On the Nature of the Cantar de mio Cid and its Place in Hispanic Medieval Epic

ON THE NATURE OF THE CANTAR DE MIO CID AND ITS PLACE IN HISPANIC MEDIEVAL EPIC Juan Carlos Bayo University of Exeter In memoliant John Cornali In die last four decades no issue on the Cid has been so controversial as the problem of die poem's composition, no doubt because of its many implications. The starting point of the modern discussions on the topic can be traced to L.P. Harvey's article of 1963, which called attention to the following paragraph of Albert B. Lord's The Singer of Tales: It is not to be wondered at that, when the singer is asked to dictate, stopping at the end ofeach verse, he is uncertain where to stop, and hesitates also as to die number of syllables in a line. Frequendy he will give a whole sentence in prose. He is, after all, telling a story. As regards the forming ofverses, songs recited for die records and songs dictated but taken down by a scribe who does not seek to obtain good rhythmic lines are about die same. They look very much like the text of the Old Spanish Cid or that of the Escorialensis manuscript of the medieval Greek Digenis Akritas with dieir "irregularities" of meter. (127) Harvey did not attempt to substantiate the remark about the Cid. Indeed, its accuracy was doubtful. There is no proof that the lines transmitted by the Vivar codex are ill-formed, even if no explanation of their structure has yet won general consent. Robert A. Hall retorted that Old Spanish poetry includes an important corpus of anisosyllabic La corónica 33.2 (Spring, 2005): 13-27 14Juan Carlos BayoLa coránica 33.2, 2005 poems, and diat there is no ground to assume diat all Spanishjongleurs were unable to dictate; however, his remark was confined to a footnote, and does not seem to have been much noticed (227 n. 3). As far as the earliest remnants of medieval Hispanic epic are concerned, Ramón Menéndez Pidal had established the original anisosyllabism of the Cid in his 1908-1 1 edition, and the discover)' of the Navarrese Roncesvalles fragment, first published by him in 1917, was further proof that he was right in this respect. Our inability to explain the system of versification of these Old Spanish epic left the doors open to die most misleading ideas. Lord's analogy between the fieldwork ofa folklorist and the writing down of a poem such as the Cid raises a number of doubts. Distortions during dictation are one of the usual problems encountered by folklorists, especially before the advent of phonographic recording; the collection of romances also provides examples.1 However, the hypothesis that the extant text of the Cid. was copied from dictation could not. develop, no doubt because there is nothing in the Hispanic oral tradition that comes close to it in form, extension and complexity, but also because there was at that time no theory that made orality the key concept for the study ofepics. As is well known, this is a framework based on principles developed by Milman Parry, Lord's mentor, who in 1928 demonstrated that the composition of the Homeric poems was enabled by a formulaic system. The concept of formula was defined by him as "a group ofwords which is regularly employed under the same metrical conditions to express a given idea" (272). Parry's success was only possible because he was able to address all the questions involved in this definition: How many words and what degree of variation are admitted? How many times must a group of words appear in order to be considered a formula? What metrical 1 At the end ofthe nineteenth century Carolina Michaelis de Vasconcellos remarked that ballads must be collected carefully; not "dissolvendo pouco a pouco em prosa incoherente e em versos estropeados o que era bellissima poesia" ( 1 890-92, 161), and criticized the rush job ofa folklorist in the following terms: "Mas como o tempo de que dispunha foi limitado, contentoii-se com os materiaes incorrectos que lhe estavnm á niño e eiam de faci] accesso, escrevendo o que Carmen e Antonia dictavam ad hoc e de...