: ■F87 HART M€M0f«AL tr&RARY CORNELL UNIVERSITY LIBRARY ENGUSH COLLECTION THE GIFT OF JAMES MORGAN HART PROFESSOR OF ENGUSH (\,aq^n^o THE PROBLEM OF THE TWO PROLOGUES TO CHAUCER'S LEGEND OF GOOD WOMEN BY JOHN C. FRENCH A DISSERTATION PRESENTED TO THE BOARD OF UNIVERSITY STUDIES OF THE JOHNS HOPKINS UNIVERSITY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY 1905 BALTIMORE J. H. FURST COMPANY 1905 Cornell University Library The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/cletails/cu31924013110832 THE PROBLEM OF THE TWO PROLOGDES TO CHAUCER'S LEGEl OF GOOD WOMEN BY JOHN C. FEENCH A DISSERTATION PRESENTED TO THE BOARD OF UNIVERSITY STUDIES OF THE JOHNS HOPKINS UNIVERSITY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY 1905 BALTIMORE J. H. FURST COMPANY 1905 M f\.'^'\ll'^o CONTENTS. A. THE PKOBLEM IN GEjSTERAL. I. Introduction ] II. The Hlstory of the Problem 3 III. Chaucer's Motive for Revising the Prologue 11 IV. The Ballad 22 V. The Plan op Legend and the Structure of the Pro- logue , 28 B. THE TWO VERSIONS COMPARED IN DETAIL. I. Lines Transposed in the Revision 40 n. Lnras Peculiar to the G-Veesion 45 III. Lines Peculiar to the F-Version 64 IV. Lines Partly Identical in Both Versions 7-5 V. Conclusion 100 THE PROBLEM OF THE TWO PROLOGUES TO CHAUCER'S LEGEND OF GOOD WOMEN. A. THE PROBLEM IN GENERAL. I. Inteoduction. The Prologue to Chaucer's Legend of Good Women has come down to us in two distinct versions. That these versions go back to distinct originals, and not to a common original greatly distorted in the course of transmission, is shown by large variations both in subject matter and form. That they are very closely related to each other is no less evident from the fact that a large number of lines, more than half of the shorter version, are exactly identical in both. Moreover, the lines which are peculiar to one version or to the other and those which are common to both, bear alike the unmistakable stamp of Chaucerian authorship. It may at once be taken for granted, therefore, — for the conclusion is irresistible, — that one of these versions is a revised or rewritten form of the other, and that the revision was made by the poet himself. We have in these two versions, so far as Chaucer texts are con- cerned, a unique possession. Chaucer has been edited and re-edi- ted and even translated into English by his critics, but in no other poem than the Legend has his own criticism of himself in any way come down to us. The importance of this possession is enhanced by the significant position which the Prologue occupies among his works. Although the date of the poem cannot be fixed with exactness, we are certain beyond conjecture that it stands near the middle of his career. In point of literary form it looks both backward and forward, for the dream, the glorification of spring, and the allegory, appear at their best and for the last time as lite- rary conventions in the Prologue, and a collection of tales bound together in a common scheme and introduced by a prologue occurs 1 2 The Problem of the Two Prologues to for the first time in the Legend of Good Women, the immediate precursor in art, and probably in time also, of the more ambitious Canterbury Tales. Considered apart from other works of Chaucer, the Prologue is still a significant poem. There is every reason to believe that when Chaucer planned the Legend, he meant to make it a monu- mental work. In that spirit the Prologue, necessarily the most vital and original part of the poem, was written. From Boccac- cio he borrowed possibly the plan of a prologue and tales, and nothing more ; from the ' Flower and Leaf romancers, to whom he gracefully alludes, no more than a few of their conventions. He passed on to his imitators more of an impulse than he received from any of them. A notable result of this is the anonymous * poem, The Flower and the Leaf, apparently inspired by Chaucer's poem and obviously an imitation of it. Clanvowe, in The Cuckoo and the Nightingale,^ owes much to the Prologue, and the anony- mous author of the Court of Love ^ betrays its influence in several lines. As an important testimony to that influence in modem times, may be instanced Tennyson's Dream of Fair Women. Since there is no external evidence to show when or why the revision was made, or even which version is the original and which the revision, the possession of the two forms, however significant the poem in itself, has thus far little meaning to us. They are valuable only for what each adds to the common store of Chau- cerian verse. The problem of the closer relation of the two Pro- logues must be settled, if at all, by an appeal to internal evidence. In regard to this, as Professor Kittredge calls it,* " very difficult question," scholars are hopelessly at variance. Since 1871, when the publication by the Chaucer Society of Ms. Cambridge Gg. 4, 27 in the six-text edition made the shorter version easily accessi- ble in print, the question of the relation of the two versions has been argued from various points of view. Conjectures have been •I do not regard as tenable Professor Skeat's attribution {Athenceum,, March 14, 1903) of this poem to Margaret Neville, sister of the Earl of Warwick. " Oxfm-d Chaucer, vn, 347 ; cf. 11. 56-80. 'Cf. W. A. Neilson, "The Origins and Sources of the Court of Love," Studies and Notes in Philology and Literature, VI, 229 ; and J. T. T. Brown, " The Author- ship of The Kingis Quair, a New Criticism." New York, Macmillan & Co. ; 1896. * Modem Philology, i, 1, n. 1, Chaucer's Legend of Good Women. 3 advanced as to the date and occasion of the revision, and from the same basis of fact different scholars have arrived at precisely oppo- site conclusions. In these investigations selections from both texts have been minutely compared ; but no one so far as I know has as yet undertaken a thorough line by line comparison of the whole of the two versions. Obviously, such a comparison must be the last resort in deter- mining the matter. If a theory based upon the poet's circum- stances, the chronology of his works, or his age does not bear the test of a critical examination of all the changes which appear in the texts, the theory is strongly discredited. If, on the other hand, a satisfactory motive for revising the poem is confirmed by the discovery of a great number of such improvements as we might expect a mature poet to make in a revision, the evidence is cumu- lative in favor of the better version as the later. It has been my purpose, accordingly, to weigh the various con- jectures why Chaucer revised the poem, and to reach a conclusion, based upon their more important variations, as to the relation of the two versions to each other. This conclusion I have endeavored to test fully by an exhaustive comparison of the two texts in detail. For the sake of convenience the longer and more widely distribu- ted form is designated throughout as F, — Ms. Fairfax 16 being the basis of the text, — and the other version, as G, after the Cam- bridge manuscript which contains it. II. The History of the Problem. The existence of the Cambridge version of the Prologue, which was discovered in 1864 in Ms. Gg. 4, 27 of the Cambridge Uni- versity Library by Mr. Henry Bradshaw, was first made generally known by Dr. Furnivall. In the summer of 1871 in a brief note in The Athermum,^ he announced Mr. Bradshaw's discovery, and a few months later published in the same periodical ^ a discussion of the two forms. " That G. is the earlier version," he says, " can hardly be doubted," and he continues to compare the two forms, 1 June 17, p. 754. ^^ October 21, p. 528. 4 The Problem of the Two Prologues to pointing out what he regards as improvements in the commonly accepted version. In Trial Forewords,^ published iu the same year, he again touched upon the variations between the two ver- sions, taking it for granted that the priority of the newly found text was unquestioned. M. Bech, in his " Quellen und Plan der ' Legende of Goode "Women ' und ihr Verhaeltnis zur ' Confessio Amantis,' " ^ adverted incidentally to the problem of the two prologues as follows : " Dass Gg. 4, 27 die friihere fassung ist, wie Furnivall im Athenmum '71, October s. 528 £F. meint, ist auch meine ansicht. Fairfax MS. 16 tragt den charakter einer endgiltigen fassung, jenes huigegen den einer vorlaufigen." He argues farther that the detailed enumera- tion of available books and stories in G (11. 267-312) has been wisely compressed into lines 556-558 of F, and that the idea of the Prologue has so developed in revision that Alcestis, who in G is praised for herself only, becomes in F a means to an end, namely, the celebration of the queen of England. In a dissertation entitled. Das Verhaeltnis der Handschriften von Chaucer's Legend of Good Women, Siegfried Kunz arrived at a similar conclusion after a brief comparison of G and F. He foimd F more consistent, fuller, and livelier than G, and the variations between the two versions such as could only be accounted for by the theory that F is a revised form. An opinion that he had ear- lier held, namely, that the poem was first written independently and that later, upon the command of the queen, it was turned to account by Chaucer as a prologue, Kunz reliaquished as untenable. Professor W. W. Skeat, in his edition of The Legend of Good Women,^ treated the question in fiirther detail but also not exhaus- tively. He concluded from a general comparison of the texts that G, which he therefore terms A, is undoubtedly the earlier, adding : " I have no doubt that a close and elaborate investigation would establish the order incontrovertibly ; but it is needless to under- take it here ; for we should at the close of it, only prove that which, for practical purposes, is already sufficiently clear." 1 P. 104 f. 2 Arujlia, v, 313-382. * Clarendon Press, 1889. Chaucer's Legend of Good Women. 5 When he came several years later to incorporate his earlier edi- tion of the Legend into the Oxford Chaucer,^ Professor Skeat found no reason to change his opinion. He reiterated his former judgment, concluding,^ "I am not aware that any one has ever doubted this result." Doubt was presently expressed with vigor, however, by ten Brink in a paper entitled, " Zur Chronologic von Chaucer's Schrif- ten," which was published in Englische Studien ^ after his untimely death. The second part of this paper was devoted to the prob- lem of the two Prologues, and to the ^date of the translation of Pope Innocent's tractate, De Contemptu Mundi. In it ten Brink, remarking that he had never seen any adequate ground for the commonly received opinion, imdertook to prove by an entirely new course of reasoning that G is a revised form. He argued (1) that since allusions to the poet's old age which are found in G are consistently wanting in F, therefore G must have been written when Chaucer was an older man and is consequently the later ver- sion ; (2) that the list of authorities named by the god of Love in G (11. 267-312) contains books with which Chaucer does not seem to have been acquainted when he wrote the first version ; (3) that the ballad, which in G (11. 203-223) has the refrain Aheste is here, existed, previous to its use in the Prologue, as an independent poem, and that it appears in the form in which it was first written in F (11. 249-269) where the refrain is My lady cometh, F being therefore the earlier version ; (4) that the plan of the Legend as originally conceived appears more clearly in F than in G. Since this plan was never carried out the poem which does not present it ftiUy may be assumed to be later. When all these arguments are admitted, it is a short step to a new date for the translation of Pope Innocent's work, which, inasmuch as it is mentioned ia G and not in F, may be supposed to have been made between the first copy and the revision. The date of the G version, from its association with the Man of Law's headlink, ten Brink regards as hardly earlier than 1393, 1 Clarendon Press, 1895. ^ Vol. m, p. xii. 'Vol. xvn (1892), pp. 1-23. 6 The Problem of the Two Prologues to and the Wreched Engendring of Mankinds, on account of its religi- ous character, he associates with the period of Chaucer's bereave- ment and financial distress, and dates 1387-1388. From the mention in the Man of Law's headlink^ of various heroines as if they had been celebrated in Chaucer's Seintes Legende of Owpyde, though their stories are not fotmd in the Legend as we have it, ten Brink concluded that the poet must have been planning, as he wrote the headlink, to complete the Legend of Good Women. Before writing any additional stories, he revised the Prologue, altering it to suit his changed relation to the court. When he had done this much, he tired of the project and gave it up, withhold- ing the new version from circulation. On the appearance of this article Dr. John Koch added to his Chronology of Chaucer's Writings, then (1892) in press for the Chaucer Society, an appendix in which he endeavored to refute ten Brink's arguments one by one. To the conclusion based on allu- sions to old age, he replied by discrediting the manuscript authority for one passage (G 31 5) and the seriousness of another (G 258-263), doubting the chronological significance of such changes in any case. The omission in F of the list of books he attributed to the inap- propriateness of some of them, and to the poet's desire to obey the injunction of the god of Love to be brief. He found in the incon- sistency of G, in that the poet pretends to be ignorant of the name of Alcestis when he has thrice heard it in the ballad, a convincing proof that F is a revision, and turned ten Brink's argument on the plan of the Legend directly about to make it prove the opposite. It seemed to him very strange that if Chaucer added the Wreched Engendi-ing of Mankinde to the list of his works because it had been written since the first version, he did not also mention the tales of Griseldis and Constance, which all admit were probably written before 1393. Upon such considerations as these and .others already advanced by Skeat, Dr. Koch saw the arguments of ten Brink crumble to pieces and was "astonished to find so much ingenuity applied to so futile an attempt." In a footnote (p. 81) to Koch's appendix, 1 Canterbury Tales, B 60 £f. Chaucer's Legend of Good Women. 7 Fumivall tersely expressed his view of ten Brink's theory by the ejaculation, " Impossible surely ! " In Germany, however, the theory of the priority of F was received with more favor. Dr. E. Koeppel, who had previously held to the commonly accepted view, promptly accepted it' and declared that he regarded the question as settled for all time. This judgment he reiterated in a review of Koch's Chronology ^ denying that any of the books named by the god of Love are inappropriate or that his warning had any bearing on the length of the Prologue itself, and asserting that the inconsistencies of G are merely evi- dences of the haste of the reworking. Continuing, Koeppel advanced the rather startling opinion that the revision of the Pro- logue was undertaken with a view to making the Legend, or part of it, serve as one of the Canierbury Tales to be told by Chaucer himself. Hence the omission in G of passages lyrical in tone and inappropriate in a narrative poem. When Chaucer had completed the revision, Koeppel suggests, he was so disappointed with the result that he abandoned the idea of using it, and laid the manu- script away among papers not intended for circulation. In 1896, Max Kaluza, reviewing^ Skeat's treatment of the problem in his iutroduction to the Legend in the Oxford Chancer, * expressed a belief in ten Brink's theory which had not been at all shaken by Koch's arguments against it. He attempted no discus- sion of the question, however, and was content with challenging the English editor to answer ten Brink's argmnents in detail. The conclusions of ten Brink as to the chronology of the Pro- logues elicited further approval in America from Frank Jewett Mather, Jr. In a foot-note to his edition of selections from the Canierbury Tales,^ published in 1899, Dr. Mather not only accepted the order of the versions as definitely fixed, but even turned ten Brink's conjecture as to the date of F into unhesitating certainty, declaring G to be "a revised version made after the ' In Englieehe Stvdien, xvu, 196. ' LUercUwrblait fiir germanische und romanische Philologie, 1893, February, col. 51. ' In Englische Studien, xxii, 271. * Vol. m, p. xxi. 5 The Prologue, The Knight's Tak, and The Nun's Priesfs Tale, Cambridge, 1899, p. xxiii, n. 8 The Problem of the Two Prologues to queen's death iu 1394, with the omission and alteration of matter personal to her." The challenge of Kaluza, which remained unanswered by Skeat, was taken up in 1900 by a French scholar, Prof Emile Legouis, in a paper entitled, "Quel fut le premier compos6 par Chaucer des deux prologues de la L6gende des Femmes Exem- plaires?"^ In this paper Legouis protested against ten Brink's "chronological preoccupation," and adopting a method already slightly used by Skeat endeavored to show (1) that an examina- tion of the plan of the Prologue discloses decided improvements in F, and (2) that the variant readings are so much better in F as to justify fiilly the conclusion that G is the earlier form. Com- bating ten Brink's arguments one by one he urged that the men- tion of Chaucer's old age is significant only when a long interval between the two Prologues is assumed, an assumption opposed to reason, and that their absence from F is better explained on the ground that Chaucer omitted as too personal the passages contain- ing them. In the probability that the ballad originally had the refrain My lady cometh, Legouis found no evidence as to the order of the Prologues, and in the greater definiteness of F as to the plan of the Legend, a positive argument for his own view. The omission of the list of books in Love's reproof, he regarded as a great gain, the reproof being in his opinion pedantic and verbose. With due caution to avoid the over-certainty which he blames in ten Brink, Legouis suggests that the obvious dedication to the queen in F may be due to a sudden change in Chaucer's relation to the court — occasioned perhaps by the grant of permission to exercise his office by deputy — which made it possible for him to dedicate the work to the queen. In a review of" Legouis' argument for Englische 8tudien,^ Koch hailed it as an independent confirmation of his own view, for Legouis did not see the appendix to Chronology until his own investigations were complete. It was now Koch's turn to declare the relation of the two Prologues settled for all time. Those who had previously agreed with ten Brink received ' Heme de V Enseignment des Langues Vivants, Avril, 1900, p. 58. ^ Vol. XXX, p. 456. Chaucer's Legend of Good Women. 9 Legouis's article with less favor, and were not inclined to change their views. Koeppel, reviewing a Chaucer monograph/ seized an opportunity to say that he still held firmly to the priority of F and to his theory that the revision was designed to serve as a Can- terbury Tale. Legouis' article was reviewed for Anglia, Beiblatt,^ by Gustav Binz, who refused to be convinced that G is an original version. He did not find it possible, however, in view of the glaring incon- sistencies of G, to regard it as a thorough-going revision. In order to save the credit of the poet, he assumed that G represents a revi- sion which was never completed, and which might therefore very naturally be inconsistent with itself. With Legouis's rather sweep- ing esthetic judgment in favor of F, Binz was disposed to disagree, regarding many of the so-called improvements as extremely doubt- ful. He held the suggestion that the ballad mighthave been twice changed, the second time back to its original form, and that the old age passages were omitted as too personal, to be equally insuf- ficient. Naturally he could not accept the conjecture of Legouis that the motive for the revision was Chaucer's desire to praise the queen, but held rather, with ten Brink, that Chaucer's coolness toward the court in his later years inspired if not the revision itself at least the changes which were made in the course of it. Evidence of a somewhat different character was next brought to bear upon the problem by J. B. Bilderbeck. This evidence, which was presented in a dissertation^ published in 1902, consisted chiefly of a detailed examination of the manuscripts. After collating the manuscripts, Mr. Bilderbeck proceeded to examine the two versions of the Prologue more narrowly, classifying the minor variations as improvements in structure or style. The results of these investi- gations convinced him that Cambridge Ms. Gg. 4, 27 presents a first draft of the separate legends up to and including that of Ariadne, and that the variations which appear in other Mss. are in large measure due to a revision made by the poet himself. The presumption thus established in favor of the priority of the G ver- 1 Englische Studien, xxx, 467. ^ Vol. xi, p. 231-237. 'Bilderbeck, Chaucer's Legend of Good Women, London (Hazell, Watson & Viney, Ld. ), 1902. 10 The Problem of the Two Prologues to sion of the Prologue, which is found only in Ms. Gg. 4, 27, Mr. Bilderbeck found to be fully confirmed by his comparison of the two versions. By means of a bolder interpretation of the alle- gorical allusions than has heretofore been attempted, he concluded further that the writing of the Legend was undertaken by the com- mand of Queen Anne in 1385, and that the story of one heroine was added to it each year. The fact that evidences of revision extend over six of the stories, together with various other conside- rations, serves to fix the date of the F version as 1390. In a recently published ' study entitled, " The Prologue to the Legend of Good Women as related to the French Marguerite Poems and the Filostrato," Mr. John L. Lowes follows ten Brink in dis- crediting the earlier view as to the priority of G, and departs even more widely from traditional opinion. He rejects altogether the assumption, first made by ten Brink ^ himself and carried still further by Bilderbeck, that F, at least, is an elaborate compliment to Queen Anne, who is allegorically represented by Alcestis and the daisy. He shows (1) that the French Marguerite poems be- long in general to a period just previous to the commonly accepted date of the Legend ; (2) that certain passages peculiar to F, usually interpreted as allegory, depead more or less directly upon these poems and the Filostrato ; (3) that F 1-196 is in many respects parallel in structure to Deschamps' Lay de Franchise, and that F from 197 to the end is similarly parallel to Froissart's Paradys d' Amours. He concludes that the apparently allegorical allusions in F are merely borrowings from French and Italian sources, and that G is a revision made when the poet's recollection of these sources had been dulled by time. The order of the two Prologues, therefore, though at least twice " settled for all time " is stUl one of the most unsettled of Chaucer problems. That it can ever be settled to the absolute conviction of every critic, without the aid of entirely new external evidence, is perhaps too much to hope for. The efibrt in this paper will be to present a conclusion based upon a more elaborate investigation of the internal evidence than has heretofore been attempted. ^ Publiealions of the Mod. Lang. Asso. of America, Vol. xrx, p. 593. * Chaucer Siudien, p. 147. Chaucer's Legend of Good Women. 