefe a mult grant mal li aturnerent. Soventes feiz a lui parlerent, qu'une gentil femme espusast e de cele se delivrast. Lie sereient, s'il eust heir ki apres lui peiist aveir sa terre e sun grant heritage Ja mes pur seignur nel tendrunt ne volentiers nel servirunt, se il ne fait lur volente Li chevaliers a graante qu'a lur cunseil femme prendra. (323—339) (3) The chamberlain. Neither chamberlain nor hermit appears in Gaimar. This character of chamberlain is prominent, however, in Horn, Tris- tan and the Lais. Cf Guigemar, (579 ff.); Le Fraisne, (403 ff.); and Eliducy in which Gilliadun's chamberlain advises her {23S~ 455) and in which Eliduc's serves him (767-805). The second mention of the chamberlain in Havelok corresponds closely to this passage in general tone. Sygar sends him to see Havelok's flame instead of going himself. ^^ Argentine's chamberlain aids her in her trouble as Gilliadun's does in Eliduc. ^^ ^"^ s^ Lanvaly ^1^-6^6. -v "^-^ .^ i 53 Vn son priue iad mande X ^ ^ Pur saueir quant cil dormira, etc. (830-831) Le chamberlenc out grant poour. (839) THE COMPOSITION OF THE LAI D'HAVELOC 133 (4) The hermit. Eliduc contains also the mention of a hermit. This hermit is dead so that the story has nothing to do with him, excepting that Eliduc had turned to him for help, and, on finding him dead, had left Gilliadun's body in his cell. His chapel is the scene of an important part of the story. The description of the hermit is as follows : Une forest aveit en tur, trente liwes ot de lungur, Uns seinz hermites i maneit E une chapele i aveit ; quarante anz i aveit este. A la chapele sunt venu, Apele i unt e batu ; Uit jurs esteit devant finiz li seinz hermites, li parfiz. (889-893, 91 1— 912, 917-918) It may be, however, that the development of the part of the hermit is due to some other literary work of the period in which hermits figured more extensively. (5) Havelok is dubbed knight. The knighting is performed by Sygar (928). Cf. Guigemar (47); Yonec (470); Milun (10, 292). (6) The single combat with Hodulf The single contest replaces the battle between the two hosts, described by Gaimar (740-744) : 'Ensemble vindrent li baron, Requistrent soi come leon. Haueloc fut de grant vertu ; Le roi Hodulf ad si feru Dune hache qil apporta, Quil labatit, puis ne leua, Iloec loccist deuant sa gent. (959—965) If we compare these lines with the description of Milun's con- test with his father we note a general resemblance. The out- come as well as the attendant circumstances of the two duels are, of course, very different. 134 THE OLD FRENCH LAI D'HAVELOC Milun le vit si cuntenir. Si bien puindre e si bien ferir : El renc se met encuntre lui, ensemble justerent amdui. Milun le fiert si durement, I'anste depiece veirement, mes ne Taveit mie abatu. Cil raveit si Milun feru que jus del cheval I'abati. (409—419) It is evident, therefore, that certain innovations which the author of the Lay has introduced into Gaimar^s material, can be explained by a desire on his part to imitate the general style of certain scenes of Marie's Lais. These scenes, we have noted, were those representing the departure of a young page, the gathering of barons at the court, the part of such personages as the chamberlains of the court or castle, and the single combat of knights. The interesting fact which this comparison has shown is that these scenes are all of the same type, inasmuch as they all afford glimpses of the important events in the life of the time in an aristocratic and courtly environment. They are scenes from Marie's Lais which contain the courtly setting of the adventure. The resemblances between the passages in the Lais and in Havelok, we have seen, were general resem- blances, sometimes involving more than one of her Lais and often differing in actual circumstances. It is impossible to say that this or that passage of Havelok is imitated exactly or directly from a certain one of Marie's Lais. What is unde- niable is that the atmosphere of her Lais has been borrowed, and that the sturdy hero of Gaimar's story in becoming the hero of a lay, has been transplanted into a more courtly en- vironment, where he moves as another Guigemar. It is curi- ously interesting to find that our author in his endeavor to make a narrative lay connected with the Bretons, has borrowed from Marie exactly what was most French and least Breton in her Lais — i.e. the more or less faithful description of the customs and life of the French or Anglo-Norman aristocratic society of her own time. To obtain this courtly coloring, our author has occasionally altered details in his material, and