The first allusion to any distinct variety of Romance is found in the life of a certain St. Mummolin who was Bishop of Noyen, succeeding St. Eloi in 659. A reference to the Spanish type of Romance is found as far back as 734; but the authenticity of the document is very doubtful. The breaking-up of Latin in Spain is certainly observable in Bishop Odoor's will under the date of 747. The celebrated Strasburg Oaths, the oldest of Romance instruments, belong to the year 842; and, in an edict of 844, Charles the Bald mentions, as a thing apart, "the customary language"—usitato vocabulo—of the Spaniards. There is, however, no existing Spanish manuscript so ancient, nor is there any monument as old, as the Italian Carta di Capua (960). The British Museum contains a curious codex from the Convent of Santo Domingo de Silos, on the margin of which a contemporary has written the vernacular equivalent of some four hundred Latin words; but this is no earlier than the eleventh century. The Charter called the Fuero de Avilés of 1155 (which is in bable or Asturian, not Castilian), has long passed for the oldest example of Spanish, on the joint and several authority of González Llanos, Ticknor, and Gayangos; but Fernández-Guerra y Orbe has proved it to be a forgery of much later date.
These intricate questions of authority and ascription may well be left unsettled, for legal documents are but the dry bones of letters. Castilian literature dates roughly from the twelfth century. Though no Castilian document of extent can be referred to that period, the Misterio de los Reyes Magos (The Mystery of the Magian Kings) and the group of cantares called the Poema del Cid can scarcely belong to any later time. These, probably, are the jetsam of a cargo of literature which has foundered. It is unlikely that the two most ancient compositions in Castilian verse should be precisely the two preserved to us, and it is manifest that the epic as set forth in the Poema del Cid could not have been a first effort. Doubtless there were other older, shorter songs or cantares on the Cid's prowess; there unquestionably were songs upon Bernaldo de Carpio and upon the Infantes de Lara which are rudely preserved in assonantic prose passages of the Crónica General. An ingenious, deceptive theory lays it down that the epic is but an amalgam of cantilenas, or short lyrics in the vulgar tongue. At most this is a pious opinion.
To judge by the analogy of other literatures, it is safe to say that as verse always precedes prose (just as man feels before he reasons), so the epic everywhere precedes the lyric form, with the possible exception of hymns. The Poema del Cid, for instance, shows no trace of lyrical descent; and it is far likelier that the many surviving romances or ballads on the Cid are detached fragments of an epic, than that the epic should be a pastiche of ballads put together nobody knows why, when, where, how, or by whom. But in any case the cantilena theory is idle; for, since no cantilenas exist, no evidence is—or can be—forthcoming to eke out an attractive but unconvincing thesis. In default of testimony and of intrinsic probability, the theory depends solely on bold assertion, and it suffices to say that the cantilena hypothesis is now abandoned by all save a knot of fanatical partisans.
The exploits of the battle-field would, in all likelihood, be the first subjects of song; and the earliest singers of these deeds—gesta—would appear in the chieftain's household. They sang to cheer the freebooters on the line of march, and a successful foray was commemorated in some war-song like Dinas Vawr's:
"Ednyfed, King of Dyfed,
His head was borne before us;
His wine and beasts supplied our feasts,
And his overthrow our chorus."
Soon the separation between combatants and singers became absolute: the division has been effected in the interval which divides the Iliad from the Odyssey. Achilles himself sings the heroes' glories; in the Odyssey the ἀοιδός or professional singer appears, to be succeeded by the rhapsode. Slowly there evolve in Spain, as elsewhere, two classes of artists known as trovadores and juglares. The trovadores are generally authors; the juglares are mere executants—singers, declaimers, mimes, or simple mountebanks. Of these lowlier performers one type has been immortalised in M. Anatole France's Le Jongleur de Notre Dame, a beautiful re-setting of the old story of El Tumbeor. But between trovadores and juglares it is not possible to draw a hard-and-fast line: their functions intermingled. Some few trovadores anticipated Wagner by eight or nine centuries, composing their own music-drama on a lesser scale. In cases of special endowment, the composer of words and music delivered them to the audience.
Subdivisions abounded. There were the juglares or singing-actors, the remendadores or mimes, the cazurros or mutes with duties undefined, resembling those of the intelligent "super." Gifted juglares at whiles produced original work; a trovador out of luck sank to delivering the lines of his happier rivals; and a stray remendador struggled into success as a juglar. There were juglares de boca (reciters) and juglares de péñola (musicians). Even an official label may deceive; thus a "Gómez trovador" is denoted in the year 1197, but the likelihood is that he was a mere juglar. The normal rule was that the juglar recited the trovador's verses; but, as already said, an occasional trovador (Alfonso Álvarez de Villasandino, at Seville, in the fifteenth century, is a case in point) would declaim his own ballad. In the juglar's hands the original was cut or padded to suit the hearers' taste. He subordinated the verses to the music, and gave them maimed, or arabesqued with estribillos (refrains), to fit a popular air. The monotonous repetition of epithet and clause, common to all early verse, is used to lessen the strain on the juglar's memory. The commonest arrangement was that the juglar de boca sang the trovador's words, the juglar de péñola accompanying on some simple instrument, while the remendador gave the story in pantomime.