11 III. Chaucer's Motive foe Revising the Prologue. Chaucer was not a man of wealth and leisure. Most of his lite- rary work seems to have been done in the intervals of a busy administrative life, and to have been inspired and carried on by an enthusiasm that overcame difficulties and led him constantly to abandon one poem unfinished in order to attempt another. It was not his habit to dwell long on any one piece of work or to revise a poem already written. His revision of the Prologue in question represents, as has been said above, apparently an exceptional and unique experience. Instances of the working over of old material are not wanting in Chaucer, but the reworking was always for the sake of making the older poetry serve a new purpose. The two Prologues, how- ever, stand in precisely the same relation to the rest of the Legend. There is no parallel here to the use of earlier poems as Canterbury Tales, or to the working in of early stanzas in the course of more mature work. For Chaucer's painstaking revision of a poem, which when revised should have the same general form and the same relation to his other poems, some special motive must have existed. What this motive was has been variously conjectured. Furni- vall,' though he noticed the lack of clear allusions to the queen in G, drew no conclusions from the fact and was satisfied with observ- ,ing that the variations show improvement in F. He believed, apparently, that Chaucer revised the Prologue merely to make it better, ten Brink,^ the first to make use of the allegorical allu- sions to the queen as a means of dating the poem, concluded that F, the only version then known to him, was designed as an ex' pression of gratitude to the queen for influence exerted by her to secure him permission to exercise his office of comptroller by deputy. The date of F was thus fixed as 1385. In his edition of the Legend published in 1889, Skeat accepted this date. He accounted for the existence of the two versions by ^Atheiumm, Oct. 21, 1871, p. 528. 'Chaucer Studien, i, 149. 12 The Problem, of the Two Prologues to supposing that when Chaucer had finished G he was not satisfied with it. The expression of gratitude which he had intended to convey did not come out with suificient clearness. "So," says Skeat, " he at once set about to amend and alter it ; the first draft, of which he had no reason to be ashamed, being at the same time preserved." A few years later ten Brink ' first expressed in print his belief in the priority of F. The occasion for the revision he assumed to be Chaucer's supposed intention about lo93 to take up the unfinished Legend and complete it. The real motive for his begin- ning this work by recasting the Prologue was his changed relation to the court. He had lost both his offices, and the royal favor which he had acknowledged in the first version was a thing of the past. Accordingly he changed the Prologue into a conventional poem, which should serve to introduce the Legend without convey- ing any play upon his earlier favorable relation to the queen. To account for the appearance of this revised version in only one manuscript, ten Brink was forced to make still another assump- tion, namely, that Chaucer gave up the project of completing the Legend before he had really begun it, and on that account withheld the later form from circulation. The most novel conjecture as to Chaucer's motive for revising the Prologue is that of Koeppel, who, having accepted ten Brink's judgment in regard to the order of the two versions, proposed a theory of his own to account for it. He assumed,^ as has been said above, that Chaucer intended at one time to use the Legend of Good Women as one of the Canterbury Talcs to be told by himself. He found the Prologue as it appears in F, on account of its lyrical tone, its personal allusions, and its obvious allegory, ill adapted to this purpose and accordingly revised it. When the work was done he was still dissatisfied with the poem, and finally gave over his intention of using it as one of the Canterbury Tales. Hence the wider distribution of F. Both Koch and Legouis reached independently a conclusion ^ Englische Studien, i, 13. 'Iiiteraturblaiif. ger. u. ram. Phil, 1893, February, C!ol. 52. Chaucer's Legend of Good Women. 13 quite the opposite of that of ten Brink, and differing in one essen- tial respecb from that of Skeat. They proposed to date G shortly before February 17, 1385, — the date of the grant of permission to Chaucer to employ a deputy, — and F shortly after that time. This theory involved the assumption, already made by ten Brink, that the queen had something to do with the grant, and that the F version was intended as an expression of the poet's gratitude. The absence of clear allusions to the queen in G was explained by the lack of an occasion for them. Although Lowes does not assign any specific motive for the revision, he regards it as having resulted in a more compact and consistent poem ; and as ten Brink's theory is not available, we may assume that he believes Chaucer to have revised F to make it a better poem.' Bilderbeck, though quite certain that F is a better version, looks elsewhere for the motive of the revision and finds it chiefly in a changed relation to King Richard. He says, "In 1390 in order to pay a high compliment to a king whose administration was winning golden opinions, and to a queen who had shown him signal marks of favour, he revised both the Prologue and the legends that he had already written." ^ But Bilderbeck had already asserted positively ^ that in G, as well as in F, Alcestis stands for Queen Anne. His view, therefore, coincides in the main with that of Skeat with which for our purpose it may be associated. There are thus open to us at least four possible views. We may believe that Chaucer revised the G Prologue to make a better poem of it, and that additions to it are merely incidental to that main purpose, or vioe versa, that he revised F for a similar reason ; that he revised G because as an expression of gratitude to the queen it was inadequate and unsuccessful ; that he revised F to get allusions to the queen out of it, either on account of his changed relations to the court, or on account of his desire to make a Canterbury Tale of the Legend ; or finally, that he revised G to 'Of. Lowes, p. 665. ^Bilderbeck, p. 109. 3 Ibid. p. 84. 14 The Problem of the Two Prologues to get allusions to the queen into a poem which did not before con- tain them. The first of these need not detain us long. Many minor vari- ations between G and F are certainly improvements in F ; but the most important changes were additions and omissions, which one does not need to study very long in order to become convinced that there was some special motive behind them. The faults of G, and they are not numerous, could have been remedied without the addition of 134 new lines. On the other hand the revision of F in the interest of better literary craftsmanship would surely not require the omission of so many good lines. The admission of Lowes, "that the B. version has the note of freshness, of spon- taneity, of composition con amore to a greater degree than A. — that it is even the more delightful version of the two ", ^ is a con- cession in favor of F not to be readily explained away by saying that that version was written currente oalamo. It is perhaps worth while to observe further that those who regard G as a revision generally find it necessary to apologize for that version as hasty work. Thus Lowes ^ agrees with Binz ' that after F 1. 390 the revision "was manifestly most perfunctorily carried out," and Koeppel* speaks of "die in manchen Kleinigheiten bemerkbare Hast der Umarbeitung." The assumption that Chaucer wrote the first version for the very purpose of expressing his gratitude to the king and the queen, and succeeded so badly that he had to rewrite the whole poem, does little credit to the mature powers of the poet. It seems strange, too, that he should, in that case, regard the imsatisfactory first draft as too good to throw away. The two versions are so different in spirit that it is very hard to believe in any hypothesis that does not involve some change of purpose. That this change of purpose was such that it resulted in an attempt to withdraw the dedication to the queen, and to remove all obvious allusions to her from the poem, is equally hard to believe. 1 Lowes, p. 683, n. 7. ''Lowes, p. 681, n. 3, p. 678, n. 4 (end). ^Anglia, Beiblait, xi, 233^. ^LUeraturblattf. ger. u. rom. Phil. 1893, February, col. 51. Chaucer's Legend of Good Women. 15 In the first place, as Koeppel has said/ it is very improbable that Chaucer ever interrupted the course of his work on the Canterbury Tales to turn back to an uncompleted poem of a different char- acter. It is still more improbable that if he did so he thought it necessary to revise the Prologue before he wrote any new legends. The intrinsic value of the poem as a whole could not be affected by the lapse of time, and the F version, — granting for the moment that it was the original, — could serve as well in 1393 as at any other time. There is, therefore, no motive left but personal pique, of which we have no grounds for accusing Chaucer. The some- what elaborate theory of ten Brink is attained by successive steps, each hypothesis being constructed to explain and support the pre- ceding one, until he has got surprisingly far from the starting point of established fact. The theory that Chaucer intended to use the Legend or part of it as a Canterbury Tale is equally unconvincing, and shows its weakness in that, like ten Brink's, it compels us to believe that Chaucer did not know his own mind. In support of his conten- tion, Koeppel cites only two passages from the text. They are equally inconclusive. I cannot believe that G U. 85-88, For myn entent is, or I fro you fare, The naked text in English to declare Of many a story, or elles of many a geste, As autors seyu ; leveth him if yow leste I is any better adapted to narration by a Canterbury pilgrim than the lines which correspond to it in F. 97-102 But wherefor that I spak, to give credence To olde stories, and doon hem reverence. And that men mosten more thing beleve Then men may seen at eye or elles preve ? That shal I seyn, whan that I see my tyme ; I may not al at ones speke in ryme. The allusion to the ladies attending Alcestis, in the words. Have hem now in thy Legend alle in minde, which appears in F (1. 557) and is wanting in G, is merely a ' LiteraturblaUf. ger. u. rem. Phil. 1893, February, col. 52. 16 The Problem of the Two Prologues to statement of what is equally implied in both versions, namely, that the story of Alcestis is to be the climax of a series of stories. We read entirely too much meaning into the word aUe, if we suppose that its omission in G implies that Chaucer meant to leave him- self free to omit a few heroines. Two lines peculiar to G allude to the legends as written : 529 Let be the chaf and wryte wel of the corn. 532 For of Alceste shulde thy wryling be. Moreover, the tone of the poem is bookish, — the allusion to the poets of the flower and leaf controversy is characteristic, — and G is even farther than F from the spirit and manner of a Canterbury prologue. It is impossible to believe that if Chaucer ever had tried to adapt F to such a purpose, he would not have succeeded far better than Dr. Koeppel would have us believe. Only one assumption remains to be considered, namely, that the G-version is the earlier, and that F represents a revision under- taken to make it an occasional poem. This theory, which Legouis advances somewhat hesitatingly, at the close of an investigation in which he has proceeded on the principle that Chaucer could only have revised the Prologue to give it greater cohesion and clear- ness, accords better with the facts than any other that has been proposed. The facts are that the two versions differ most sig- nificantly, not in allusions to the poet's age, or in the mention of works by Chaucer or by his favorite authors, not primarily in literary merit, but in the relation of the two poems to the queen. In order to make this clear we will proceed to examine the two Prologues with regard to their larger variations. A cursory reading of the two forms side by side suggests that G, which is shorter, is also simpler, and more conventional than F. In G, after an intro- duction in which he discusses his fondness for old books and for daisies, hinting that love and ancient lore are to be associated in the work which he is about to begin, the poet takes up in a more specific manner the preparation for the Legend. He describes under the pretty fiction of a dream, inspired by a May-day in the fields, the impulse that moved him to write and the plan of the work. The daisy is a type of Alcestis, and Alcestis, the model of Chaucer's Legend of Good Women. 17 purity and fidelity among women. The whole is an appropriate prologue, and nothing else. F, on the other hand, is more elaborate than G, more subjective, and in the poet's expression of his own feelings, much more fer- vent. If we compare with the introduction in G, lines 1-88, the corresponding lines in F, 1-103, we shall find that the most sig- nificant differences consist of additions in F. Examples of such additions are : F 50 That blisful sighte softneth al my sorwe, So glad am I whan that I have presence Of hit, to doon al maner reverence, 56 And I love hit, and ever y-lyke newe. And ever shal, til that myn herte dye ; Al swere I nat, of this I wol nat lye, Ther loved no wight hotter in his lyve. 82 ... the flour. Whom that I serve as I have wit or might. She is the clemesse and the verray light. That in this derke worlde me wynt and ledeth. The herte inwith my sorowful brest you dredeth. And loveth so sore, that ye ben verrayly The maistresse of my wit, and nothing I. My word, my werk, is knit so in your bonde. That, as an harpe obeyeth to the honde And maketh hit soune after his fingeringe, Bight so mowe ye out of myn herte bringe Swich vois, right as you list, to laughe or pleyne. Be ye my gyde and lady sovereyne ; As to myn erthly god, to yow I caUe, Both in this werke and in my sorwes alle. These passages indicate unmistakably that the daisy, which is in G merely a ' flower white and red,' here represents a lady whom Chaucer wishes to compliment as highly as possible, and whose patronage he invokes for his poem. Although these avowals are couched somewhat in the language of the conventional love-com- plaint, such expressions as " al maner reverence," " the herte you dredeth," "my lady sovereyne," "myn erthly god," lead us to suspect a different relation from that of a lover to his mistress, a suspicion which, as we shall see, is confirmed farther on. 18 The Problem of the Two Prologues to The objection of Lowes ^ to this allegorical interpretation, on the ground that the passages just cited echo similar phrases in cer- tain French poems and the Filostrato, I do not regard as at all conclusive. The Marguerite poems in question, as Lowes admits, are all allegorical. The invocation of the Filostrato, also, is addressed to a real person. This Chaucer must have taken for granted in both cases, if, indeed, he did not know it perfectly well. Now it is beside the point to argue, as Lowes does,^ that Alcestis is a real person in any parallel sense ; for she is no more real to Chaucer than the heroines of love who accompany her. Nothing seems to me more natural than that if Chaucer had determined to make an occasional poem of the G-Prohgue, directing its allegori- cal meaning upon a living person, such phrases, already happily used for a similar purpose, should at once come into his mind. It seems quite unnatural that, if neither Prologue is allegorical, he should have carefully omitted such phrases merely because his recollection of the originals was dulled. Chaucer borrowed freely from other poets ; but he did not borrow blindly, and his method in other cases justifies our belief that in this instance he made use of these lines, especially those from the Filostrato, because they specifically suited his purpose. It is to be remembered, too, that Chaucer had a tradition of his own, which, aside from the influence of other poems, led in the direction of allegorical compliment. No one has yet ventured to deny that the Duchesse and the Parlement are allegorical, and that the latter is a celebration of Queen Anne. The only other difference of importance up to this point is the omission in F of the announcement (G 85-88) of the author's intention to tell in English stories which he has gathered from old authors. This revelation he postpones in F with, "That ahal I seyn whan that I see my tyme." This difference concerns us now only as it indicates more clearly the single purpose of G. Verses 85-88 in G are a fitting climax ' Lowes, p. 669 £E, The relation of the passages to their sources will be consid- ered later. 2 Lowes, p. 673. Chaucer's Legend of Good Women. 19 to the introduction. The lack of coherence between 88 and 89 Chaucer could easily have remedied without rewriting the whole, and G thus amended would be so far a perfectly satisfactory pro- logue as such. As an expression of devotion to a lady of high rank, it would stand in need of important changes. If we suppose the introduction of the avowal in F 83-96 to be one of these changes, the omission of G 85-88 becomes desirable. Passiag on from the introduction we find, besides various minor changes, many of them in the direction of greater devoutness on the part of the poet, two considerable additions in F. In F 152- 170 the poet dwells much more fully on the mating of the birds, and in F 171-187 he introduces a new description of the delight- fulness of the flowery meadow. Neither of these seems necessary to add coherence or clearness to G, but both are in keeping with the more elaborate and subjective character of F. The allusion to "... Daunger for a tyme a lord," and "Pitee" that made " Mercy passen right," may with no great stretch of the imagina- tion be regarded as an allegorical reminiscence of the experience of one dependent on the bounty of a royal patron.^ The second addition is a glorification of the daisy, which echoes the avowal in F 83—96, and suggests palpably the same allegory. Note particularly lines 184—187 : " The dayesye or elles the 'ye of day,' The emperice and flour of floures alle. I pray to god that faire mot she falle, And alle that loven floures, for hir sake ! " The next considerable variation between G and F is in the ballad. In G the ballad is sung " carolwyse " by the attendants of the queen with the refrain, " Alceste is here, that al that may desteyne ; " and in F the ballad is sung by the poet himself, " ia preysiag of this lady fre," with the refrain, "My lady cometh, that al this may disteyne." ' Cf. Skeat, Oxford Ghaucer, m, p. xxiii ; Bilderbeok, p. 102. 20 The Problem of the Two Prologues to Obviously the poet is more directly responsible for the senti- ments of a ballad when it is sung in his own person, and the ballad itself is more likely to have an allegorical significance when the subject of it is unidentified. If the queen is to be identified with Alcestis in G, there is no hint in the poem that this is the author's intention. In F, " My lady cometh " is instantly asso- ciated with "... my gyde and lady sovereyne," and the allegory is clear. As if to make it clearer still, the poet adds at the close of the ballad in F five lines (271-275) in which he repeats the thought of the ballad with the conclusion : " So passeth al my lady sovereyne." In place of the ballad as sung by the ladies in G, there appears in F a song which they sing (U. 296-299) : " Hele and honour To trouthe of womanhede, and to this flour That berth our alder prys in figuringe I Hir whyte coroun berth the witnessinge ! " in which a similar thought is expressed. Another important difiference between G and F lies in the tirade of the god of Love against the poet. The allusion to the poet's age and the long discussion of his books (G 258-312), are wholly wanting in F. Now neither of these passages is out of place in a conventional prologue. It is not unnatural that the god should accuse the poet of despising love because he could no longer feel its fire, and it is quite natural that after Chaucer's appreciative allusion to old books in the introduction, we should now hear more of them. In a prologue that is intended to serve as a compliment and a dedication, however, these passages are least essential. They can be dropped out without injury to the narrative, and they leave the prologue free to center more definitely about its heroine. From this point on for nearly 200 lines the variations are very slight. Then we come in F 496 to the following couplet : " And whan this book is maad, yive hit the quene On my behalf e, at Eltham or at Shene." This is as clear a dedication as we have any right to expect, clearer than any allusion found in the Booh of the Duohesse or the Chaucer's Legend of Good Women. 