has presented scenes which were not in his source, but which could easily have been inspired by the general tone of the Lais, THE COMPOSITION OF THE LAI D'HAVELOC 135 (b) Details of description The changes made in the Lay to modify the general tone, by the addition or suppression of details, are not such as corre- spond to any special lay of Marie's but reflect rather the court- ly tone of the whole collection. Sometimes such changes have affected an important alteration in the poem ; sometimes they have merely softened and refined a description. Such modifi- cations of tone are the following: (i) Grim is made a baron instead of a fisherman. (2) The queen is buried beside the king. This is an instance of the care to show chivalrous deference to woman, and to mention her where it is possible. (3) The treatment of the wedding night is much more re- fined than Gaimar's. Argentille does not repine and complain to the king of Havelok after her marriage, as she does in Gaimar's account. Havelok turns from her but it is because he wishes to hide his flame from her. The author adds that they were gradually reconciled to each other: Mes puis sasseurerent tant, Et par parole et par semblant, Qu'il ama. (4) Gaimar's passage in which Kelloc and her husband dis- cuss the wisdom of telling Havelok about himself is omitted, and simply reflected in the one line of Kelloc's speech : Bien te ciele, si ieo le te di (592). This hesitation on the part of Kelloc, and the tone which she used in Gaimar's account, when she spoke of Havelok, was not that of a respectful vassal. (5) Kelloc's husband himself accompanies Havelok to Den- mark. In Gaimar's story, Kelloc and her husband equip him and send him with Don Alger the merchant. Kelloc says to him in the Lay : Mon seignur vus i conduiera, Dedenz sa nef vus passera. This also shows more respectful deference to a feudal lord. (6) Sygar himself first does homage to Havelok, thus show- ing the same deference. In G he is the last. (7) In the scene of the shipwreck, the author of the Lay emphasizes the fierceness of the attack and the valor of those with Havelok. 136 THE OLD FRENCH LAI D'HAVELOC (8) Argentine is given a more prominent part in details un- important in themselves. (9) Other modifications have in turn resulted from these. Grimes high estate may account for his guardianship, although it is more likely that his position of guardian brought about his change of estate. His rank is certainly responsible for the invention of the castle, the plans for flight, and the scene of the embarcation. Certain other details added to the narrative are undoubtedly due to the author's own fancy and invention because he wishes to enliven his story, or make it more clear and connected. Compare for example his description of the meal at Sygar's, which Gaimar merely mentioned in a later part of his account, or Alsi's jests, or the allusions to the founding of Grimsby. These changes abound and are sometimes not free from in- consistency or absurdity, as in the case of Baron Grim, already referred to, who incongruously caught fish in Grimsby si com il soloity and who was spared by the outlaws because he was their conoissant. The same thing is true of the dream, which repre- sents the events which were to befall Havelok. The author of the Lay changed these points, so that the battle became a single contest, but the dream he neglected to change ; the foxes who represent Hodulf's men in the dream are destroyed, as in Gaimar's version. THE COMPOSITION OF THE LAI D'HAVELOC 137 CONCLUSION These modifications may be largely due to Marie, but they are not exclusively the result of her influence. Since we know, however, the debt the author owed Marie in other ways, it is not assuming too much perhaps to give her Lais credit for the greater part of such changes in general tone as we have just been considering. The author's taste and environment and other literature of the period contributed the rest. The composition of this Lay proceeded differently from that of the majority of the anonymous lays already examined by others. All but a few of these lays were the product of the more or less skillful fusion of two or more lays of Marie's collection, sometimes with changes invented by the author, or with the introduction of material derived from another anonymous lay, or from a source such as 'Tristan, Such stories were furnished with prologue, epilogue, and references to the Bretons closely copied from Marie. This Lay has the same external form, but it has as its source one definite poem, which was reworked according to the fancy and poetic