21
It has always seemed strange that this should be included in the list, since we have no way of knowing whether or not the Cid had such a vision.
22
La España del Cid, 5th ed. (Madrid: Espasa-Calpe, 1956), II, pp. 555, 556 and 562.
23
Menéndez Pidal, 64 pp. 86-7, analyses the 987 lines whose number of syllables can be determined beyond doubt (there are objections to this method, as Jules Horrent shows in Roncesvalles [Paris, 1951], p. 69n, but it is difficult to see any alternative). Of these lines, some 3.5 per cent have 10 or 11 syllables, some 7.5 per cent have between 17 and 20, and the remainder are symmetrically distributed, with 14 syllables as the most frequent length.
24
Leonard's work was published in 1928-31. Another aspect of his theory of Germanic influence is revived and developed in Strausser's work on alliteration, 113. Strausser's evidence falls well short of conviction, partly because she is forced to postulate a type of alliteration so different from the Germanic that the connection becomes tenuous, and partly because she finds only 60 lines (less than two per cent of the Poema) with three or more alliterating, sounds. See also F. Maldonado de Guevara, «Knittelvers "Verso nudoso"» RFE, XLVIII (1965), 39-59.
25
See also Jean-Claude Chevalier, «Architecture temporelle du Romancero tradicional», Bulletin Hispanique, LXXIII (1971), 50-103.
26
Ruth House Webber applies this test to Roncesvalles: «The Diction of the Roncesvalles Fragment», in Homenaje a Rodríguez-Moñino (Madrid: Castalia, 1966), II, pp. 311-21; and her article on epic epithets, 124, is also relevant here.
27
De Chasca himself refers to «17% de hemistiquios formularios en 7.460 hemistiquios», but also to 32.9 as the «porcentaje de versos formularios en el Poema»
(24 pp. 332-4; the figures in 25 pp. 90-2 are slightly different). On the basis of de Chasca's own figures, however, it seems to me that by his criteria 28.1 per cent of the Poema's hemistichs are formulaic.
28
«Formulas in the Chronicle of the Morea», Dumbarton Oaks Papers, XXVII (1973), 163-95. For other studies, see Edward R. Haymes, A Bibliography of Studies Relating to Parry's and Lord's Oral Theory (Cambridge, Mass.: Harvard Univ., 1973); Samuel G. Armistead's review, to appear in MLN, will fill some gaps in Haymes's coverage of Hispanic material.
29
El arte juglaresco en el CMC (24) began as a reworking of 23, but is so much longer, and has so different an emphasis, that it is better to consider them as separate books.
30
Materia mirable: estudio de la composición numérico-simbólica en las dos obras contemplativas de Juan de Padilla, el Cartujano (Groningen: privately publ., 1972), pp. 254-5.