21 Parliamgmi, of Birds, both of which are admittedly occasional poems. We have already seen that F differs from G chiefly by the addition of passages expressing romantic devotion to a lady under the figure of the daisy. The tone of these passages has suggested the suspicion that they are aimed at a royal personage. This couplet settles the matter. It would be absurd for the poet to dedicate to the queen a poem in which some other " lady sove- reyne" is lauded as his "erthly god," "emperice and flour of floures alle " surpassing all others in truth and beauty. Prologue F, therefore, is an occasional poem designed as an elaborate compliment to the queen of England,^ whose betrothal Chaucer has already celebrated in the Parliament. Prologue Q is not. Either Chaucer revised F in order to get the allusions to the queen out of the poem, or he revised G in order to dedicate the work to her, and to make the Prologue an allegory in her honor. That in a moment of pique over his lack of favor at court, Chaucer rewrote F in order to undo his work, is a suppo- sition that suits neither the spirit of the poet as it is elsewhere revealed to us, nor the chronology of his life and writings. Chaucer assuredly had too much good sense to attempt to withdraw a dedi- cation already made and circulated, even if we can think him mean-spirited enough to wish to do so. It is much easier to believe that after the first draft of the con- ventional prologue was finished, and had passed iuto the hands of a few of his friends, Chaucer determined for some reason to dedi- cate the work to the queen, and rewrote the Prologue accordingly. Whether his motive for doing so was gratitude for permission to exercise his office of Comptroller by deputy, it is vain for us to conjecture, for we do not know that the queen had anything to do with the matter.^ Some other specific favor or some interest manifested by the queen in the work which he had imdertaken is quite as likely to have furnished an occasion. Whatever the first motive, a reasonable and sufficient purpose in the revision is apparent, and that purpose creates a strong presumption in favor of F as the revised form. 'Cf. Bilderbeck, p. 100. ^d. Tatlock, Mod. Philol. i, 327-8. 22 The Prohhm of the Two Prologues to IV. The Ballad. So much has been made of the ballad in its relation to the two versions of the Prologue as to justify a separate treatment of that problem. The ballad itself is substantially the same in both Prologues, with the exception of a significant variation in the refrain. The variation is as follows : G Alceste is here, that al that may desteyne. F My lady cometh, that al this may disteyne. As the meaning of the ballad depends wholly upon its refrain, this variation is of considerable importance. In G the fair one who so far surpasses in beauty and fidelity all the good women of history, is the Greek Alcestis ; in F she is the unidentified ' My lady.' Equally important is a variation in the context which gives the ballad a different relation to the poem in each. In G the ballad is prefeced by the words (11. 199-202) : And after that they wenten in compas, Daunsinge aboute this flour an esy pas, And songen, as it were in carol-wyse. This balade, which that I shal yow devyse. In F, instead of these lines appear the following (11. 247—248) : And therfor may I seyn, as thinketh me, This song, in preysing of this lady fre. In these two variations alone, ten Brink ^ saw conclusive proof of the priority of F. For, he reasoned, the reproof of the god of Love in F, 537 ff. : . . . . A f ul gret negligence Was hit to thee, that ilke tyme thou made "Hyd, Absolon, thy tresses" in balade. That thou forgete hir in thy song to sette, can have no meaning unless, according to the god's interpretation of it, the refrain alludes to some one else than Alcestis. More- ^ Englisehe Stvdien, xvn, 17. Chaucer's Legend of Good Women. 23 over, since the god of Love is familiar with the ballad, although it was not addressed to him but merely communicated to the reader, he must know it in an original form independent of the Legend. From his reproof it is clear that he knows only the form which has the refrain, My lady comeih. Hence F contains the original ballad and is consequently beyond question the original prologue. All this was to Koch ^ just as certain proof of the precise oppo- site, though his answer to it is less satisfactory than that of Legouis. Legouis " accuses ten Brink of forgetting the rights of fiction, and argues that whether the ballad was written before the Prologue or not, it is here conceived as improvised on the spot, and must be so accepted both by the reader and by the god of Love, who is presumed to be, like other gods, omniscient. This view seems to me to be thoroughly just, ten Brink has in fact strangely con- fused literal fact and poetic fiction, and his conclusion, so far as it depends upon the god of Love's acquaintance with Chaucer's verse, has a very unstable foundation. The god does, indeed, elsewhere in the poem, allude to the Romance of the Hose and to Troilus and Cressida, and his queen names several more of Chaucer's works; yet we do not feel it necessary to conceive of either of them as having read a Chaucer manuscript. We must remember that Love is a god, and a dream-god at that, and not bound by the restrictions to which mortal readers of Chaucer are subject. The fact that the ballad is addressed to the reader and not to the god of Love is, therefore, unimportant. In G the poet uses a very simple convention in having a lyrical passage in a narrative poem sung by some of the actors. In F he abandons the convention and, with a word of explanation, boldly adopts the lyrical form himself. To the ordi- nary reader, and one might add, to the poet himself, no necessary incongruity is involved in the second method. One must be in a particular frame of mind before it occurs to one, on reading this poem, to ask whether the poet composed the three stanzas of the ' Chranology of Chaucer's Writings, p. 84. 'Legouis, p. 61. 24 The Problem of the Two Prologues to ballad on the spot, and whether the god of Love overheard him, or read his heart. Besides concurring with ten Brink in the conclusions just cited, Lowes * endeavors to give the ballad special significance by bring- ing it into relation with a ballad of Froissart'' in the Paradys d' Amours. The first stanza of this ballad recites the virtues of various flowers, and concludes with the refrain (1. 1635), Sus toutes flours j'aime la margherite, and the remaining stanzas are a fiirther glorification of the marguerite with the same refrain. Lowes argues that inasmuch as Chaucer had already adequately celebrated the daisy he could not follow Froissart closely, but was forced to substitute for the flowers the names of women, whom in a sense they typify. The question of the source of the ballad, apart from its possi- ble bearing on the order of the Prologues, does not now concern us, and need not be discussed at length. It should be said, however, that the only resemblance between the ballad of Chaucer and that of Froissart is a hkeness in method, and that not remark- ably close. The source already suggested by Skeat,^ if we are to look for a specific source, is much more reasonable. This ballad, printed by Raynaud,^ among works " aMributahles a Deschamps," we may reasonably suppose Chaucer to have known ; though the fact that the citation of well-known characters, who are compared with one's sweetheart to her advantage, is a common convention,' makes it impossible to say that this poem served as his model. In any case nothing is proved as to the order of the ballads in date of composition. In G we have Chaucer's first use of the type for one purpose, — the celebration of a heroine of antiquity, who must be named in the refrain if it is to be understood, — and in F a second use of it for a different purpose, the celebration of a woman then living. In the second case his purpose coincides with that of the conventional type, and consequently the type is followed more closely. 1 Lowes, p. 655 ff. * Oeuvres (ed. Scheler), i, 49. ' Oxfcfrd Chaucer, m, 298. * Deschamps, Oeuvrea CompUtes, x, p. xlix. 5 Of. Balades xivn and lx in the same volume of Deschamps. Chaucer's Legend of Good Women. 25 Whatever we may think as to the source of Chaucer's graceful praise of his " lady sovereyne " it is unnecessary to assume that these verses existed as an independent poem before they were used in the Prologue. Absalom ^ and Jonathan, who are named in the ballad, are not available for a Legend of Good "Women ; but the ballad does not pretend to be a table of contents. It occurs in the Prologue, in both versions, before the injunction to write the Legend has been mentioned, and is in no way limited in its range by the plan of the poem as a whole. Indeed, the mention of Absalom and Jonathan is here especially appropriate, for of Absalom it is said/ " In all Israel there was none to be so much praised as Absalom for his beauty " ; and David in his lamentation over Jonathan said of him,^ " Thy love to me was wonderful, passing the love of women." From the siiperlative types of beauty and fidelity among men, the poet passes on to recount examples of heroic and ill-fated virtue among women, reaching his climax in the wifely devotion of Alcestis. In all this there is no reason for believing that the ballad was not composed specifically for use in the G version, and that it was so composed is the more natural assumption. It is, however, much more than a mere assumption, for a close scrutiny of the text of the ballad should convince any one that it was first written for the glorification of Alcestis. The line common to both versions (F 253, G 207), Mak of your wyfhod no comparisoun, is clearly an allusion to the devotion of Alcestis in dying in her husband's stead. In F, regarded apart from its context as a typical love lyric, this line would be inexplicable. The ballad contains other expressions such as, " trouthe of love," " of love swich peyne" "your trouthe," which would be difficult if not meaningless in a poem addressed to an unidentified mistress. These facts seem to me to preclude the possibility that the ballad as it appears in F ever had an independent existence, and to ren- der nugatory ten Brink's argument based on that assumption. ' Cf. allusions to his beauty in balades of Deschamps, especially i, 150 ; iv, 347. ' II Samuel xiv, 25. ^ II Samuel i, 26. 26 The Problem of the Two Prologues to It is equally evident that the ballad was not originally written for the place which it occupies in F, whence it follows as a corol- lary that its relation to that version was in no way influenced by Froissart's ballad. In F the ballad is recited avowedly in order to give fitting praise to the beauty of the queen whom the god of Love leads with him. It is introduced by the lines (244-248) : .... In this world, thogh that men wolde seke, Half hir beautee shulde men nat finde In creature that formed is by kinde. And therfor may I seyn, as thinketh me, This song, in preysing of this lady fre. Yet the ballad is not devoted primarily to the praise of her beauty, though some allusion is made to it, but to her fidelity in love and her wifehood, qualities which could not be discerned at first sight. But neither the reader nor the dreamer has had any intimation that this lady is Alcestis. The ballad in F is there- fore somewhat out of harmony with its context, and bears the appearance of a passage wrested from its former connection to serve a new purpose. Especially is this the case if, as Lowes urges,' no allegorical significance is to be admitted, and the reader may not assume that the daisy-crowned queen is the patroness already invoked under the figure of the daisy, and now indicated a little more clearly as queen and wife. In G, on the other hand, the ballad is perfectly in place. We are plainly told in G 179 that, Hir name was Alceste the debonayre ; and Alcestis is named in the refrain by those who are supposed to know all her virtues. The ballad is in fact an Alcestis poem in both versions, having no connection in se with the Marguerite cult, and it is necessary to assume that it was first written in the version which frankly names Alcestis. It is quite possible, of course, that the ballad as it appears in G first existed as an independent poem. One might even argue with some show of reason that the ballad thus independently composed represents the germ of the Legend. We know from his allusion 1 Lowes, p. 670 f. Chaucer's Legend of Good Women. 27 to her in Troilus (Bk. v, 11. 1527-1533) that Chaucer knew the story of Alcestis and was interested in it. Suppose that this interest led him to celebrate her, as the preeminent example of wifely devotion, in a ballad after the manner of Deschamps. The success of the ballad suggested a larger scheme, and he thought of telling in some detail the stories of the women whom he had merely named in the shorter poem. The new undertaking took the form of a succession of stories of heroic women with Alcestis, as before, the climax. The idea of representing the whole work as a penance imposed by the god of Love on account of his having written Troilus and the Romance of the Rose, gave the Prologue. The Legend is thus the ballad expanded. After the Prologue was written, and possibly a tale or two, it occurred to him for some reason to make Alcestis a type of the queen of England, and thus to dedicate the poem to her. Accordingly the Prologue was rewritten, and the ballad appeared in the revision in a more subjective form with a refrain suited to the new purpose. I do not say that this theory is convincing, or indeed that it is anything more than possible, yet it is after all as plausible as that upon which ten Brink has based so much of his argument for the priority of F. Leaving the region of pure conjecture, we may notice that the variations between the two ballads are just such as we might expect in a change from the literal to the allegorical. If the verses are to serve as a compliment to the queen, they must not too literally praise Alcestis, and they must be in line with the preceding allegory. The indefiniteness of F and its subjective character are, therefore, ends to be gained by the revision. The evidence of the text is that both have been gained at the cost of some slight disturbance of a previously adopted plan, and the allusions to Alcestis make it impossible to assume the opposite process as an effort at improvement in structure. If my reasoning in regard to the ballad has been correct, certain inferences appear to be inevitable. These are (1) that the incongruity involved in the allusion of the god of Love to the ballad in F is more fancied than real, and is in any case not to be explained by assuming the priority of F ; (2) that the ballad in 28 The Problem of the Two Prologues to F is not an imitation of Froissart's ballad in the Pwradys d' Amours ; (3) that the ballad as it appears in G represents its original form, written presumably for use in that version; (4) that the F ballad shows distinct traces of adaptation for use in that version, where it is saved from incongruity only by its allegorical ^ interpretation. V. The Plan op the Legend and the Structure op THE Prologue. It has already been said, in the preceding chapter, that the ballad is not to be taken literally as representing the plan of the Legend. This is evident enough from the fact that in the nine legends that are extant Chaucer has told the stories of two hero- ines, Medea and Philomela, who are not named in the ballad. In a line common to both versions (G 186, F 283), he has told us of nineteen ladies who constitute an inner circle about Alcestis, and who are, generally speaking, identical with the ladies named in the ballad. In his injunction to the poet, however, the god of Love has not committed him to a definite list of names, but has given him wide latitude both as to metres and subject matter. A too literal interpretation of a similar enumeration in the Introduction to the Man of Law's Tale {Canterbury Tales, B GO- VT) has given rise to an unnecessary difficulty. Speaking of Chaucer's many stories of lovers, the Man of Law says : ' That is, the allusions to wifehood and fidelity are somewhat justified if they are seen to apply to the actual patroness of the poem already invoked and glori- fied under the figure of the daisy. Neither Richard nor his queen need have been offended by the later identification with Alcestis in F 513 as, She that for her husbonde chus to dye, And eek to goon to belle, rather than he ; for, since Alcestis is not a creation of Chaucer's own imagination, he must take her as she is, if at all. There is no necessity to press the allegory to the literal fulfilment of details. Anne is merely credited with devotion sufBcient for a similar sacrifice had need required. The royal pair might as reasonably have been indignant at being likened to tercel and formel in the Pm-Uameni of Birds. Chaucer's Legend of Good Women. 29 Who-so that wol his large volume seek Cleped the Seintes Legende of Cupyde, There may he seen the large woundes wyde Of Lucresse, and of Babilan Tisbee ; The swerd of Dido for the fals Enee ; The tree of Phillis for hir Demophon ; The pleinte of Dianire and Hermion, Of Adriane and of Isiphilee ; The bareyne yle stonding in the see ; The dreynte Leander for his Ero ; The teres of Eleyne, and eek the wo Of Brixseyde, and of thee, Ladomea ; The crueltee of thee, queen Medea, Thy litel children hanging by the hals For thy lason, that was of love so fals ! O Ypermistra, Penelopee, Alceste Your wyfhod he comendeth with the beste ! But certeinly no word ne wryteth he Of thilke wikke ensample of Cauacee. It is obvious that this list does not coincide either with the legends as we have them, or with the names mentioned in the ballad. The discrepancy in names and the fact that the Man of Law speaks of the book as a "large volume," led Diiring^ to assume that the ten legends necessary to complete the scheme as suggested in the Prologue were actually written, though now lost. His view, however, did not meet with general acceptance. Con- cerning the problem, Skeat remarks,'^ " There is no reason for sup- posing that he ever wrote complete tales about Deianira, Hermione, Hero, Helen, Briseis, Laodomia, or Penelope, any more than he did about Alcestis. But it is highly probable that just at the period of writing his Introduction to the Man of Law's Prologue he was seriously intending to take up again his 'Legend' and was planning how to continue it." A similar opinion was held by ten Brink,^ who connected with Chaucer's supposed intention to complete the Legend his revision of the Prologue, and assumed that the variations between the ballad and the list just cited were indications of a change of plan. The fact that lines 554-559 of F, which seem to mention the original plan of the Legend, are ' Chaucer's Werhe, I, 272. ^ Oxford Chaucer, in, xxvi. ' Englische Studien, xvn, 18. 30 The Problem of the Two Prologues to wanting in G, he regarded as strong confirmation of these indica- tions and a clear proof that F represents the original plan. The lines in question are as follows : Thise other ladies sittinge here arowe Ben in thy balade, if thou canat hem knowe, And in thy bokes alle thou shalt hem finde ; Have hem now in thy Legend alle in minde. To this Koch replied that it was all quite true, but that it merely served to prove that after Chaucer had deviated from his original plan, he thought it necessary to add these lines as a justification of the greater freedom which he had assumed. That two such diametrically opposed conclusions can be drawn from the same set of facts is due, apparently, to a decided difference in the interpretation of the lines ; ten Brink takes them literally assuming that thise other ladies means the nineteen heroines, and that the injunction of the poet is in effect, " Remember in your Legend all the ladies who are named in the ballad." Koch, it would seem, restricts the passage to no such literal sense, regard- ing it as an allusion to the whole company of faithful women, many of whom are properly included in the sense of the ballad, though they are not named in it. The poet is enjoined to have all of them in mind, and to glorify them all, so far as possible, in his Legend. In favor of this latter interpretation it may be remarked that the qualifying clause of the line, Ben in thy balade, if thou comsi hem knowe, has little meaning if the line is to be taken literally, and that the immediate context, I mene of hem that ben in thy knowinge For her ben twenty thousand mo sittinge Than thou knowest, certainly implies that the whole passage is meant to be taken as comprehensive and general. Moreover, ten Brink has failed to note that the real plan of the Legend is indicated in a passage which is common to both versions (G 538-540, F 549-551) : Chaucer's Legend of Good Women. 31 But now I charge thee, upon thy lyf, That in thy Legend thou make of this wyf, Whan thou hast others smale mad before ; Any other plans beyond this, Chaucer must have regarded as subject to change at any time before the stories were actually written, and it does not seem clear that so long as this original injunction was obeyed he could have felt it binding upon him either to add to or to subtract from the text of the Prologue. It is pertinent now to inquire further what the real significance of the allusion in the Introduction to the Man of Law's Prologue, is. We have seen that taken literally it presents difficulties which both Skeat and ten Brink pass over lightly. Why does Chaucer in speaking of his Legend of Good Women use a title which is not even hinted at in the work itself? Why does he call the rather slender collection of tales a large volume? Their answer to the latter question is, apparently, because he meant to proceed at once to make the work fulfil the prophetic description which conceived it as already completed. Now Chaucer had already abandoned enough unfinished poems to make him aware of the danger of such borrowing from the future, and, moreover, he was at this time engaged upon a work which drew more heavily upon his interest and his powers than any other that he had ever undertaken. That he ever seriously iatended to break in upon his work on the Tales to take up the Legend, with the promise of making it a large volume, is, on the face of it, highly improbable. Such an improbable conclusion is by no means forced upon us by the allusion of the Man of Law to Chaucer's Legend, for that allusion is admittedly inexact, both in naming the book and in describing its bulk, and we have no right to demand of it an exact enumeration of its contents. The description of the subject matter is, like the title, poetical ; and some of the names rest upon no other foundation than their mention in the ballad or their implied inclusion among the host of women who have been as " trewe as Steele." Chaucer has merely given himself free rein in writing on a fevorite subject, and not being concerned about future students of his chronology, has taken no pains to confine his poetry within the limits of exact and literal description. 32 The Problem of the Two Prologues to A more determined effort to convict Chaucer of a change of plan from F to G, in this case with reference to the structure of the Prologue itself, is made by Lowes.' He says : " Structurally regarded, the B-version^ of the Prologue falls into two clearly distinguishable parts. The first includes 11. 