ideal of the author. Our author has sometimes marred the beauty of the prim- itive details of Gaimar, as we have observed. Especially is this shown in his making Grim a baron, and replacing the poetic flight of the queen to the shore with her son, and the protection given her by the fisherman Grim, by the inconsis- tent scene of the guardianship of the baron Grim and the formal embarcation, details which are not in harmony with the prim- itive features of the story. This general contrast between the artificial, courtly atmos- phere of the work and the primitive traits of the story, espe- cially when the author fails to carry out his modifications consistently, is often wearying. He has, however, on the whole not injured his material by his treatment of it. The clearer, more polished and graceful Lay is a better literary pro- duct than Gaimar's account, which is often crude and obscure, now giving unnecessary detail (as in the case of the enumera- 138 THE OLD FRENCH LAI D'HAVELOC tion of the fishes caught by Grim), now condensing until pic- turesque effect or even clearness is lost. When we compare the Lay and Gaimar's version we see that the writer has with the courtly tone and coloring of the Lais of Marie apparently imbibed something of her grace and ease. We see too that he had more talent than Gaimar for this kind of narration. In justice to Gaimar it must be said, however, that his story of Havelok is but a small episode in a long work, whereas the Lay is a short poem studied in every detail and elaborated with the greatest care. The method followed by the author of Havelok suggests a question concerning his predecessor in this form of literature. If this poet, who presumably wrote the Lai d'Haveloc not long after the appearance of Marie's lays, composed a lay in this fashion, and nevertheless asserted firmly that the Bretons made a lay on Havelok, may it not be that our faith in Marie's own statements of the same kind do not always rest on a solid basis P We see from the study of the later anonymous lays and from Havelok that this type of literature — the lay — practically begins and ends with Marie. May she not, at least occasion- ally, have used material as foreign to the Bretons as is pur story, and may she not have made the same assertions as did the author of Havelok, with no better foundation for them ? In any case, since Marie is left in a more and more isolated position, as the conventionality and insincerity of the allusions to the Bretons and their lays become known, we cannot escape the fear that some day our faith in her words about the Bretons and their lays will be as rudely shaken as has been our trust in the statements of the author of the Lai d'Haveloc. Vita I was born in Harrisburg, Pennsylvania, November i, 1872. My father was Wallace Weir Fahnestock, and my mother, Mary Knight Nutting. After completing my college preparatory work in the schools of Cleveland, Ohio, I entered the College for Women, Western Reserve University, where I was graduated in 1894. From 1894 to 1896 I studied abroad, chiefly at the University of Zurich (where I was a student three semesters), and in Paris. In 1897 ^ was Fellow in Romance Philology at Bryn Mawr College, and I returned to Bryn Mawr in 1 901-1902, and 1906-1907. My graduate study was pursued principally under Professor Morf and the late Professor Ulrich, both formerly of the University of Zurich, and under the late Professor Menger, Professor Foulet, Professor De Haan, and Professor Holbrook at Bryn Mawr. My major subject is French Literature, my first minor Old French Philology, and my second minor Spanish. My thesis is A Study of the Sources and Composition of the Old French Lai d'Haveloc. I taught for seven years in the State College for Women, Columbus, Mississippi, where I had charge of the department of Modern Languages, and I have been during the past year instructor in Romance Languages at Mount Holyoke College. It would be impossible adequately to .thank Professor Morf, Professor Foulet, and Professor De Haan. I am especially grateful to Professor Foulet for his most helpful criticism of this dissertation, which was begun at his suggestion. Edith Fahnestock. Mount Holyoke CollegSy May, 1908. MAR 7 1939 'M^iSiem^sM UNIVERSITY OF CALIFORNIA LIBRARY \ BERKELEY ^ Return to desk from which borrowed. This book is DUE on the last date stamped below. ^1r, AN 28 1957 a ^ ^OMiff ^\^^' l50ct63tS JjEC'D LD 5May'55nj P" 151969 5 6 REC'D LD DEC J/iW 2 6 197? 85 lb '69 -10 All wSih'^m LD 21-100m-ll,'49 (37146816)' ..P'flW'^MAR?' 72 .4PM 4 8 ^ur IVERSITY OF CAUFORNIA LIBRARY •i ^ if!?