composition heading his Cancionero inédito ', an example of one of those coplas which he himself characterized as llenas de mocedades, runs as follows: Que vos que en penas n.... y males que me vi.... por que nos adol.... á mis ojos que... pues que no... 3 It will be observed that while Cervantes cuts only the feminine ending of the rhyme-word, Juan Álvarez Gato expects the reader to find even more than the last unaccented syllable. The fact that neither Amador de los Ríos nor Menéndez y Pelayo nor the editor Cotarelo y Mori have called attention to the peculiar form of this stanza, may be due to their ascribing it to a mere corruption of the text in the only ms. in which it is extant 4, they do not say anything to this effect but far more likely to their taking little, if any, notice of the matter of metrical form. As it is not very probable that the absence of one or more syllables of the rhyme-words in each of the ten lines quoted is due to a mere chance, we may consider that it was intended by the poet and that it is, therefore, a case in point. Ascending from the time of Álvarez Gato, we come to another variety of the verso de cabo roto in the poetry of the Catalan school whose relations with Castilian literature were scarcely 1. Published by Cotarelo y Mori, Madrid, 1901. 2. Historia, VII, 123 ff. 3. 1. c. 4. Academia de la Historia, C. 114. less intimate, if somewhat later, than had been those of the Gallego-Portuguese lyric'. The first instance is a composition by Pere Galvany, a poet of the beginning of the fifteenth century 2, printed in the so-called Cancionero de Zaragoza (edited Zaragoza 1896), p. 2343: In this example, as will be seem the rhyme-words, identical in the corresponding lines of the two stanzas, are cut off from the lines themselves, but written opposite them on the margin. A similar case occurs in an anonymous composition of the same Cancionero, p. 244-246, of the ten stanzas of which the first two may answer the purpose of illustration : I. I shall have occasion to recur to this point further on. 2. In Milá y Fontanals, Obras completas, III, 354. 3. The rhyme-words which in the ms. are written out on the margin, are here put in italics. The second stanza is omitted here as having the same rhymes and the same system. 4. Expression equivalent to « fulano y zutano ». See Milá, 1. c. 5. Vas sepulcro, e. g., see Milá, 1. c. = In this poem which has the rhyme-order abba cddc, the accented rhyme-vowels are the same in the corresponding lines of all the stanzas, only the consonants before the atonic final syllables changing from stanza to stanza. The final syllable, which I have marked by italics, is in the ms. written on the margin opposite the stanza. With this final syllable, we obtain the effect of assonance between the corresponding lines of the several stanzas; without it that of masculine rhyme. I Both of these Catalan poems agree with two compositions of the Picara Justina (Rivad. p. 69 and p. 131) in this that the truncated syllables are written on the margin. In the two Tercetos de pies cortados (ib. 131), four of the six lines have no consonance at all. In view of the concordance of practice observable in the compositions cited from Catalan poetry 2, from Álvarez Gato and 1. In his note to the anonymous piece, the editor of the Cancionero de Zaragoza (p. 244) refers to other Catalan instances of this artifice, contained in the so-called Cançoner d'amor of Paris (in Morel-Fatio, Catal., p. 196 b), and in the works of the Valentian school, instances which are not accessible to me. 2. In commenting on the poem of Pere Galvany, Milá (Jahrbuch für romanische und englische Literatur, V, 187-8; Obras, III, 354) remarks: «< Todos los the Picara Justina, it would seem safe to assume that the principle of the versos de cabo roto existed in the poetical tradition of the sixteenth century and that it was to this tradition that Cervantes and his less illustrious contemporaries were equally indebted for this artifice 1. H. R. LANG versos que en las estrofas se corresponden, terminan con la misma palabra cas, payre, etc. Esta combinación no se halla prevista en las Leys d'amors, las cuales ofrecen como primer modelo de « coblas retronchadas » una estrofa cuyos versos terminan todos con la misma palabra (I, 280). » It is obvious that Milá, probably misled by Raynouard (Lexique Roman, V, 80 and 481), wholly misunderstood the meaning of the term coblas retronchadas, taking it in the sense of stanzas with truncated rhyme-syllables, which it did not have, instead of strophes with a refrain, a sense which is perfectly clear from the definition given in the Leys d'amors (I, 286). This definition accords with that found in the Catalan Doctrina de compendre dictats edited by P. Meyer in Romania, VI, 355-358, but also with the form of the compositions so named. Cf. e. g. Wolf, Studien zur span. u. port. National Literatur, 262 and P. Meyer, Romania, XIV, 38-41. For the etymology of the term retroncha or retroencha, see Suchier, Zeitschrift. f. rom. Philol., XVIII, 282. As to the Provençal expression for a truncated », see following note. 1. Provençal poetry knew another form of broken rhyme, too different in character from the one under discussion to be considered here. This is the one which in the Leys d'amors (1, 52 and 196) is called rims trencatz e sillabicatz and which consists in dividing the final word of a line so that its first syllable rhymes, and its last syllable forms the beginning of the next line. The rims trencatz does not occur to my knowledge, in the now extant verse of the Castilian lyric, but was practiced by the Gallego-Portuguese School (11751350). See Liederbuch des Königs Denis von Portugal, p. cxxvi, and Tobler's correction of the text of no LIX, in Archiv für das Stud. d. n. Sp., 1895, 472. REVUE HISPANIQUE. XV. 7 COMPRESSION IN THE POEMA DEL CID'. Does the Poema del Cid in its present form represent with completeness the original composition? Baist in the Grundriss der Rom. Phil. says: «Wir besitzen im Wesentlichen das Lied noch so wie es um die Mitte des 12 Jhrs. oder kurz nach ihr gedichtet ist ». Contrary to Baist the purpose of this thesis is to show that there is sufficient evidence to support the belief that during the process of transmission, many passages of varying length have either dropped out or been compressed into sum maries. Close scrutiny of the evidence seems also to suggest that the author of the Poema based his work on other poems. That these poems were not ballads nor of the nature of lyrico-epic poems sprung from the immediate impression of events, has been sufficiently proved by Milá y Fontanals and others. And the unity of the Poema itself precludes the idea that the author was a mere compiler of existing material. On the contrary the Poema is a work of art. But like other literary works of art, it must have had predecessors. That such existed seems to be borne out by the existence of other poems on the Cid and by the internal evidence of compression. 1. This thesis was accepted by the faculty of Harvard University, 1906, in partial fulfillment of the requirements for the degree of Ph. D. Thanks are due by the author to Prof. J. D. M. Ford, who suggested the subject and offered several kind criticisms during revision for the press. 2. Grund. Rom. Phil. 12, p. 397. |