1-196, and, without any hint whatever of a vision, is devoted to the panegyric of the daisy and the detailed account of a day spent ia its honor in the mede The second part, including the last 383 lines of the version, is given up to the vision of the god of Love, the nineteen ladies, and Alceste." Lowes endeavors to show that the plan of the first part of F is imitated from Deschamps' Lay de Franchise, and that of the second, from Froissart's Paradys d' Amours, and argues that G adheres less closely to this derived plan, and may therefore be regarded as a revision made when the influence of the originals was no longer strongly felt. To this it should be answered first of all that the bifurcation of F at line 196 is entirely arbitrary. The two parts are indeed distinguishable, as are the actual and dream portions of the Duch- esse ; but they are not structurally distinct, nor is the plan faulty. In the first part of F according to this division, we have (1) an expression of the poet's regard for old books, (11. 1-39) ; (2) an account of his habitual devotion in the spring time to the daisy, which wins him away from books, (11. 40—105) ; (3) the story of a specific day's devotion in an actual meadow, beginning early in the morning (11. 106-107), ... In my herte I fele yit the fyr, That made me to ryse er hit wer day. To these words line 197 directly corresponds, marking the end of the same day : Whan that the sonne out of the south gan weste, Hoom to myn hous ful swiftly I me spedde To goon to reste. "When the poet has fallen asleep in his " litel herber," the expe- riences of the day bear fruit in a dream which transports him at 1 Lowes, p. 635. »i. „. p. Chaucer's Legend of Good Women. 33 once to the same flowery meadow. There he sees approaching him the god of Love aud his queen, who is instantly seen to represent the daisy. Such is the connection between the two parts of F, a connection which, in my opinion, is entirely adequate both in thought and in form. Leaving aside for the moment, however, the question of the structure of F, we may next inquire to what extent the influence of the two French poems is more apparent in F than in G. The assertion that Chaucer has imitated the plan of the Lay de Franchise is based upon the following of its characteristics : (1) Its concentration of May-day observances on the daisy. (2) Its lack of a dream setting. (3) Its use of the personified daisy as the central figure. (4) The fact that she speaks not as mistress but as a favored subject. (5) Her individualization by dress. These characteristics, however, all belong to G as well as to F. It is true that no hint of a dream occurs in lines 1—196 of F, but that feature is represented in the latter part of F where we must look also for the third, fourth, and fifth of the list cited above. In G we have, up to the appearance of the god of Love, a brief description of the actual day, 8 lines (89-96), and a rather full description of the dream day, 40 lines (104-143). In F the relation is reversed, and there is a full account of the real day, 100 lines (103-202), and a very brief account, 3 lines (209-211) of the dream, into which the god of Love comes at once. The difference between the two versions, therefore, is not so great as might seem, for it is merely a difference in the treat- ment of the same material. The use of actual experiences to furnish a motive for a highly imaginative dream is so conventional that no importance need be attached to the presence or absence of a dream setting in the Lay. The really important question is this ; which of these arrange- ments is the better and therefore most likely to represent a revision ? I cannot but feel that in F as compared with G, two advantages are gained : an adequate occasion is provided for the very signifi- cant vision of the god of Love ; and the dream is reserved for the 34 The Probkm of the Two Prologues to supernatural part of the poem, which is heightened by its contrast with the real. To justify his assertion of a close agreement between the second part of F and the Paradys d' Amours, Lowes cites ^ the following parallels: "the offender found in Love's domain and charged with trespassing ; the further charge of heresy agaLust Love's law, based on what the offender has said or sung ; the distinct recognition, on the part of the lady in royal habit, that her master owes mercy to the suppliant ; ignorance on the part of the offender that this lady is after all someone of whom he has already known ; the plea of repentance on the offender's part or on his behalf, and the specific glorification of the poet's lady, centering in a balade under the name or form of the daisy. That is to say," continues Lowes, "the framework, the cadre of the Paradys is in striking agree- ment with that of the second part of the B-version of the Pro- logue." Now all this is quite true, but it is also true that the framework of the Paradys agrees equally well with the plan of G. It is possible that Lowes considerably overestimates the direct influence of the French poem, but whatever that influence was there is no adequate ground for assigning more of it to one version than to the other. In the list quoted above the first feature that is not literally represented in G is the plea of repentance. The line (F 368), Or him repenteth utterly of this, is peculiar to F ; but it is to be noted that this is not a plea of repentance, either by the culprit or in his behalf, but one of several possible grounds of forgiveness recited by the queen. "He mighte doon hit," she says, inadvertantly, or because he was bidden, or he may be penitent. As a matter of fact such a plea is never made in either version, but the poet to the last insists that he meant no harm, and seeks to defend himself until silenced by the queen with the words, " Lat be thyn arguinge." What is apparently a more important variation in plan is to be found in the ballad, which in G is certainly less of "a specific 'Lowes, p. 652. Chaucer's Legend of Good Women. 36 glorification of the poet's lady" than in F. I have already- shown, however, that the ballad in F is an Alcestis poem and not an imitation of Froissart's ballad, nor in itself associated with the daisy cult. "What remains of the parallel is merely that there is in each a ballad recited in praise of a lady, not even iu each the mistress of the singer. This is of course no parallel at all ; and moreover the fact of the insertion of a ballad into the poem is true of both versions. When we remember that although the assertion of direct influ- ence upon the plan of the Prologue was supported by an arbitrary division of F into two parts and the assigning of one French poem to each, it was found necessary to look in the second part ot F for three of the five parallels with the Lay de Franchise, we need not be surprised to find the results of such an assertion somewhat inconclusive. All the essential parallels are, in fact, found in G as well as in F, and the whole argument based upon the plan of the poems proves nothing more than a general rather than a specific influence of the French poems upon Chaucer, without the slightest evidence of the priority of F. There still remains to fortify the argument which I have just discussed, the series of verbal parallels, first pointed out by Lowes,' with various French poems including the two in question. Lowes uses the F version only, however, and before we accept any conclusion as to the bearing of his comparison upon the question of the plan of G, it will be fair to put the G version in parallel columns with his citations from the two French poems with which we are concerned. G Pbologue. Pakadts d'Amoubs, 11. 1633-5. ' 40 Now have I thereto this condicioun Cascune flour a par li son merite ; 41 That of alle the floures in the mede, M6s je vous di, tant que pour me partie, 42 Than love I most these floures whyte Sus toutes flours j' aime la margherite. and rede, lines 1621-2 .... flours petites 43 Swiche as men caUen daysies in our Que nous appelons margerites. town, Lay de Feanchise, U. 14, 27-30. 44 To hem have I so greet afEeccioun, Le premier jour de ce mois de plaisance. 'Lowes, p. 612. 36 The Problem of the Two Prologues to 45 As I seyd erst, whan comen is the May, 46 That in my bed ther daweth me no De mon hostel me pars au point du jour day 47 That I nam up, and walking in the Prins et sousprins d'amoureuse dolour mede 48 To seen these floures. . . . M'acheminay pensant par une plaine A la beauts de la tresdouoe flour lines 44^50 51 And whan the sonne ginneth for to Et au vespre, quant il (soleil) fait son weste, retour 52 Than closeth hit, and draweth hit Ses fueUles clot, que nul ne la mal- to reste, maine 53 So sore hit is afered of the night. En demonstrant qu'eUe est vrai et cer- 54 Til on the morwe, that hit is dayes taine light. Et qu' en clart^ veult monstrer son atour; Mais en obscur tient si clos son demaine Qu'il n'estmesdis n' autre chose villaine Qui nul temps puist en lui faire demour. The right-hand column above includes all the lines cited by Lowes from the Paradys and the Lay as parallel to the F version. The relation of these lines to G may now be considered in detail. Since lines 40-48 are identical, with one important exception/ in both versions, we may pass them at once. Lines 51 and 52, however, deserve closer examination, for they are much nearer to the French than are the corresponding lines of F. Those lines (F 60-63) are as follows : And whan that hit is eve, I renne blyve, As sone as ever the sonne ginneth weste, To seen this flour, how it wol go to reste, For fere of night, so hateth she derknesse ! Here the description is not concerned about the daisy at first hand, as in G and in the Lay, but about the poet, the point of view being changed. The verbal and structural parallels involved in the statement that, when the sun goes to the west the daisy closes, are consequently wanting. This seems to me remarkably like the ' The variation between these fioures in G and this flour in F cannot be assigned to the French, because the allusion in Chaucer is purely literal, and there is, of course, no parallel in sense or structure. Its significance as an evidence of revision is considered later. Chaucer's Legend of Good Women. 37 revision, with regard only to the context, of lines originally written under the influence of a foreign original.^ Such a revision seems to me quite sufficient to accoimt for the lines which immediately follow, and which may at first sight seem to be closer to the French than G 53 and 54. These are F 64, 65 : Hir chere is pleynly sprad in the brightness Of the sanne, for ther hit wol unclose. The only verbal parallel here is brightness — elarU, for hir chere and son atour are certainly not equivalent save in the sense that they are different figures of speech for the same literal original. Chaucer can hardly be said to have followed the sense of the French, unless it be in the juxtaposition of light and darkness, a feature which is fully represented in G 53, 54. If a French source were needed for these lines of F a better might be found in lines 41, 42 of the Lay, C!ar au souleil quant il rent sa luour S'euvre la flour, tant est humble et humaine ; I believe, however, that an adequate source can be found nearer at hand in the line of G (54), Til on the morowe that hit is dayes light, and perhaps line 48, To seen these floures agein the sonne aprede. ' What may perhaps be fairly called another parallel, and one that is restricted to G, is found in line 58, As wel in winter as in somer newe. This should be compared with Paradys (P Amours, 11. 1636-8, Car en tous temps, plueve, gresille ou gelle, Soit la saisons ou fresque, ou laide, ou nette, Ceste flour est gracieuse et nouvelle. Lowes, who mentions the parallel, rejects it as not verbal ; but the fact that the word nmivdle is translated by newe in the same relation, together with the fact that Chaucer has apparently used the lines of the Paradys just preceding this, furnishes reasonable ground for thinking this direct borrowing. If it be so, we have another indication of priority in G. 38 The Problem of the Two Prologues to Chaucer has simply expanded the antithesis of G, night vs. dayes light, into For fere of night, so hateth she derknesse ! Hir chere is pleynly sprad in . . . brightnesse. It is such expansions that one expects in a revision. Had Chaucer written F as an original "with his eye on Deschamps," these lines (64 and 65) would have been the first in which he described the daisy directly, as Deschamps does throughout, and not from the point of view of the poet. Here, if anywhere, we should have expected unmistakable evidences of borrowing; yet it is here that we find them least convincing. Unless I greatly misjudge the evidence submitted above, the appeal to verbal parallels to prove that the plan of F, and not that of G, was based on the Paradys d' Amours and the Lay de Franchise not only does not strengthen the argument, but posi- tively weakens it. It does seem probable that Chaucer drew upon those poems for certain features of the Prologue ; but neither in that probability nor in any other consideration that has been brought to my notice, do I find any evidence that invalidates the priority of G. Oiaucer's Legend of Good Women. 39 B. THE TWO VEESIONS COMPAEED IN DETAIL. In order to test the conclusions reached by an examination of the larger variations between the Prologues, and at the same time to find surer ground for an opinion as to their order, a detailed comparison of variants, line by line, is next undertaken. The following statistics will serve to give a clearer idea of the nature of these variants : Lines in G 545 Lines in F 579 Lines common to G and F 280 Lines partly identical in both 165 Lines peculiar to G 100 Lines peculiar to F 134 It is to be noted that of the 100 liues peculiar to G 52, or more than half, occur in a single passage, the reproof of the god of Love ; whereas these additions to the common stock which are peculiar to F are not only more munerous but also more evenly distributed. Besides the three classes of minor variants already mentioned, namely, lines peculiar to one version, or to the other, or partly identical in both, there is a fourth class consisting of lines which, though common to both versions, have been trans- posed in the course of revision to a new context. Of the 545 lines of G 55 are reproduced, either wholly or in part, in F with a different relation to the plan of the Prologue. Taking advan- tage of the presumption already established in favor of F, I shall discuss, from the point of view of G as an original, first, the lines differently placed in the two versions ; second, the lines peculiar to G and those which displace them in F ; third, the liues pecu- liar to F which have not been otherwise noticed ; and finally, the lines which are partly identical in both. 40 The Problem of the Two Prologues to I. Lines Transposed in the Eevision. G 55-58. This dayeseye, of alle floures flour, Fulflld of vertu and of alle honour, And ever y-lyke fair and fresh of hewe, As wel in winter as in somer newe. This description of the daisy immediately follows the account of the flower's going to rest, and is concluded in G by the line, Fain wolde I preisen, if I coude aright ; In r the same lines precede the going to rest, and are used not as in G merely as a description of the flower which the poet would gladly praise, but as a ground for the reverential devotion which he pays to it. With their context they appear in F as follows : . F 51-56 So glad am I whan that I have presence Of hit (to) doon al maner reverence, As she, that is of cdle floures flour, FidflUed of al vertu and honour. And ever y-lyke fair, and fresh of hewe; And Have hit, and ever y-lyke newe, — Obviously the passage, as it is placed in F, lends itself readily to interpretation as an allegorical compliment to a lady typified by the flower, whereas in G it is disconnected from what precedes it, and provides a motive only for the poet's complaint of inability to praise the daisy fittingly. The implication of the passage in F with the context, which contains numerous lines peculiar to that version, and its service as a motive for what precedes and as a logical transition to the daisy's good night, as compared with the plain conventionality of its use in G, make it hard to believe that in this case the process of revision could be from F to G. G 71-80. A second transposed passage is G 71—80, ten lines which make up Chaucer's disavowal of partisanship in the Flower and the Leaf controversy. These ten lines appear in F condensed into nine and transposed to 188—196. In G the passage with its immediate context reads : Chaucer's Legend of Good Women. 41 G 68-80 I hope that they wil nat ben evel apayd, Sith hit is seid in foi-thering and honour Of hem that either serven leef or flour. For trusteth wel, I ne have nat undertake As of the leef, ageyn the flour to make ; Ne of the flour to make, ageyn the leef. No more than of the corn ageyn the sheef. For, as to me, is leefer noon ne lother ; I am with-holde yit with never nother. I not who serveth leef, ne who the flour ; That nis nothing the entent of my labour. For this werk is al of another tunne. Of olde story, er swich stryf was begunne. In F the couplet corresponding to G 69-70 is made to serve another purpose, as follows : F 81-83 Sin that ye see I do hit in the honour Of love, and eek in service of the flour. Whom that I serve as I have wit or might. That is, it is made to introduce the ardent invocation of the daisy which occupies U. 83—96 of F, and which is wholly wanting in G. After this invocation come the four lines beginning, But wherefor that I spak, to give credence — which are in the main common to both versions, though differently employed; next comes the description of the actual May scene, which is much fuller in F than in G ; and finally, the disavowal introduced as follows : F 186-196 I pray to god that faire mot she falle. And all that loven floures, for hir sake ! But natheles, ne wene nat that I make In preysing of the flour agayn the leef No more than of the com agayn the sheef. For as to me nis lever noon ne lother. I nam with-holden yit with never nother ; Ne I not who serveth leef, ne who the flour ; Wel brouken they hir service or labour ; For this thing is al of another tonne. Of olde story, er swich thing was be-gonne. 42 The Problem of the Two Prologues to The changes that have been made within these lines will be dis- cussed in another chapter. We are concerned here only with the evidence that may be found in their transposition. The difference between the two Prologues in that regard may be illustrated as follows : In G In F The apology to earlier writers. The apology to earlier writers. The disavowal of partisanship. The invocation of the daisy. The allusion to belief in old books. The allusion to belief in old books. The actual May scene. The actual May scene (enlarged). The poet's return and his slumber in the The disavowal of partisanship. arbor. The dream May scene. The poet's return and his slumber in the arbor. The coming of the god of Love. The dream scene and the coming of the god of Love. Other considerations being equal, it is perhaps more probable that the reviser removed a passage to a later point in order to make room for something new than that he did the converse. There are other considerations, however, which weigh in favor of the same conclusion. In G the disavowal is introduced by the assertion that the poem is written in honor of them " that either serven leef or flour." In F that first allusion to the allegorical controversy appears as part of a fervent appeal to lovers "that can make of sentement " to help whether they be with the leaf or with the flower. All other reference to the strife of flower and leaf is deferred imtil Chaucer's praise of the daisy has, in F 183-186, reached its climax, a climax which it never does reach in G, when the disavowal with its, "But natheles," comes in most appropriately. Chaucer says in effect : 'Although I do praise the daisy thus fervently, it must not be assumed on that account that I am a partisan in the Flower and Leaf controversy.' The disavowal is thus logically related to the main purpose of F, whereas in G it is merely incidental to an apology to the earlier poets after whom Chaucer comes " glening here and there." It is much easier to believe, therefore, that it was displaced from its position in G in order to make way for the invocation, which is inserted in accordance with a change of purpose, and that it was Giaueer's Legend of Good Women. 43 put after the poet's actual devout celebration of the daisy because that arrangement seemed to him a better one, than it is to believe the converse process. G 93-106. Another transposition appears in the account of the poet's return home in the evening, his slumber in the arbor, and his dream, which are deferred in F until after the description of the May scene and are found in lines 197-211. This produces an im- portant change in the plan of the poem. In each version there are two May-day experiences, one real, the other dreamed. In G the real May-day is briefly described m four lines, 89-92, which are followed by an account of the home-coming and the sleep in the arbor, in the course of which the poet dreams of a May-day in the meadow, and of the appearance of the god of Love. In F the real May-day is very fully described, — much more fully than the dream scene in G, — and is followed by a dream scene in which the god of Love at once appears with his retinue. This transfer of the description of the spring-time scene from dream to reality has been regarded as an improvement by both Skeat^ and Legouis.^ Binz, in his review of Legouis,^ though he admits the superiority of the F version in certain respects, is dis- posed to doubt it here. He believes that the wonderful vision of the god of Love is better prepared for in G than in F, where it is too abruptly introduced. This question, which has already been touched on in a preceding chapter, is too much a matter of sub- jective opinion to be settled by an appeal to facts ; but it cannot be denied that the description in F is ftdler and richer than in G, that it is, as Legouis says, the picture of a real scene, and that there is a gain in concreteness and life in making it an actual expe- rience. Moreover, in my opinion, the account of the poet's all- day-long devotion to the daisy, which is the climax of the descrip- tion, provides a fiilly adequate motive for the immediate appear- ance in his dream of the personified daisy, Alcestis, " with florouns smale, ... for al the world right as a dayesye." I have no ' Oxford ChxMcer, m, xviii. ' Legouis, p. 62. ' Anglia, BeihUitt, xi, 234 f. 44 I'he Problem of the Two Prologues to hesitation, therefore, ia regarding this transposition also as an after-thought of the poet's which has resulted in a more pictur- esque as well as a more consistent and symmetrical poem in the F version. G 338-339. The last transposition to be noticed affects two lines of G which are moved back in F. The couplet, This man to you may wrongly been accused, Ther as by right him oghte been excused, follows in G' the plea of Alcestis based on the presence of false accusers in the court of Love, which is included in lines 328—337. In F these two lines, with two new ones, precede this passage instead of following it, and take the place of four lines omitted in F. Accordingly, in F we read, F 348-352 And if ye nere a god, that knowen al, Than might hit be, as I you tellen shal ; This man io yaw mayfcdsdy been aceused Thereas by right him oghte been excused, For in your court is many a losengour, and so on. The value of the lines peculiar to the two versions will be discussed later. It is quite obvious, I think, that the arrange- ment of G, in which the couplet follows the allusion to false accusers without any grammatical connection with it, has been improved in F by transposing the couplet so that it precedes the allusion and is bound to it by a causal connective. Without pushing the argument too far, it can be fairly said that every change of lines from one position in G to another in F can be accounted for as improvements, on the hypothesis of a revision proceeding from G to F, inspired by one or both of two motives, namely, the improvement of the poem, and the celebration of the lady who is symbolized by the daisy. If we suppose the revision to have been from F to G, we must assume that Chaucer weak- ened his poem in order to accomplish some object important enough to override other considerations, an assumption for which no external evidence exists. Chaucer's Legend of Good Women. 45 II. Lines Peculiae to the G-Veesion. Of the 545 lines of the Prologue in G 100 are not reproduced either wholly or in part in F. F, however, is longer by 34 Imes than G, and contains 134 lines not found in G. Lines which stand alone in either version are significant as to the nature of the revision, for they must either have been deliberately rejected or deliberately inserted. I sh^U first discuss lines peculiar to G, taking up F only when it is evident that one passage has dis- placed another. G50. 44^50 To hem have I so greet affeccioun, As I seyde erst, whan comen is the May, That in my bed ther daweth me no day That 1 nam up, and walking in the mede To seen these floures again the sonne sprede, Whan hit upriseth by the morwe shene, *The kmge day, thus walking in the grene. ' Line 50 is the last of a sentence, and is so loosely connected with what precedes it that it can be dropped out with no violence to the structure. The line does add something to the thought but nothing that is absolutely essential to it. One might even think of it as having been written chiefly for the sake of completing the couplet. Hence it is not at all such a line as would be naturally added in a revision. On the other hand, it is precisely such a line as could be readily dropped out to make way for others. The lines which, as I believe, have taken its place read with their context as follows. F 49-52 Whan hit upryseth erly by the morwe ; *That blisfut sigkte softneth al my some, *So glad am I whan that I have presence *0f hit, to doon al maner reverence. If we suppose that F represents an earlier form we must explain why these three lines were rejected to make room for the prosaic and unnecessary line which corresponds to them in G. Clearly ' The asterisk is used in accordance with the system of Skeat to indicate lines which occur in one version only. 46 The Problem of the Two Prologues to the lines are a part of the poet's glorification of the Queen, but that is clear to us only when the whole Prologue is considered. With other more obvious allusions to the Queen removed, these lines might have been left untouched. Neither ten Brink's hypo- theses, therefore, nor Koeppel's, afibrds us a satisfactory motive for Chaucer's changing F 49-52 to G 50. The enrichment and improvement of the text is ground enough for believing that he did substitute F 49-52 for G 50. G54. 51-54 And whan the sonne ginneth for to weste, Than closeth hit, and draweth hit to reste, So sone hit is afered of the night, — *TU on the morwe, that hit is daiyes light. Line 54, like line 50 above, is the last line of a sentence and is not very essential to the thought. It can, therefore, be easily dropped out in revision. This passage of four lines appears in F without it, introduced by a new line which, in a sense, takes its •place. F 60-64 *And whan that hit is eve, I renne blyve, As sone as ever the sonne ginneth weste, To seen this flour, how it wol go to reste, For fere of night, so hateth she derknesse ! The starred line in F introduces the passage and links it to the context by telling of the interest of the poet in the daisy's going to rest. It is a most essential line and the least likely to be sac- rificed in a revision undertaken to improve the poem. As a whole the passage in F is more compact and vigorous. The rather awkward parenthesis of line G 53 is avoided. The sudden change of number from "to seen these flowres" in 1. 48 to "Than closeth hit" in 1. 52, and the consequent liability of confusion of the antecedents of hit in 1. 49, which refers to the sun, and hit used three times in 52 and 53 to refer to the daisy are other faults in G removed by the revision. G60. 55-60 This dayesye, of alle floures flour, Fulfild of vertu and of alle honour, And ever y-lyke fair and fresh of hewe, Chaucer's Legend of Good Women. 47 As wel in winter as in somer newe, Fain wolde I preisen if I coude aright ; *Bid TOO is me, Mi lyth nat in my might I We have seen above that lines 55-58 of this passage are trans- posed in F to a different context. This transposition of course made it necessary to recast lines 59 and 60, which depend gram- matically on what precedes them. Chaucer did this, I believe, as follows : F 64-67 *Hir chere is pleynly sprad in the brightnesse *0f the Sonne, for ther hit wol unclose. * Alias ! that I ne had English, ryme or prose, Suffisant this flour to preyse aright ! After the allusion to the daisy's fear of darkness (1. 63), he has inserted two new lines describing her love of light, and then the sprightly interjection in which he laments his inability to praise her worthily. This interjection seems to me much better than the colder and more formal " Fain wolde I preisen " of G, and the changes throughout the passage suggest decidedly a revision of G to produce F. Turning from the consideration of single lines we may next notice the general effect of the changes just discussed. Lowes ^ endeavors to show that G 48—60 are superior in structure to the corresponding lines of F because they follow the time order more closely, and are the result of changing a "loosely-linked cento into a compact close-knit unit." With his sweeping judgment in their favor, I cannot at all agree. Judged by his own somewhat arbitrary standard of chron- ological sequence, F is quite as consistent as G, and to my mind more coherent. In F we have two hints of the time of day, each from the point of view of the poet. In lines 46-48, the morning, . . . ther daweth me no day That I nam up and walking in the made To seen this flour agein the sonne sprede : and in lines 60-61, the evening. And whan that hit is eve, I renne blyve. As sone as ever the sonne ginneth weste To seen this flour, how it wol go to reste. ^ Lowes, p. 659 £E. 48 The Problem of the Two Prologues to Lines 64-65, Hir chere is pleynly sprad in the brightnesse Of the Sonne, for ther hit wol unclose, are not, as Lowes says, a return to the day again, — ^though he justifies such a return in G 54 as " completing the circle of the twenty-four hours," — but merely a counter statement of the daisy's love of light in contrast with her fear of darkness described in the preceding line. No more logical transition is needed, and there is, as we have seen, no necessity for making Deschamps respons- ible for the appearance of the lines at this point. The first half of the description is equally coherent. After telling us, as also in G, the fact that he is up early in the morning to see the daisy, he continues, That blisful sight softneth all my sorwe, and goes on to explain and justify this early morning devotion, which in G is merely stated as a fact. The whole passage (lines 50-59) is a consistent elaboration of this thought, and all belongs by structure as well as by logical connection with what precedes it. The general description of the daisy in G consists of three distinct sentences, all abruptly ended and without graceftil transi- tions, and. lacks moreover some ten lines of F the omission of which is not accounted for. G 85-88. 81-88 But wherfor that I spak, to yeve credence To bokes olde and doon hem reverence, Is for men shulde autoritees beleve, Thereas ther lyth non other assay by preve. *For myn intent is, or I fro yow fa/re, *The naked text in English to declare * Of many a story, or elles of many a geste, *As auUmrs seyn ; leveth hem if yow leste ! The straightforward and conventional character of these lines has been already noticed. They constitute the climax of a sort of prologue to the Prologue, and very properly state the purpose of the poem. The old book and the dream are a favorite convention of Chaucer's for introducing a poem. Here he has used them to Chaucer's Legend of Good Women, 49 introduce and justify a collection of old stories. There is nothing in these lines to indicate that the poet had more than a single purpose as he wrote them, and nothing whatever to justify us in thinkiag that they represent a second thought. The corresponding passage in F is quite different. F 97-102 But wherfor that I spak, to give credence To olde stories, and doon hem reverence, And that men mosten more thing beleve Then men may seen at eye or elles preve ? *That shall I seyn, whan that I see my tyme ; *I may not al at ones speke in ryms. These lines, particularly the last couplet, do look very much like a second thought. Certainly the poet would be much more likely to half-apologize for not doiag something that he had done in the earlier draft, than to make such an apology a part of a first version. It is easy to imagine the mental process by which G was changed into F, and not at all easy to imagine the converse. Moreover F, in this case, as often, shows the better art. Chaucer intends later to represent himself as directed by Alcestis to write as penance for his offenses against love, the stories of faithftil women, gathering his material from history. There is no reason why he should anticipate this poetical device by telling us in the beginning of the Prologue, and that in a prosaic and conventional manner, precisely what he is going to do. G91. 90-91 And I had romed al the someres day, *The grene medew, of which that I yam tolde, This single starred line needs no special comment. It will be seen presently that the whole passage of which it is a part is displaced in F by much livelier and fuller description of the scene. G 105 and 107. 105-108 *And that I romed in that same gyse, To seen that flour, as ye han herd devyse. *Fair was this medew, as thoughle me overal ; With floures swote enbrowded was it al. In F the poet does not roam either in reality or in dream ; he 50 The Probkm of the Two Prologues to kneels at daybreak beside the daisy to see it open its petals to the sunlight. Lines 107-108 are represented in F by, — smale, softe, swote gras F 118-122 That was with floures swote enbrouded al, *0/ smich swetnesse and sudch odour over-al, That, for to speke of gomme, or herbe, or tree, Comparisoun may noon y-maked be ; The line peculiar to G (1. 107) is a padded line, general in its description, and the first line of the sentence. Line 120 in F, which has the same rime-word, is closely involved in the structure, is specific in its description, and contains no superfluous words. G 138-143. *So ech of hem [^doth weT} to creature. *This song to herkne I dide al myn entente, *For-why Imette Iwiste what they mente. *I'il at the laste a larJce song above : *'I see ' quod she, ' the mighty god of love ! *Lo! yond he cometh, I see his wings spredeP This is merely a transition passage by means of which we get on rapidly to the coming of the god of Love. The introduction of the lark to announce his appearance is an awkward convention, a reminiscence apparently of the Parliament of Briddes, in which the birds allegorically represent the personages whom the poem celebrates. Here they are merely incidental, and nowhere else in the G-version are they represented as speaking. The allusion to his spreading wings is also incongruous, for it is hard to conceive him at one moment as fljong through the air and the next as walking beside his queen attended by a multitude of ladies. Certainly, as he is described in this passage, the god of Love is ill-adapted to the allegorical representation of the king of Eng- land. In F, in which the poet sees the royal pair, from a-fer com walking in the mede, the description besides being more consistent with itself is more appropriate to an allegorical interpretation. G 161-164. 158-164 Y-clothed was this mighty god of love Of silk, y-brouded ful of grene greves ; Chaucer's Legend 0/ Good Women. 51 A garlond on his heed of rose-leves *Steked al vdth Kliefloures name; *But of hisfaee I mn not seyn the heme. For aekirly his face shoon so brighte, *That with the gleem a-etoned wos the sighte; The three starred lines are wanting in F, where the rose-leves are transferred from the garland to the embroidery of Love's silken robe, and their place on his head is taken by a crown. The lines peculiar to F are, F 228-231 . . . — A fret of rede rose-leves, *The fresshest sin the world was first bigonne. *JBis gilte heer was corouned with a sarnie, *Instede of gold, for lievinesse and toighte; That the change from a fresh garland of roses and lilies to a glittering crown points to a change in the allegory has already been noted by Legouis/ approved by Koch,^ and made still more of by Bilderbeck.' It is another one of numerous indications that F is an occasional poem and G is not. The four lines in G which describe the brightness of Love's face (11. 162-5) are condensed into two in F, an undeniable improvement and an indication of revision from G to F. They read in F, F 232 Therewith me thoughte his face shoon so brighte That wel unnethes mighte I him beholde ; G 199-202. *And after that they wenten in compas, *Daurmnge aboute thisfixmr an esy pas, *And songen, as it were in carol-vn/se, *This balade, which thai I shal yow devyse. Since the ballad in F is not sung by the women, but is recited by the poet himself, these lines will not serve as an introduction to it. In F, consequently, their place is taken by the couplet, *And therfor may I seyn, as thinketh me *This song in preysing of my ladyfre 'Legouis, p. 68. ' Englische Stvdien, xxx, 457. »Bilderbeck, p. 85 f. 52 The Problem of the Tido Prologues to The significance of the variations in the use and the form of ballad have been discussed fully in another chapter. It may be noted here that though the dance is wanting in F, the song is preserved in the following lines not found in G, ^ And kneled down — /Al' F 295-299 * And scmgen with o fow, 'Hde and honour *To trouthe of wrnnan hede, and to this flour *That berth our alder prys infiguringe/ *jSir whyte coroun berth the vriinessinge.' The homage of the attendant ladies is not lost, therefore, though their ballad has been transferred to the poet because of a changed purpose in the poem. The next starred line in G is 225 *Uponthe softe and swote grene gras, a line which may have been omitted in F because the grass had already been so described in the description of the meadow (F 1. 1 1 8), Upon the snude softe swote gras, G 247. ' For thou,' quod he, ' art therto nothing able. *My servaunts been aXle wyse and honourable, Thou art my mortal fo, and me warreyest.' These lines appear in F as follows : F 320-322 ' For thou,' quod he, 'art ther-to nothing able. *IIit is my relik, digne and ddytable, And thou my fo, and al my folk werreyest.' The sense is better in F than in G, for the three lines 246, 247, and 248 are all more or less disconnected, and the transitions in sense are abrupt. There is a logical connection in thought in F between lines 320 and 321 which is wanting in G, and both the sense and the structure of the passage are improved by the anti- theses of hit and thou. The change has the further effect of increasing the prominence of the daisy. G 258-64. 255-66 Thou hast translated the Romauns of the Kose, That is an heresye ageyns my lawe, And makest wyse folk fro me withdrawe. *And Ihinkest in thy wit, that isful cool, Chaucer's Legend of Good Women. 53 *That he nis but a verray proprefool *That loveth •paramours, to horde and hote. * Wd wot I ther-by thou beginnest dote *As oldefoles, whan Mr spirit fayleih ; Than blame they folk, and wite not what hem ayleth. *Hast thou nat mad in English eek the book How that Crisseyde Troilus forsook, In shewinge how that wemen han don mis 1 The lines in this passage which are peculiar to G constitute ten Brink's strongest evidence that G is a revised and later form. He notes ^ that it is one of three allusions to old age which are wanting in F. The others are, G 315 As othere oldefoles many a day, which appears in F as, F 337 As other wrecches han doon many a day ; and the third, upon which ten Brink does not depend in his argument, G 400-401 *Whyl he was yong, he kepte your estat ; *I not wher he be now a renegat, a couplet which is wholly wanting in F. Lines 106 and 107 of F, *That in my herte I fele, yit the fyr *That made me to ryse er hit wer day — which Fumivall ^ takes to be an interesting allusion to Chaucer's age, ten Brink regards as indicating nothing more than an interval of time between the writing of the Prologue and the incident which it describes. When it is remembered that Chaucer has already ' told us that his May-day rambles are habitual, it is fair to assume that his eagerness to see the daisy open to the simshine on this occasion was no unusual thing. The " fyr " that impelled him was, therefore, the fire of youth and no isolated and remarkable impulse to which he could allude so specifically after a short interval. Although these lines are not the words of a man in his dotage, they, nevertheless, smack too strongly of the reminiscence ■ Englische Studien, xvn, 14, n. 1. ' Tried Forewords, p. 105. ' Lines 36-39 of both versions. 54 The Problem of the Two Prologues to of middle-age to be lightly brushed aside as having no reference whatever to a lapse of time. Though it be granted, however, that the F version is not wholly without allusion to Chaucer's age, it cannot be denied that two distinct allusions to the dotiag folly of men who are no longer lovers are consistently wantiug in F/ If these allusions are to be taken as serious autobiographical hints, there is some justice in ten Brink's argument that the version which contains them is, on the face of it, likely to be the earlier. There is, however, nothing that warrants us in taking them so seriously. We might as well take literally Chaucer's self- depreciation in F 414, " Al be hit that he can nat wel endyte." as additional proof, showing that he wrote F in his youth while conscious of his immaturity ; or in G 29, "But as for me though that my wit be lyte," as another allusion to dotage. For some light on Chaucer's sober earnestness when he speaks of himself as a lover, compare his disclaimer in the Parliament, 11. 889—890, " For al be that I knowe not love in dede Ne wot how that he quyteth folk her hyre,' ' with his hint at an eight-year-long love-sickness in the Duchesse, 11. 36—40. Chaucer is not to be held strictly to account for every- thing that he says about himself, for he is never half so serious as his critics. Moreover, even if we do allow an element of autobiography in these allusions, it is hard to see why, since they are both rela- tive and apply to Chaucer only as a lover, they may not as well be dated 1385 as 1393. ten Brink remarks that Chaucer does not allude to his old age in poems earlier than the Legend, and that he does so freely in later ones, namely, Scogan and the Com/pUynt of Venus. ' ten Brink, p. 14 : Wir stehen daher vor der bezeichnenden thatsache, dass Gg. eine zweimalige anspielung auf Chaucer's alter (und altersschwache) enthalt, die Vulgata keine einzige. Chaucer's Legend of Good Women. 55 Now it is generally agreed ' that the Envoy to Scogan was proba- bly written in 1393, the very year to which ten Brink assigns Prologue F. Chaucer was then probably not more than fifty-three years of age. In the Envoy he is writiag a personal poem addressed to a friend, and the situation is therefore different from that of the Legend of Good Women. Still, if Chaucer can speak of himself as a gray-beard, too old to wake his muse again, when he is a little over fi%, he might easUy be old enough at forty-five to be accused of lack of sympathy with lovers. The truth is that Scogan is jest and raillery from beginning to end, addressed to a man twenty years ^ his junior ; and little of value as to Chaucer's real feeling is to be inferred from it. The Envoy to the Compleynt of Venv^, which may have been written in 1394,' serves to illustrate further the spirit in which such allusions are made. In it Chaucer apologises for the work : " For eld, that in my spirit dulleth me, Hath of endyting al the soteltee Wei ny bereft out of my remembraunce,'' declaring that he cannot find rimes enough to follow Granson word by word, when he has actually outdone his original and made his translation an example of the highest skill in riming, The allusion is like the others purely jocular. It stUl remains to be explained why, if these allusions to age are not serious enough to be taken as sufficient evidence for dating the poem, they are to be found only in the version which I regard on other grounds as the earlier. We have already seen that F is formally dedicated to the Queen, and that it is marked by a tone of courtliness consistent with that dedication and largely wanting in G. It seems to me not at all unlikely that all three of the passages mentioned by ten Brink were omitted in the course of the revision, because of their unnecessary personal bearing and their possible tendency toward coarseness. I am aware that I may in turn be accused of pushing the sense of these passages too far when I ' Eoch, Chronology of Chaucer's Writings, p. 70, and note. 'Cf. Kittredge in Harvard Studies in Philol. and Lit., l, 116 ff. 'Koch, Chronology of Chaucer's Writings, p. 72, n. 56 The Problem of the Two Prologues to speak of their showing a tendency toward coarseness ; but when I recall the sixteenth and seventeenth statutes of the Court of Love,^ and remember the Ovidian origin of so much of the romantic tradition, I do not wonder that even in the fourteenth century Chaucer might have felt it to be in better taste to omit them from a poem dedicated to the queen whose marriage he had already celebrated in the Parliament. G 267-312. The long reproof of the god of Love beginning 267 But natheles, answere me now to this, and endiag, 311-312 But yit I sey, what eyleth thee to wryte The draf of stories, and forgo the com ? is wholly wanting in F. The substance of the reproof is that Chaucer has neglected the stories of faithful women, of which he had an abundance ia his books, and has preferred to write of the perfidy of Cressida. As this is the most considerable passage peculiar to G, it has naturally provoked numerous conjectures. Skeat ^ at first found it hard to understand why Chaucer should here appeal to works against women ; but afterward ^ modified his opiuion. ten Brink based upon the mention of such books his most cogent argument for the later date of G. " These books," he said,^ " are such as, so far as one can see, Chaucer knew only slightly or not at all during his second period, though they play a very significant r6le in his third period." Legouis,^ insisting upon Chaucer's right to be judged first of all as a literary artist, argues that this tirade of the god of Love, including 11. 258-264, is out of keeping with the character of the god, whom it transforms into a verbose ' Skeat' s Chaucerian and Other Pieces, p. 421 f. Cf. lines 435-469. This poem is, of course, non-Chaucerian and later than the Legend of Good Women, but it imdoubtedly goes back to an older tradition quite familiar to Chaucer. 'Legend of Oood Women (1889), p. xi. ' Oxford Ohaucer, m, p. 302, n. * Englische Studien, xvn, 15. ^Legouis, p. 61. Chaucer's Legend of Good Women. 57 pedant. Hence, Legouis would say these passages were omitted in the revision. Whatever may be said of the literary merit of the lines, it must be admitted, as Koeppel has pointed out,^ that the books mentioned by the god of Love are not inappropriate here, for even Jerome concedes some goodness to women, ten Brink's chronological argument based upon them is at first sight convinc- ing, but demands closer scrutiny. The lines in which authors of stories about good women are named are as follows : 280-281 *What seith Valerie, Titus, or Claudian? *What seith lerome ageyns lovinian ? The next twenty-four lines are a description of two chapters of Jerome against Jovinian. Then follows, 305-307 *What seith also the epistels of Ovyde *0f trewe wyves, and of Mr labour ? *'What Vincent, in his Storial Mirour ? Six authors are here mentioned by name. Let us examine each in turn and see if any significance as to the chronology of the two Prologues can be found in them. Valerie, I take to be Valerius Maximus,'' whom Chaucer twice ^ mentions by name as one of his authorities for classical stories. Professor Bright has shown * that the otherwise inexplicable Eleanor of the House of Fame (1. 516) refers to Hamilcar, the account of whose dream Chaucer got from Valerius Maximus. 'Anglia, xni, 174£f. ' Skeat is certainly wrong in supposing that the reference is to Walter Mapes' Valerius ad B/afinum. See Lounsbury, Chaucer Studies, ii, 276, and Koppel, Anglia, xm, p. 182. As Koppel intimates, the epistle of Valerius is quite inap- propriate as a source of stories about true women, and Valerius Maximus is not. In his Factarum et dietorum Memorabiiwm, Lib. iv, cap. vi, 1, the story of Alcestis is briefly told along with those of other women faithful in love. In Lib. n, cap. iv, 14, Valerius extravagantly praises the devotion of the Hindoo widows, who strive for the honor of cremation upon the pyre of the dead husband, and in other passages, n, i, 3 ; n, ii, 8 ; and m, ii, 9, narrates incidents showing the virtue and the courage of women. »In the Mmiii's Tale, B 3910, and in the Wife of Bath's Tale, D, 1165. * Modem Language Notes, IX, 241. 58 The Problem of the Two Prologues to We may, therefore, be certain that Chaucer knew Valerius before he wrote either version of the Prologue, and the mention of Valerie has no bearing whatever on the problem of their relation to each other. Livy, whether Chaucer actually used his works or not, is men- tioned ia the Booh of the Duchesse (1. 1084) and Claudian, not to multiply evidence, in the House of Fame (1. 449). We come next to Jerome ageyns Jovinian, the work upon which particularly ten Brink bases his argument. This is evidently the book which Chaucer has especially in mind in the tirade of the god of Love, for fully half of that passage is a discussion of the contents of two chapters of Jerome's work. It is quite true that the same book is very extensively used in the Wife of Bath's Prologue (D 674 ff.) and in the Franklin's Tale (F 1367-1456) ; but it does not neces- sarily follow that the three allusions to the work are therefore to be assigned to the same date, or even to the same period ia Chaucer's life. Positive evidence that Chaucer knew Jerome's treatise before he wrote either version is afforded by his mention of Marcia Catoun in both forms of the ballad (G 206, F 252). The reference here is, as Lounsbury points out,^ to the daughter of Cato, the account of whose refusal to marry a second time is given iQ Jerome.^ The fact that in the passage from which this account of Marcia Catoun is taken (the end of chapter 45 and the beginning of 46) Alcestis, Penelope, Laodamia, and Lucretia are also mentioned as heroic wives, may even imply that Chaucer had Jerome before him when he wrote the ballad, and derived some of his inspiration from him. The argument that Chaucer did not know a work at the time of writing a certain poem, based solely on the fact that he appar- ently made no use of it, is in itself of doubtful force. In this case even that argument cannot apply, for we have evidence that Chaucer knew every one of the authorities named in the reproof before he wrote either version of the Prologue. G 324-327 : *He shal nat rightfully his yre wreke *0r he have herd the tother party speke. ' Lounsbury, Stvdiea in dumem; n, 249. ^Adverms Jomnianwm, i, 46. Chaucer' 8 Legend of Good Women. 59 *A1 ne is nat gospel that is to yow pleyned ; *The god of love berth many a tale y-feyned. These four lines, which form part of the appeal of Alcestis in behalf of the accused poet, are wanting in F. In their place are found two new lines, F 348 and 349 : *And if ye nere a god that knowen al, *Than mighte hit be, as I yow tellen shal ; and a couplet carried back, with slight change, from G 338-339 : This man to yow may falsly been accused, Ther as by right him oghte been excused. In G, as I conceive it to have been written, the appeal of Alcestis including the four starred lines had reference only to the idea of a court of love in which Cupid himself is judge. There is no good reason for interpreting it to mean also the court of King Richard. F, however, is distinctly dedicated to the Queen, who is represented in it by Alcestis. In F, therefore, whatever Alcestis says to her lord is, to say the least, lq danger of being applied to the king of England, and Chaucer may well have felt like toning down somewhat his admonitions to royalty. Yet there was no necessity for cutting out the whole passage. It could do no harm to intimate to the king that he was liable to hear envious slanders, or that a righteous lord should not be like the tyrants of Lombardy ; but to tell him that he ought not to wreak vengeance before hearing both sides, or refiise to listen to the complaints and petitions of his people would be too specific to be wise. G 344-345 : *Ther^or he wroot the Base and eek Orisseyde * Of innocence, and niste what he seyde. Or him was boden maken thilke tweye Of som persone, and durste hit nat with-seye. This couplet is wanting in F. Without it, however, thilke tweye of line 366 must refer to the mention of the two poems 33 lines back, which seems unlikely. If the presence of the lines in G is to be used as an argument that they were inserted in the course of a revision, we must suppose that Chaucer did not write them in 60 The Problem of the Two Prologue to his original. It seems to me much more likely that the couplet has been lost from F by the carelessness of a scribe, or inadvert- antly omitted by Chaucer in making his revision. G348. *For he hath writen many a book er this, In place of this line we have in F the line, *0r him repenteth utterly of this. Both lines are so disconnected from the context that either might have been readily substituted for the other. It is true, however, that the line in F adds a new thought, a new possible ground of excuse for the poet, and that the line m G does not, and is moreover rather meaningless and unessential as it stands. It is fair to assume that the better line is the later one. G 360-364. G 368-369. *And that him oweth, of verray duetee, *Shewen his peple pleyn benignitee, *And wel to here hir excusaciouns, *And hir compleyntes and peticiouns, *In duewe tyme, whan they shal hit profre. *And therto is a king ful depe y-sworn *Ful many a hundred winter heer-biforn. These lines are to be associated with 11. 324-327 discussed above. With them they make up eleven lines which I believe were omitted because they seemed inappropriate in an allegorical poem. If I understand ten Brink's view perfectly, he would have us believe that Chaucer revised the Prologue to make it less of an allegory, and at the same time on account of his changed relations to the court inserted these further criticisms of royal justice, which, if they are to express his personal pique, demand an allegorical interpretation. So practised a diplomatist as Chaucer could hardly have failed to recognize that criticism of royalty is at all times unfruitfal if not unwise.^ ' Bilderbeck (p. 96 £E. ) argues that these allusions were removed because they had done their work and were no longer applicable to King Bichard's adminis- tration. It seems doubtful, however, that Chaucer would have ventured to be so plain-spoken in the first instance. CJMuoer's Legend of Good Women. 61 G 400-401. *Whyl he was yong, he kepte your estat ; *I not wher he be now a renegat. In a part of the Prologue in which there are few variations from G, F wholly lacks this couplet. It is either a slight expression of pique inserted by the disgruntled courtier in G, or it is a passage natural enough in G as a conventional prologue, but so personal as to be in bad taste in a poem in which the god of Love becomes a type of the King of England. It is m any case such a personal reference as the poet might have discarded on second thought. That he inserted it because he was seven or eight years older than when he wrote the first draft seems very unlikely. G 414-415. *And of the Wreohed Engendiing of Mankinde, *As man may in pope Innocent y-finde ; The presence of this couplet in G and not in F has given rise to various conjectures. Koppel, when he regarded G as the earlier version, supposed* that Chaucer began the Wrecked Engend- ring before he wrote G and expecting to finish it mentioned it there, and that he later tired of the work and gave it up. Hence in the process of revision he dropped it out of F, and some years later made use of fragments of the incomplete translation in the Man of Law's Tale. At the same time he carried over the chaff figure in G, lines 311 and 529 to the 3Ian of Lav^s Tale, 1. 701. ten Brink, having satisfied himself that G is to be regarded as a revision, of course rejected this explanation, and assumed that the mention of a work of Chaucer's in what he regarded as the later version was strong confirmation of his view. Now Chaucer lost his offices in December, 1386.^ From that time until July 12, 1389, he was not, so far as we know, engaged on any public service. This fact and the significant one that in May, 1388, his pension was assigned to another justifies the belief that he was, during part of this period at least, in financial straits. Hence, ' Herrig's Arehiv, Lxxxiv, 406 ; lxxxv, 48. ' Life Becords of Chaucer, p. xxsv and pp. 268, 269. 62 The Problem of the Two Frologues to argues ten Brink, Chaucer's interest in the treatise of Pope Innocent, which he translated at this time, and hence his use in the Man of Law's Tale, written soon after, of the inappropriate stanzas on poverty. All this hangs together excellently, but is it therefore trust- worthy? If F was written in 1385 and G in 1393, Chaucer must certainly have done in the interim some work better worth mentioning and more appropriate to the context than the transla- tion of Pope Innocent. Without attempting to fix the date of this translation, I may say that I am not at all convinced that it must have been made while Chaucer was himself suffering the pangs of poverty. He was not, so far as we know, in prison when he translated Boethius. Chaucer's fondness for Boethius and the presence of the Melibeus and the Parson's Tale among the Canter- bury Tales may show us that his own taste and the taste of his age made it possible for him to find the De Miseria Conditionis Humanae interesting at any time. That his translation of it must have been made about 1387-1388 is therefore merely a conjecture, and a weak link in ten Brink's chain of argument. Another such link is the assumption that the stanzas on poverty translated from Innocent and prefixed to the Man of Law's Tale were dragged in because of the poet's personal feelings. However we may feel about it, Chaucer evidently regarded these stanzas as a suitable introduction to the tale, which, it must be remembered, is told in character. They are quite consistent in style with the tale itself, and a remote and formal approach to the subject is not inconsistent with the character of the Man of Law. It is improbable that Chaucer ever suffered such poverty as he apostrophises here, and is perhaps unlikely that, if he had, he would speak of it as he does in these stanzas. Last of all it may be remarked that ten Brink's date for the Man of Law's Tale, 1391, is also purely conjectural. Koch's guess' puts it about 1386, and Skeat^ thinks that it was composed in 1380 and revised in 1387. My conclusion is that the couplet mentioning the Wreohed Engendring of Mankinde has no significance as to the relative date of the two Prologues, '■ Chronology of Ghaucer's Writings, p. 79. ^ Oxford Chaucer, m, 413. Chaucer's Legend of Good Women. 63 and that its absence in F is due either to the error of a scribe or to Chaucer's feeling that it added nothing to the poem and might well be dropped out. G 527-532. 525-532 Than seyde Love, a ful gret negligence Was hit to thee, to write unstedfastnease *Of women, sit thou knowest hir goodnesse *By preef, and eek by stories heer-biforn ; *Let be the chat, and wryt wel of the com, *"Why noldest thou han writen of Alceste, *And leten Criseide been a-slepe and reste ? *For of Alceste shulde thy wryting be, . . . This passage repeats substantially the reproof of Love in 11. 266 ff., narrowing it down to Alcestis. It is therefore quite con- sistent with the purpose of the poem if that purpose be the glorification of wifely fidelity with Alcestis as its type. Any change of purpose which would make it possible to omit the first of these two passages, i. e., 11. 266 ff., would also make it easy to change the second. The lines which take the place of these in F, reflect a previous change in the ballad. They are : F 539-543. 537-543 Than seyde Love, ' a ful gret negligence ' Was hit to thee, that ilke tyme thou made *' 'Hyd, Absolon, thy tresses," in balade, *That ihouforgete hir in thy song to selte *Sin that thou art so gredy in hir dette, And wost so wel, that kalender is she *To any woman that wol lover be. In F the poet sings his baUad in praise of a lovely lady whose identity he does not yet know. Among the beautiful and faithful women mentioned in it the name of Alcestis does not appear, since in the original form it occupied the place of honor in the refrain. This justifies the reproof; for whether Love interprets the refrain, "My lady cometh," as applying originally to Alcestis or not, the poet has failed to " set her in his song." Very naturally there follows immediately the injunction to make good his neglect by giving her the place of climax in the Legend. 64 The Problem of the Two Prologues to III. Lines Peculiar to the F-Veesion. Some of the lines whicli are peculiar to the F-version have been considered already in their relation to those lines of G to which, in a sense, they correspond, and which, according to my view, they have displaced in the process of revision. Many others, however, cannot be regarded as having taken the place of lines omitted from G. If F is a revised version, they are absolute additions designed to enrich the poem or to contribute to the better fulfilment of its purpose. If F is an original version and G a revision of it, these lines are deliberate omissions dropped out of the poem for some specific reason which it should be possible at least to conjecture. That such a reason is to be found in their intrinsic value as poetic lines, no one has ventured to maintain, for it is these lines that justify the assertion that F is the less conventional, the more elaborate, and the more subjective poem. No more vigorous or efiective passages occur in either Prologue than some of those that appear to be additions in F. Three reasons have been suggested as possible grounds for the omission of such lines. These are, first, that they allude too clearly to the Queen to suit Chaucer's mood at the time of revision,^ second, that they are too lyrical in tone to be suitable in a Canterbury Tale,^ and, third, that they were omitted in the course of revision in order to give G a more compact and coherent structure.* The latter view, that of Lowes, assumes that certain parts of F lack imity and coherence because they have been composed " with the eye on a foreign original." Some attention was paid to this assumption in the preceding chapter, and the relation of the Prologue to two of the French poems has been discussed in connec- tion with the plan of the poem. The other French originals may be considered now. 1 ten Brink, Englisclie Siudien, xvn, 19. ^ Koeppel, Literaturblatf. ram. v,. ger. PhUot, 1893 (February), col. 52. ' Lowes, p. 661. Cha'noer's Legend of Good Women. 65 Besides the Lay de Franchise and the Paradys d' Amours, which have been discussed, Lowes cites parallels from four different poems. These are Froissart's, BiUi6 de la flour de la Margherite, Machault's, Bit de la Marguerite, Froissart's, Lejoli mois deMay, and one of Desehamps' balades. The last two mentioned may be dismissed from consideration by showing that they are parallel to G as well as to F. G-Prologue, 11. 53-55. Lejoli mois de May, 11. 289-290. This dayesye of alle floures flour, Car elle est la flour souverainne Fulfild of vertu and of alle honour, De bout^ et de beauts plainne. Balade No. 532, 1. 16. And ever y-lyke fair, and fresh of hewe. Qui en tons temps belle et fresche sera. In the case of another the parallel is so far from convincing that it is likewise to be dismissed from consideration. Prologue F, 11. 50-52. DU de la Margherite. That bUsful sighte softneth al my sorwe, Car elle m'a gari d'oultre la mer. De ma doulour. So glad am I whan that I have presence Si la doy bien servir et honnourer Of hit, to doon al maner reverence. Et mettre en li cuer, et corps, et penser. In the absence of any confirmatory evidence that Chaucer made use of the Dit, this citation, parallel to Chaucer in only what is commonplace, has no meaning whatever. The lover's doulour has been healed by the sight of his sweetheart from time imme- morial. Turning now to the DittiS, we find that the first parallel, which is undeniably close, is represented in both versions : Prologue O, 11. 48-49. Ditiie de la Marguerite, 11. 162-166. To seen these floures agein the sonne . . . Car n'ai aultre desir sprede, Que de 1' avoir pour veoir k loisir Whan hit up-riseth by the morwe Au vespre clore et au matin ouvrir, shene. Et la soleil de tout le jour sievir Et ses florons contre lui espanir. What we have left consists of two brief citations from the Dittii, which are parallel to four lines peculiar to F. They are as follows : 66 The Problem of the Two Prologues to F Prologue, 11. 56-59. Dittie de la Marguerite, 11. 81-82. And I love hit, and ever y-lyke newe, Comme celle est que j'aim d'entente And ever shal, til that myn herte dye ; pure, Et aimerai tous jours, quoi que j'endure. 11. 159-162. Al swere I nat, of this I wol nat lye, S'en ce parti vivoie, nul millour Ther loved no wight hotter in his lyve. Ne doit querir Horns, ce m'est vis, qui tant aime et desir La flour que fai. Now if we had before us forty lines peculiar to F which could be traced to French sources, instead of four, or admitting 50-52, seven at most, there would still be lacking any proof that the version containing them was the earlier. I have followed out these parallels in such detail simply to make it clear that the "loosely-linked cento" of quotations in F reduces itself on exam- ination to a somewhat insignificant matter after all. That either in the structure of F, or in the relation of the Pro- logue to Froissart's DittU, any satisfactory explanation of the omissions from F has been offered, I cannot believe. If there is any passage in the Legend that deserves the epithet loosely-linked it is G 48—60, with its abrupt transitions, its floating, parenthetic lines (50 and 53), and its sudden and unexplained shifts in point of view at lines 51 and 55. F, on the other hand, is logically coherent throughout and maintains a single point of view, that of the poet. Moreover, each of the passages is obviously an expan- sion of the idea preceding it. "I go early to see the daisy open," says the poet, "and (1. 50) that sight softens my sorrow, so glad am I to do it reverence." "I love it," he says again, "and (1. 57) ever shal, til my heart die." The fact that the second passage is related to the Ditti6, proves, as I have intimated, nothing at all as to the order of the versions. We have no reason for believing that when Chaucer once used a poem as a source, he laid it aside forever. The very work of revising the Prologue would inevitably recall to his mind the works that suggested his first essay at a Marguerite poem, and would probably send him to them for improvements. If, as I believe, he revised with a changed pur- pose, the Ditti^j whioh (jelebrett^s a living Margaret, would naturally occur to him. Chaucer's Legend of Good Women. 67 We may next consider the lines in the light of the conjectures of Koeppel and ten Brmk, quoting them again for the sake of clearness. F 50-57. *That blisful sighte softneth al my sorwe, *So glad am I whan that I have presence *0f hit, to doon al maner reverence, And I love hit. . . . *Aiid ever shal til that myn herte dye ; *A1 swer I nat, of this I wol nat lye, *Ther loved no wight hotter in his lyve. This fervent avowal of love for the daisy, though the last couplet of it suggests an allegory, contains in itself no such clear refer- ence to the Queen as would require its omission were such references deliberately omitted, nor is it more lyrical than many parts of the Canterbury Tales. In themselves the lines suggest no reason why they should be sacrificed. That they add life and genuineness to the rather tame, so greet affecdoun, of G is an obvious reason why they should have been added. When these lines are considered along with other clearer allegorical hints, we see that they fore- shadow the celebration of the Queen under the type of the daisy, and we may readily conclude that they were added chiefly because the poet had a new purpose in mind as he wrote them. F 68-72. *But helpeth, ye that han conning and might, *Ye lovers, that can make of sentement ; *In this cas oghte ye he diligent *To forthren me somwhat in my labour, *Whether ye ben with the leef or with the flour. In G the allusion to the love-poets who have preceded Chaucer, speaks of them in the third person, and contains no such invo- cation as this.' In F the whole passage is enlivened by being ■ The absence of these lines in G is accounted for by Lowes on the ground that by the time of the revision the Marguente poems had become an old story to Chaucer. The reason is entirely inadequate for the eflect. The appeal and the direct address are better poetry than the comparatively lifeless passage in G, and Chaucer had no motive that we can think of for such a change of feeling toward 68 . The Problem of the Two Prologues to thrown into the second person and enriched by the addition of these lines. The appeal to "lovers that can make of sentement " is based upon the special character of the poet's aim, an aim^ which, it is to be noticed carefully, is quite different in the two Prologues. In G, . . . hit is seid in forthering and honour Of hem that either serven leef or flour. This rather feeble and general dedication of the poet's work gives place in F to, . . . . ye see I do hit in the honour Of love, and eek in service of the flour, Whom that I serve .... Eemembering that when such a phrase as " service of the flour " occurs in the French Marguerite poems, it signifies devotion to a mistress of flesh and blood, and never to a mythological character like Alcestis, we may observe that Chaucer appeals to lover-poets on the ground of a double purpose — the glorification of love, and of a living person symbolized by the daisy. Who this person is the next passage gives us an intimation, which becomes perfectly clear only when the poem is dedicated to the Queen. F 83-96. 81-96 Sin that ye see I do hit in the honour Of love, and eek in service of the flour, *Whom that I serve as I have wit or might. *She is the clernesse and the verray light, *That in this derke worlde me wynt and ledeth, *The herte in-with my sorowful brest you dredeth, *Aiid loveth so sore, that ye ben verrayly *The maistresse of my wit, and nothing I. *My word, my werk, is knit so in your bonde, *That, as an harpe obeyeth to the honde the Flower and Leaf poets as would induce him to give up the better for the worse. Mere lapse of time is no ground for the omission of F 68-72 or for the change of second person to third. If we suppose that after the somewhat tame allusion in G, Chaucer's interest in the French poems was quickened by a re- reading of them, or by his correspondence with one of the poets, we can readily see how the revised poem would show a livelier feeling and would contain the direct appeal wanting in the first draft. Otherwise we must suppose Chaucer to have revised capriciously and unsuccessfully. Chaucer's Legend of Good Women. 69 *And maketh hit Sonne after his fingeringe, *Eight so mowe ye out of myn herte bringe *Swich vois, right as yow list to laugh or pleyne, *Be ye my gyde and lady sovereyne ; *As to myn erthly god, to yow I calle, *Both in this werke and in my sorwes alle. There can be no doubt that this allusion to the daisy, which drifts so easily into the second person and becomes an earnest invocation of a "gyde and lady sovereyne," is allegorical, and that, in view of lines 496-497, the daisy can only symbolize the Queen of England. No such allegorical interpretation of G is possible. If the daisy in the corresponding part of that version stands for any person, it can only symbolize the virtuous Alcestis. If this passage in F has been omitted in the course of a revision, therefore, it is fair to assume that the allusion to the Queen is responsible for its rejection. Nothing but the .strongest kind of external evidence, however, would justify us in believing that Chaucer sacrificed these fine lines merely because he was out of suits with fortune. The conclusion that accords best with the other evidence, as well as with common sense, is that F here shows the efiects of revision with a slightly changed purpose, that is, the invocation of the Queen as patroness of the poem.^ F 103-118 : *My busy gost, that thrusteth alwey newe *To seen this flour so yong, so fresh of hewe, *Constreyned me with so gledy desyr, *That in my herte I fele yit the fyr, *That made me to ryse er hit wer day — And this was now the firste morwe of May — *With dredful herte and glad devocioun, *For to ben at the resurreeioun *Qf this flour, whan that it shulde unclose *Agayn the sonne, that roos as rede as rose, *That in the brest was of the beste that day, *That Agenores doghter ladde away. *And doun on knees anon-right I me sette, *And, as I coude, this fresshe flour I grette ; *Kneling alwey, til hit unclosed was, *Upon the smale softe swote gras. ^ The relation of these lines to the FUoslrato, pointed out by Lowes, seems to me rather to confirm this hypothesis than to weaken it, for both passages have the same purpose. 70 The Problem of the TSioo Prologues to These lines are consistent with the allegory in that they repre- sent a much greater devotion to the daisy than is displayed in G, yet they are necessarily so much more Kteral that a reviser bent on removing all references to the Queen would certainly have been tempted to spare them for their poetic value. Much more certainly would Chaucer, in revising G, have regarded such lines as these as an undoubted improvement. F 143-144 : *tJpon the braunches ful of blosmes softe, *In hir delyt, they turned hem ful ofte. In the part of G that corresponds to this passage in F, the text is so corrupt that the couplet wanting in G may possibly have fallen out. It is quite as possible, however, that these two lines, giving as they do a thoroughly good descriptive touch, were added as an improvement. F 152-187 : *Construeth that as yow list, I do no cure. *And tho that hadde doon unkindenesse — *As doth the tydif, for new-fangelnesse — 155 *Besoghte mercy of hir trespassinge, *And humblely songen hir repentinge, *And sworen on the blosmes to be trewe, *So that hir makes wolde upon hem rewe, *And at the laste maden hir acord. 160 *A1 founde they Daunger for a tyme a lord, *Yet Pitee, through his stronge gentil might, *Forgaf, and made Mercy passen Eight *Through innocence and ruled curtesye *But I ne clepe nat innocence folye, 165 *Ne fals pitee, for ' vertu is the mene.' *As Etik saith, in swich maner I mene. *And thus thise foules, voide of al malyce, *Acordeden to love, and laften vyce *0f hate, and songen alle of oon acord, 170 * Welcome, somer, our governour and lord ! ' *And Zephirus and Flora gentilly *Yaf to the floures, softe and tenderly, *Hir swote breth, and made hem for to sprede *As god and goddesse of the floury mede ; 175 *In which methoughte I mighte, day by day, *Dwellen alwey, the loly month of May, *Withouten sleep, withouten mete or drinke. Chaucer's Legend of Good Women. 71 *A-doun ful softely I gan to sinke ; *And, leninge on myn elbow and my syde, 180 The longe day I shoop me for to abyde *For nothing elles, and I shal nat lye, But for to loke upon the dayeseye, *That wel by reson men hit calle may *The ' dayeseye ' or elles the ' ye of day,' 185 *The emperice and flour of floures alle. *I pray to god that faire mot she falle, *And alle that loven floures, for hir sake ! This thoroughly Chaucerian passage gives us a much fiiUer description of the birds' May-festival, which is quite in the spirit of the poets whose help he has invoked, continues the note of romantic devotion to the daisy, and in the concluding verses once more makes clear its allegorical significance. The more elaborate description of the May-day scene is consistent with the other changes by which the date is definitely fixed as the first day of May, and the whole scene is transferred from dream to reality. It ought not to be necessary to insist that all these changes are improvements, and that no passage omitted from G contributed so much to the atmosphere of the poem and to its total effect as this episode of the bird-mating on May-day. F201. *To goon to reste and erlyfor to ryse. To seen this flour to sprede, as I devyse. This couplet consists of a line brought from a different context in G (1. 106) and a new line (F 201) which rimes with it. The lines could very easily be dropped out without destroying the sense (see G 95-96), yet the fact that the line which is common to both versions occurs earlier in G, and that the description of the May scene is consistently fuller and better in F justifies the belief that the other process has taken place, and that this natural touch was added to improve the passage. This couplet makes Chaucer's early morning walk into the meadows an habitual thing and not a special occurrence. F 271-275. 270-275 This balade may ful wel y-songen be *As I have seyd erst, by my lady free ; 72 The Problem of the Two Prologues to *For certeynly, alle these mow nat suffyse *To apperen with my lady in no wyse. *For as the sonne wol the fyr disteyne, *So passeth al my lady sovereyne. If F is to be regarded as the original version, we must account for the omission of these lines. They could not have been omitted as weak lines, nor even as tautological in that they repeat the sense of 11. 247—248, for the introduction of the new figure of speech saves them from bald repetition, and the words, "as I have seyd erst," show that the repetition was conscious and delib- erate. They need not have been omitted because of a change from Ballad F to Ballad G, for they might easily have been recast to apply to Alcestis. We have left only ten Brink's unsatisfactory hypothesis that Chaucer revised in order to get rid of personal allusions. The truth is, I feel sure, that the lines were added to adapt the meaning of the ballad to the Queen, and that they are a direct consequence of the change of " Alceste is here " to " My lady Cometh." The repetition in the devout line (277), I prey to god that ever falle hir faire which is almost identical with F 1 86, I pray to god that faire mot she falle, is probably unconscious, since line 277 is taken over from G with its context. F335. 332-335 And of Criseyede thou hast seyd as thee liste, That maketh men to wommen lasse triste, That ben as trewe as ever was any steel. *0f thyn answer e avyse thee right wed ; The long tirade of the God of Love in G, which Chaucer omitted, probably because it was least essential to the new purpose of the second version, is represented in F by a few significant lines of G and one entirely new line to complete the second couplet. The allusion to the fidelity of women is made general instead of specific, and the harshness of the reproof is greatly mitigated. In spite of the interest of the list of books for the Chaucer student, it is not unfair to say that the loss to the poem is not very great. Oiauoer's Legend of Good Women. 73 F357. 353-358 And many a queyhte totelere accusour, That tabouren in your eres many a soun, Bight after hir imaginacioun, To have your daliance, and for envye ; *These been the comes, and I shed not lye. Envye is lavender of the court alway ; The corresponding passage in G reads : 332-334 And for to han with yow some daliaunce, Envye (I prey to god yeve hir mischaunce 1 ) Is lavender in the grete court alway. The one new line in F is made necessary by the change of the riming word in 1. 356, and is much in the nature of a stop-gap. The general result, however, is advantageous, for the mention of envy m line 356 prepares the way for the fuller description of that vice which immediately follows, links the thought of the two sentences, and does away with the abruptness of the transition from G 332 to G 333. F380. 379-381. He moste think hit is his lege man, *And is his tresour, and his gold in cofre. This is the sentence of the philosophre. Line 380 is inserted to supply the rime lost by the rejection of lines 360-364 of G, which, as has been already said, were probably omitted because of their too specific allusion to the duty of a king to his subjects. F 496-497. And whan this book is maad yive hit the quene On my behalf e, at Eltham, or at Shene. The absence of this couplet from G might possibly be ascribed to accident, if it were not that it agrees with numerous allegorical allusions to the Queen which are also wanting in G. These make it conclusive that the poem is dedicated to the Queen as an allegory in which she is the central figure. The fact that Alcestis, who speaks these lines, is herself in a sense a type of the Queen, does not invalidate this interpretation. This couplet is a touch of reality 74 The Problem 0/ the Two Prologues to suddenly interjected into the account of supernatural events, and is hardly conceived or spoken in character. Whatever incongruity is found in it, however, only serves to heighten the impression, already conveyed by the use of the ballad, that F is an adaptation of G to a purpose for which it was not at first intended. The abruptness of the introduction of the couplet in a part of the Prologue in which no other changes have been made is additional confirmation. However this allusion to the Queen got into F, it is there and must be dealt with. To say that it shows that F was to be read at court, contributes nothing to the solution of the problem. To say that it merely shows that the poem was written at the com- mand of the Queen, is to admit that the G-version, which lacks the allusion, was not so written, and the admission goes far toward confirming the theory that the allegory in F bears a special rela- tion to the Queen. The couplet is, in fact, I feel sure, nothing less than a dedication of the F-version to the Queen of England, and in the light of other passages, an intimation, that the whole poem is to be regarded as an elaborate compliment to her. F 552-577. 552 *But er I go, thus muche I wol thee telle *Ne shal no trewe lover come in helle. *Thise other ladies sittinge here arowe 555 *Ben in thy balade, if thou canst hem knowe, *And in thy bokes alle thou shalt hem finde ; *Have hem now in thy Legend alle in minde, *I mene of hem that been in thy knowinge. *For heer ben twenty thousand mo sittinge 560 *Than thou knowest, that been good women alle *And trewe of love, for aught that may befalle ; *Make the metres of hem as thee leste. *I mot gon hoom, the sonne draweth weste, *To Paradys, with al this companye ; 565 *And serve alwey the fresshe dayesye At Cleopatre I wol that thou beginne ; And so forth ; and my love so shalt thou winne. *For lat see now what man that lover be *Wol doon so strong a peyne for love as she. 570 *I wot it wel that thou mayst nat al hit ryme, *That swiche lovers diden in hir tyme ; *It were to long to reden and to here ; Chaucer's Legend of Good Women. 75 *Suffyceth me, thou make in this manere, *That thou reherce of al hir lyf the grete, 675 *After thise olde auctors listen to trete. *For who-so shal so many a storie telle, *Sey shortly, or he shal to longe dwelle In a part of this long passage peculiar (except 566 and 567) to F, ten Brink saw evidence of a conscious change of plan between the first version and the revision, — a change which made some of these lines no longer appropriate. In a preceding chapter I have endeavored to show that there is no necessity for assuming that Chaucer's plan as to the heroines whose legends he meant to write ever suffered any change that is evident to us. Even if it be granted that some modification of his original plan developed as he proceeded, it is hard to see why any of the lines in this passage should have been taken so literally as to seem inconsistent with the modified scheme. In fact any good reason for the omission of F 552-577 is hard to find. The fact that they make the Prologue fuller and more specific where it is somewhat bald and abrupt in G, and that in their allusion to the poet's books and to an example of fidelity in love, they in some degree take the place of G 267 ff., is good ground for believing them a later addition. IV. Lines Partly Identical in Both Versions. Many minor variations between the two Prologues affect only parts of lines. There remain to be considered, therefore, the lines which are partly identical in the two versions. In these we may expect to find valuable evidence; for whatever the purpose with which the poet undertook his revision, he can hardly have failed in the course of it to make numerous minor improvements, some of which we can hope to discern. Some lines, however, will be found equally good in both versions, and may be considered to differ merely because they were written at different times, and yet others may owe their differences to the vagaries of scribes. In the classification of these variations, which I have regarded as changes made in the G-yersion, the over-lapping of divisions has been avoided so far as possible; but in some cases it has 76 The Problem of the Two Prologues to seemed desirable to discuss a line under two different categories. Improvements in sense and meter often coincide. Changes made chiefly for the sake of consistency with other changes may involve incidental improvement in structure or rhythm. It is inevitable that some variations should be open to controversy, and to one familiar with the history of the problem of the two Prologues it will not be surprising if a few of the arguments should be made in other hands to work the other way. Mathematically exact proof is impossible. What the classification is expected to prove is that the general direction of the evidence is clearly in favor of r as the revised version, and that the conclusions of the preceding chapters are strikingly confirmed. The nature of the classification ^ which I have undertaken may be indicated in general as follows : I. Changes for Metrical Improvement. 1. Changes which avoids harsh or unpleasant repetitions. 2. Changes which obviate direct attack. 3. Changes which result in more sonorous or more fluent lines. II. Changes for Improvement in Grammatical Structure. III. Changes for Improvement in Sense. IV. Changes for Consistency. 1. Changes which conform to the allegorical purpose of F. 2. Changes which conform to other changes. V. Incidental Variations. I. CHANGES FOE MBTBICAL IMPEOVEMENT. 1. Changes which avoid harsh or unpleasant repetitions. G 1 A thousand sythes have I herd men telle, F 1 A thousand tymes have I herd men telle. The repetition of the And /or the newe blisful somers sake .... they turned hem . . . t And songen . . . G is corrupt and apparently incomplete, yet, as it was written, it was inferior to F, which shows decided improvements in sense. In F 139 and al his craft is a valuable addition. Clere belongs with songen and not with braunches as in G, where its adverbial relation is obscured, and the branches are mentioned later in F in a more appropriate sense. G 130 seems to be an illogical combi- nation of the two expressions of the somer and for the somer's sake. This is remedied in F where the line looks forward to a new context. G 145-146 And saw him come, and in his hond a quene, Clothed in ryal abite al of grene. F 213-214 The god of love, and in his hande a quene ; And she was clad in real habit grene. F 214 specializes the queen at once and better prepares for the following description of her. Al of, the only insignificant words in G, are given up. G 149 With many fioures, and I shal nat lye ; F 217 With, fiorouns smale, and I shal nat lye ; Florouns ^ smale is a more specific description than many fioures. G 150 For al the world right as the dayesye I-coroned is ... . P 218 For al the world right as a dayesye Y-corouned is ... . As the comparison here is purely literal, and the daisy has in F already some allegorical significance, the indefinite article is more appropriate. ^If, as seems likely, the word florouns is a borrowing from Froissart's Ditlii (cf. Lowes, p. 631-4) , it by no means follows that F is the original line. Chaucer knew French and might at any time adopt a French word that suited his purpose. But why should he give up the sonorous and descriptive florouns smale for many fioures? Chaucer's Legend of Good Women. 89 G 166-168 But at the last in hand I saw him holde t Two fyry dartes, as the gledes rede ; And aungellich his winges gan he sprede. F 234-236 And in his hande methoughte 1 saugh him holde t Two £yry dartes, as the gledes rede ; And aungellyke his winges saugh J sprede. The delayed perception of what the god of Love had in his hand comes in somewhat oddly after a detailed account of the embroidery of his gown and the garland on his head. This is improved in F. Another improvement is the change to saugh I in F 236, which continues the point of view of vision established by line G, 166. G 221 Mak of your trouthe in love no host ne somi, F 267 Maketh of your trouthe neyther host ne soun. Maketh is an improvement in sense, for being plural it carries out the alle y-fere of F 263, whereas mak can apply gramatically only to Ysiphile. The occurrence of the phrase trouthe in love in G 214 above and the mention of love as the subject of the baUad in G 212 and 215 are perhaps sufficient justification of its omission here. G 222 Nor Ypermistre or Adriane, ne pleyne ; F 268 Nor Ypermistre or Adriane, ye tweyne ; Ne pleyne seems inappropriate' here. The women are not represented as complaining of their fate but as glorying in their fidelity. Hence the propriety of replacing the expression even by the meaningless ye tweyne. The ending of the verse is little more than a tag in both cases. G 226-227 f They setten hem .... By ordre alle in compas, alle enveroun. F 300 And with thai word a-compas enviroun, They setten hem And with that word is inserted primarily as a transition from the preceding passage peculiar to F. By ordre alle may be very well given up on other grounds, however, since the sitting in order is more fully explained later on. ' Cf. Bilderheck, p. 79. 90 The Problem, of the Two Prologues to .... ' What dostow heer G 242 In my presence, and thai so boldely ? ' .... ' What dostow heer F 316 So nigh myn owne flour, so boldely.' There is decided improvement in sense here in that interest at once centers in the flower instead of in the god of Love. It is to be noted that in general the god occupies a more prominent place in G than in F. Changes consistent with this difference in the versions have been made in G 244, 248, and 253. G 313 By seint | Venus | of whom that I was bom. F 338 By selynt Venus that my moder is. Of whom that I was born is a heavy round-about expression which has been well replaced by the more compact that my moder is. G 320 Ageyns these points ' that ye han to him meved. F 344 Agayns al this that ye han to him meved. Al this is less formal and, in the mouth of Alcestis, more appro- priate than the these points of G. G 330 That tabouren in your eres many a ( F 354 That tabouren in your eres many a soun. Soun, besides being more specific than thing, carries out the figure in tabouren. G 351-352 tAs thogh that he of malice wolde endyten Despyt of love, and hadde him-self y-wroght. F 371-372 fAs thogh that he of malice wolde endyten Despyt of love, and hadde him-self hit wroght. The sense of G is obscure and difficult, because an object for hadde y-wroght must be supplied. F is made clear by the insertion of hit. G 387 To dampne a man withoute answere or word. F 401 To dampne a man withoute answere of word. Ansivere or word is illogical. The sense is improved in F by the change to answere of word. ' Was the choice of the word in G influenced by the Lay de Franchise, 11. 222-3, Et quant li Eoys ces doulz mos escouta Touz ces .m. pains a dit qu'il retendra? Chauoer^s Legend of Good Women. 91 G 412 And for to speke of other besinesse, F 424 And for to speke o£ other holynesse, The reason for this change viewed from the standpoint of either version is not obvious. The more religious character of the works which are named after this line has been noticed, and Legouis suggests ^ that holynesse was used in order to be consistent with the mention of Boece, which could not be easily dropped out. It must be remembered, however, that the phrase other holynesse looks backward as well as forward, and the term holynesse applies equally to all the works named. It is more probable that we have here simply another change to the more specific and picturesque term. The poet is represented as on trial before an angry god, and his advocate is reciting his virtuous acts in his behalf. The works which she first names were written, she tells the god, " To serve you, in preysing of your name," and in F the fiction of piety is kept up by calling the poems holynesse instead of keeping the colorless term besiness. Q 440 And al foryeve, with-outen lenger space ; F 450 I al foryeve, with-outen lenger space ; The offense of the poet has been committed against the god of Love, not against Alcestis, who intercedes in his behalf. His forgiveness, therefore, should come from the god of Love, and the change, slight as it is, was probably not accidental, but was made as an improvement in sense.^ G 471-472 fThou shalt, whyl that thou livest, yeer by yeer, The moste party of thy lyve spende F 481-482 fThou shalt, whyl that thou livest, yeer by yeer, The moste party of thy tyme spende The change from lyve to tyme is an improvement in sense. Whyl 'Legouis, p. 68. " Lowes (p. 671, n. 4) regards this change as a " rude blow" to the theory that Alcestis is Queen Anne. This is pushing both the literal meaning and the alle- gory entirely too far. The avowed purpose of Alcestis was to reconcile her lord to the poet (cf. Unes 419 fl.), and her responsibility for the reconciliation is fully recognized in both versions. Though the god may easily depute to Alcestis the assignment of penance, forgiveness belongs, in the nature of things, to himself alone. 92 The Problem of the Tim Prologues to that thou livest is sufficient intimation that the task is to be life-long. Lyve in the next line is needless repetition, which is obviated by the change to tyme, making better sense and more fluent verse. .... Kalender is she G 543-544 Of goodnesse, for she taughte of fyn lovinge, tAnd namely of wyfhood the livinge. F 544-545 Por she taught al the craft of fyn lovinge. fAnd namely of wyfhood the livinge. The sense is somewhat clearer and simpler in F. G paraphrased means, ' for she taught the living of fine loving and especially of wifehood,' whereas in F the sense is, ' for she taught the craft (or art) of fyn loving and the living of wifehood.' rV. CHANGES FOB CONSISTENCY. 1. Changes which conform to the aUegorical purpose of F. A general comparison of the two versions (Chapter in) has seemed to justify the conclusion that F is marked by an allegorical purpose which is wanting in G. This conclusion is borne out by various minor changes, which are to be satisfactorily accounted for only on the ground that they were made for the sake of consistency with that purpose. Such changes are the following : G 48 To seen these floures agein the sonne sprede. F 48 To seen this flour agein the sonne sprede. The change to the singular is not only a gain in definiteness, but a first hint of the personification of the daisy in F. G 53 So sone hit is afered of the night, F 63 For fere of night so hateth she derknesse. The tendency toward personification in F here becomes clear. It is a part of the enlivening which the whole passage has received from revision. G 57-58 fAnd ever y-lyke fair and fresh of hewe, As wel in winter as in somer newe, F 55-56 tAnd ever y-lyke fair, and fresh of hewe, And I love hit, and ever y-lyke newe, Chaucer's Legend of Good Women. 93 The propriety of the mention of winter in G 58 has already- been fully shown by Lowes.^ It is given up here, not as has been surmised, because it is inconsistent with fact or literary convention, but because an avowal of constant love for the daisy is more im- portant for the new purpose than a bit of description which is susceptible of literal interpretation only. G 61 For wel I wot thai folk han her-befom F 73 For wel I wot that ye han her-biforn The direct appeal in F to " lovers that can make of sentement " for help " in this cas " enlivens the following lines by throwing them all into the second person. We may pass over the substitu- tion of ye for they in G 65 and 67 and proceed to a comparison of the concluding lines of the passage. G 68-80 I hope that they wil nat ben evel apayed Sith hit is seid in forthering and honour Of hem that either serven leet or flour. F 80-2 For-bereth me, and beth nat evel apayed Sin that ye see I do hit in the honour Of love, and eek in service of the flour, *Whom that I serve These lines indicate distinctly the difference in purpose between the two versions. One is simply a "flower and leaf" poem glori- fying wifely fidelity with Alcestis as its crowning type ; the other is written in the service of a flower, which is fervently invoked in terms of personification. The whole succession of changes con- tributes to the allegory. G 158-62 fY-clothed was this mighty god of love Of silk, y-brouded ful of grene greves ; A garlond on his head of rose-leves *Steked al with lilie floures newe. F 227-30 In silk, enbrouded ful of grene greves In-with a fret of rede rose-leves *The fresshest sin the world was first bigonne *His gUte heer was corouned with a sonne, In F the rose-leves are transferred from the garland on the god's head to the embroidery of his robe. A crown of gold is 'Lowes, p. 617, n. 1. 94 The Problem, of the Two Prologues to now his due as a sign of royalty, but on account of his ethereal origin he wears a sun instead, " for hevinesse and wighte." That the change has been made to accommodate the description to an allegorical interpretation in which the god of Love becomes the king of England is possibly a fair inference. This is noted by Koch,^ approved by Bilderbeek,^ and rejected by Lowes.^ Legouis* suggests, and not unreasonably, that the sun is more consistent with the following allusion to brightness. The objection of Lowes based upon the heat of the sun is meaningless, for it is " a sun " that is here mentioned and allusion is poetical. G 179 Sir name was Aleeste, the debonayre ; F 275-276 *So passeth al my lady sovereyne, That is so good, so fair, so debonaire ; The suppression of the name of Alcestis in the ballad makes a change here necessary. It is to be noticed that the epithet debonaire is now applied, not to Alcestis but to my ladye sovereyne. G 234 I, lening faste by under a bente. F 308 I kneling by this flour, in good entente In G the poet is merely the conventional spectator, who remains half-concealed to watch the company and "knowen what this peple mente," and who is surprised in the act." In F he is a worshipper of the daisy, and is discovered kneeling devoutly beside the flower. This greater devotion is characteristic of F. G 242 In my presence, and that so boldely. F 316 So nigh myn owne flour, so boldely. In F the flower stands for the Queen, into whose royal presence the poet has ventured. In G no such allegorical meaning attaches to it, and the god of Love rebukes the intruder for being in his presence. G 355 That usen vnlfulhed and tirannye. F 375 That ban no reward but at tirannye. The sweeping expression of F, that care for nothing but tyranny, ^Englische Studien, xxx, 457. ^Bilderbeck, p. 86. 'Lowes, p. 675, n. 1. *Legouis, p. 68. 5 a. Paradys d' Amours, 11. 60-1, 228-36. Chaucer's Legend of Good Women. 95 would be less likely to give offense to a royal patron than the more tempered expression, that practise wilfulness and tyranny, which might be construed to apply to a headstrong king of Eng- land. This is only a part of a general toning down noted by Skeat.» 2. Changes which conform to other changes. Q 51 And whan the sonne ginneth for to weste, F 61 As sone as ever the sonne ginneth weste, This change, an improvement in itself, is needed to conform to the introduction in F of several new lines the last of which, F 60, is And whan that hit is eve, I renne blyve. After this line the and whan of G must be given up. If the change had proceeded in the other direction and the lines peculiar to F had been dropped out, no alteration of F 61 would have been necessary or even desirable. G 55 This dayesye, of alle floures flour, 59 Fain wolde I preisen. F 52-53 ... to doon al maner reverence As she, that is of aUe floures flour. The introduction in F of three lines, 50-52, of enthusiastic devotion to the daisy is responsible for the change in G 56, which is made to continue the sense and the spirit of the passage. The change is away from conventionality to vigor and liveliness of expression.^ G 71-72 For tmsteth wd, Ine have nat undertake As of the leef, ageyn the flour, to make ; F 186-189 *I pray to god that faire mot she falle, *And alle that loven floures, for hir sake ! Bvi naihdes, ne wene nat that I make In preysing of the flour agayn the leef. ' Oxford Chaucer, in, xxv ; Bilderbeck, 95 S. 'The "utter shipwreck of grammar" noted by Lowes (p. 660) is after all rather an innocent slip. For a similar anacoluthon cf . Canterbury Tales, A 4083, Unthank come on his hand that bond him so And he that bettre sholde han knit the reyne. 96 The Problem of the Two Prologues to The almost complete change of G 71 to F 188 is due to the fact that the passage introduced in G by line 71 has been trans- posed to a new context in F, where it follows an impassioned invo- cation of the daisy. This makes the adversative sense of hut natheks appropriate. G 81-84 fBut wherfor that I spak, to yeve credence To bokes olde and doon hem reverence, Is for men shvlde autoritees behoe, Ther as ther lyth non other .assay by preve. F 98-101 To olde stories, and doon hem reverence, And that men mosten more thing beleve Then m«n may seen at eye or elles jpreoe f *That shal I seyn, whan that I see my tyme. The changes in G 83 and 84, which alter the whole trend of the passage making it interrogative, are made for the sake of con- sistency with the postponement which follows in F 101, and are part of a larger improvement already discussed. G 108-110 With floures swote enbrowded was it al ; As for to speke of gomme, or erbe, or tree tCompaiisoun may noon y-maked be ; * F 120-122 *0f svdch swetnesse and swich odour overal, That, for to speke of gomme, or herbe, or tree, tComparisoun may noon y-maked be ; The change of G 109, as for to spelee to thai, for to speke is made possible by the use of swieh 'va. the preceding liae peculiar to F. The result is a clearer and firmer construction. G 144-145 Tho gan I lohen enddong the mede, And saw him come, and in his hond a quene, F 212-213 And from a-fer com walking in the made The god of love, and in his hande a quene ; These lines in G immediately follow the exclamation of the lark, which is omitted in F. The omission of this awkward device leads to the use of the more direct lines in F and the im- mediate appearance of the real subject, the Queen. G 224 Whan that this balade al y-songen was. F 270 This balade may ful wel y-songen be, The changed purpose of the ballad and the introduction of five Chaucer's Legend of Good Women. 97 new lines of praise for " my lady free," have necessitated a com- plete change in G 224. G 238 And seyde, ' who resteth ther ? ' and I answerede F 312 And sejde, ' who kneleih ther ? ' and I answerede This change is made for the sake of consistency with F 308. G 266-268 *Hast thou nat mad in English eek the book How that Crisseyde Troilus forsook, In shewinge how that wemen han don mis? F 332-333 And of Criseyde thou hast seyd as thee liste That maketh men to wommen lasse triste, The omission of the "old age passage," G 258-264, has deprived lines G 266—268 of their context and made it necessary to recast these lines in F. G 288 For to hir love were they so trewe, F 334 That ben as trewe as ever was any steel. These lines are not exactly parallel. The variation is due to the omission of lines G 267-312. The substance of those lines, which recount instances of fidelity, is tersely expressed in the single line of F. G 315 As other oldefoles many a day, F 337 As other vrrecches han doon many a day. The omission of the allusion to olde foles in G 262 may be responsible for the change of G 315. Both changes were probably due to a desire to tone down the somewhat blunt character of this part of the Prologue. G 370 And for to kepe his lordes hir degree, F 384 Al wol he kepe his lordes hir degree, G 370 follows a couplet which alludes to a king's oath of ofiice, and which is wanting in F. The omission of it makes necessary a change in F 384, which becomes the first line of a new sentence. G 532-533 *For of Alceste shulde thy wryting be Sin that thou wast that kalender is she F 541-540 *Sin that thou art so gretly in her dette And woit so viel, that kalender is she 98 The Problem of the Two Prologues to G 533 has been changed to adapt it to the new context which it has in F. V. CHANGES INCIDENTAL TO THE EEVISION AND NOT SIGNIFICANT AS TO ITS DIRECTION. A number of minor variations between the two versions have found no place in the classification thus far, many, because they are so slight as to seem almost accidental, and others, because no argument as to the priority of either form can safely be based upon them. For the sake of completeness these changes are here set down in the order of their occurrence in the G Prologue. G 29 And as for me, though that my wit be lyte. F 29 And as for me, thogh that lean but lyte. G 77 I not who serveth leef, ne who the flour. F 193 Ne I not who serveth leef, ne who the flour. G 78 That nis nothing the entente of my labour. F 194 Wei brouken they hir senrice or labour. G 131 That songen, ' blissed be seynt Valentyn 1 ' F 145 And songen, ' blessed be seynt Valentyn 1 ' G 132 [For] at his day I chees yow to be myn, F 146 For on his day I chees yow to be myn, G 135-136 And after diden other observaunces. Eight [plesing] un-to love and to nature ; F 150-151 To love, and diden hir other observaunces, That longeth un-to love and to nature ; G 153 For of o perle/y» amd oriental F 221 For of o yeile/yne, oriental G 157 Considered eek the fret of gold above. F 225 Considered eek hir fret of gold above. G 170 Al-gate me thoughe he mighte wd y-see ; F 238 Al-gate me thoughte that he mighte see G 180 I prey to god that ever faUe she fayre ! F 277 I prey to god that ever falle hir faire ! G 214 Hyde ye your trouthe in love and your renoun ; F 260 Hyde ye your trouthe of love and your renoun ; G 215 And thou, Tisbe, that hast /or love swich peyne. F 261 And thou, Tisbe, that hast of love swich peyne. Chaucer's Legend of Good Women. 99 G 217 Herro, Dido, Landomia, alle wr-fere, F 263 Herro, Dido, Laudomia, alle y-fere, G 218 Eek Phyllis, hanging for thy Demophoun, F 264 And Phyllis, hanging for thy Demophoun, G 231 As they were of degree, ful curteisly ; F 305 As they were of eataat, ful curteisly ; G 237 The god of love on me his eye caste, F 311 This god of love on me his eyen caste, G 239 Un-to his wmig, whan that I him herde, F 313 Unto his asking, what that I hit herde, G 251 And lettest folk to han devocioun F 325 And lettest folk from, hir devocioun G 253 To troste on me. Thou mayst hit nat denye ; F 327 To serve Love. Thou mayst hit nat denye ; G 254 For in pleyn text, hit nedeth nat to glose, F 328 For in pleyn text, withouten nede of glose, G 314 Although [thai] thou reneyed hast my lay, F 336 For, thogh that thou reneyed hast my lay, G 338 This man to yow may wrongly been accused, F 350 This man to yow ma,jfakly been acciised, G 340 Or eUes, sir, for that this man is nyce F 362 And eek, paraunter, for this man is nyce G 341 He may translate a thing in no malyce, F 363 He mighte doon hit, gessing no malyce, G 342 But for he useth bohes for to make, F 364 But for he useth thinges for to make G 375 Al be that here stat be nat a-liche, F 389 Al be that hir estai be nat y-Uche, G 439 Al lyth in yow, doth with him what yow leste F 449 Al lyth in yow, doth with him as yow leste G 451 That han me holpen, and put in swich degree, F 461 That han me holpe and put in this degree, G 467 In right ne wrong ; and leme tMs at me 1 F 477 In right ne wrong ; and lerne thai of me ! G 479 For in your world that is now holden game. F 489 For in your world that is now holde a game. G 480 And thogh thee lesteth nat a lover be. F 490 And thogh thee lyke nat a lover be. 100 Chaucer's Legend of Good Women. V. CoKCLtrsiON. If my inferences from the evidence presented in the preceding chapters have been just, the cumulation of proof makes certain conclusions appear inevitable. In the first place, it is clear that the version which I have called Gr, the text of Ms. Gg. 4, 27, is an earlier draft of the Prologue, and that F, the commonly accepted text, is the later. This has been proved, not, perhaps, for all time nor beyond all question, but with as high a degree of certainty as can reasonably be expected in the solution of such a problem. The first impressions of Furnivall, Kunz, Skeat, and others have been fully confirmed. The arguments of ten Brink and Lowes, which make against the traditional opinion, have been examined in detail, and have been found to be not only inadequate in themselves but also overwhelmingly contradicted by evidence on the other side. It has been shown, in the second place, that the later version is an adaptation of the earlier to meet the demands of a new purpose. This purpose is the celebration, under the figure of the daisy, and » through the daisy, of Alcestis, of the virtues of the queen of Eng- land, to whom the later version is dedicated. The assumption that such a purpose existed is not only justified by the language of the F-version, but is demanded by many of the changes, which are inexplicable without it. These two conclusions, constituting in themselves the end which was to be attained by this investigation, are not inter-dependent. If the allegorical interpretation of F should be denied, or extended to G also, the evidence in favor of F as a revision would remain intact. Data for determining the time or the occasion of the revi- sion are not available, and the question of date is, moreover, not pertinent to this discussion. There is, however, no evidence to indicate that the two versions are separated by any considerable lapse of time. This study of the relation of the two Prologues to each other was suggested by a general discussion of the problem in the English Seminary of the Johns Hopkins University. My thanks are due to Mr. Jas. E. Routh, Mr. B. A. Wise, and other members of the seminary for helpful notes and suggestions placed at my disposal at the outset. LIFE. I was bom in Warren County, Illinois, in the year 1875; attended the public schools of that county and of Baltimore ; and received the degree of A. B. from the Johns Hopkins University in 1899. The next year I spent at Harvard as a graduate student in English. In 1900 I returned to Johns Hopkins, entered the English seminary, was appointed Student Assistant in English, and reappointed in 1901 and 1902. I held a Hopkins Scholar- ship during my undergraduate course and was Fellow in English in 1903-1904. I gratefully acknowledge my indebtedness to the authorities of the University, and my specific obligations to Professors Greene and Griffin, Associate Professors Armstrong and Ogden, and especially to Professor Bright, under whose direction this disserta- tion was written, for his inspiring example of sound scholarship, and his patient and penetrating criticism. Cornell University Library PR 1882.F87 The problem of the two prologues to Chau 3 1924 013 110 832 ©DR ^8''<*2(T^BOfeATior^ '^H FJ"'* JA^i^ m^.A^ ■^AA^^^^*- — iiliMMHiMk. iAPW 7(386. CAVLORO PRINT CD INU.B.A.