Selecciona una palabra y presiona la tecla d para obtener su definición.
Indice Abajo

The Singer of Tales and Mediaeval Spanish Epic

Alan Deyermond





Dr L. P. Harvey's recent article is the first attempt to assess the relevance of recent work on oral-formulaic poetry to the study of mediaeval Spanish epic1. Investigation of modern Yugoslav oral poetry was begun in the early 1930s by Milman Parry, and continued after Parry's death by Albert B. Lord, whose book The Singer of Tales (Cambridge, Mass., 1960) summarizes the results so far. Parry and Lord were chiefly concerned with the effect of their investigations on Homeric scholarship, but Lord includes in his book a chapter on mediaeval epic with, as Harvey says, two almost casual references to the Cantar de Mio Cid. At the same time, there have been attempts to apply Lord's methods, which had previously been outlined in a number of articles, to Anglo-Saxon and French material2.

Harvey has convincingly demonstrated the relevance of the Yugoslav evidence to the problem of the Cantar de Mio Cid's metrical irregularity. My aim in this article is to continue the discussion by examining some of the difficulties involved in any attempt to use Lord's work as a key to other problems of Spanish epic. First, however, it may be useful to indicate briefly some more of the problems to which the Parry-Lord type of investigation may provide answers. Comparison of the unique manuscript of CMC with the version prosified in the Primera Crónica General shows a fairly close agreement in the early part of the poem, but a widening divergence later on. This was one of the facts relied on by E. C. Hills in his argument for dual authorship3. However, even assuming that the PCG really does incorporate a reworking of CMC4, an alternative and, it seems to me, more likely explanation is provided by Lord: «One of the reasons also why different singings of the same song by the same man vary most in their endings is that the end of a song is sung less often by the singer» [The Singer of Tales, 17), the reason for this being the impatience of the audience on most occasions. The same consideration would, presumably, apply even more forcibly to performances by different singers: the early part of a poem would, by frequent and extensive performance, become relatively stable, while the less frequently performed and usually more summary later part would make a greater call on any singer's powers of improvisation.

Another worrying feature of CMC has been its occasional inconsistency: the capture and unexplained reappearance of Álbar Salvadórez (lines 1681 and 1994), the unexplained appearance of the escaño torniño (line 3115), and perhaps -depending on the view one takes of the incident- the failure to reimburse Raquel and Vidas5. Lord explains that an oral poet draws continually on his stock of 'themes' (e. g. assemblies, weddings, rescues) and formulas: «In a traditional poem, therefore, there is a pull in two directions: one is toward the song being sung and the other is toward the previous uses of the same theme. The result is that characteristic of oral poetry which literary scholars have found hardest to understand and to accept, namely, an occasional inconsistency, the famous nod of a Homer» (94). The inconsistencies of CMC are, however, fewer and less serious than those of the Yugoslav poems cited by Lord. This might perhaps suggest that, while the Per Abbat manuscript seems fairly close to a dictated oral text, its ancestry may not be wholly oral.

CMC is not the only poem that can be fruitfully approached in this way. Our opportunities for observation are, of course, limited by the fact that most Spanish epics have survived only as prosified fragments in chronicles, so that little or nothing can be learned about the regularity of their verse, and the identifiable fragments are not often large enough to help us much in a study of formulas and themes. Menéndez Pidal has, however, been able to reconstitute substantial parts of the Siete Infantes de Lara6, and here, and even more with the fragment of Roncesvalles7 and the Mocedades de Rodrigo8, some conclusions are possible. Even if we hesitate to accept Lord's view (218) that the mixed blood of the mediaeval Greek epic hero Digenis Akritas reflects the semi-divine origin of such heroes as Gilgamesh, and hesitate even more to transfer this concept to Mudarra in the Siete Infantes de Lara (a poem not mentioned by Lord), we may nevertheless feel that the similarities between the Siete Infantes and Roncesvalles9 are easier to understand in the light of Lord's findings:

Since the singer hears many songs, he uses the language and formulas that belong to them all; for the accomplished singer whom he has been imitating does not have one set of expressions for one song and another for another, except when there are themes in the one that are not in the other, and even in these cases the formulas and formulaic techniques are the same in all songs.


(24)                


Again, what is true within a single singer's repertory is probably true within a tradition: striking themes and formulas will be transmitted and used in new contexts

The last and worst of the extant mediaeval Spanish epics, the late fourteenth-century Mocedades de Rodrigo, provides an even better illustration than does CMC of «the singer's difficulties in making normal verses when he is deprived of singing» (Lord, 126) -in other words, when he is dictating to a scribe. Lord continues:

A mixture of prose and verse, parts of verses interspersed with parts of prose sentences and vice versa, are the result. This is true especially at the beginning of the song, but even when the singer has accustomed himself to reciting, the number of lines that are irregular or poorly formed rhythmically and formulaically still remains high.


(126)10                


We cannot tell whether special difficulties of this son were experienced at the beginning of CMC or Roncesvalles, since the beginnings of both poems have been lost, but we are fortunately able to test Lord's statement against the extant text of MR, where the beginning is in prose, and moreover prose that only gradually acquires traces of assonance. The verse, when it comes, is markedly irregular.

Consideration of MR, at first sight a simple case of the dictated oral text, leads ultimately to substantial difficulties. Lord qualifies his remarks about irregularity in dictated texts:

In the hands of a good singer and competent scribe this method produces a longer and technically better text than actual performance... we can differentiate between those [dictated texts] skilfully and those ineptly done. The first will have regular lines and fullness of telling. The second will have many irregularities in line and the general structure will be apocopated.


(149)                


MR fits comfortably into this second category: apart from the metrical irregularity, there are considerable gaps, parts of lines are sometimes omitted, and the end of the poem has been lost11. One might expect to find, on the basis of Lord's comments, that an ineptly dictated oral text would also be artistically weak in other ways, and there is a prima facie case for applying here Harvey's remark, in a different context, that a text that «is extremely defective metrically, is then extremely defective in all other aspects»12. Harvey is referring to the possibility of a corrupt manuscript tradition, but the state of the extant MR would support the application of his views to the ineptly dictated text. MR is a bad poem, bad in structure, style and characterization. It is an interestingly bad poem, but the important point here is that its artistic quality matches its fragmentation and its metrical irregularity. This makes it rather disturbing when we find no such matching in CMC (nor, possibly, in Roncesvalles, though it is hazardous to form such judgements on the basis of a hundred-line fragment). Whether or not one returns to the neglected theory of Alfred Coester, that the Per Abbat manuscript is a compressed version13, there is still a flagrant contradiction between the marked metrical irregularity and the equally marked skill of the poet in style, structure and character. Effective literary criticism of CMC is only a recent development, but enough work has already been done to show that what we have here is in no sense an artistically defective poem14.

There appear, if this really is a dictated text, to be only two ways out of this impasse: either there is no necessary connexion between defective metre (and structural apocopation?) on the one hand, and artistic failure on the other; or else there is a fundamental irregularity in early Spanish oral verse that is merely intensified by the process of dictation, and that is inherited by some works in cuaderna vía. In favour of the second solution are two of the best-known pieces of mediaeval Spanish literary criticism. The metrical regularity of the Libro de Alexandre is stressed in the poem's second stanza:


fablar curso rimado          por la quaderna uia
a sillauas cuntadas          ca es grant maestria15


and one does not have to believe in the traditional clerecía /juglaría dichotomy to see that this regularity is one quality that the Alexandre poet does not ascribe to popular poetry, which we may assume to have been orally diffused. Two centuries later, the Marqués de Santillana tells Dom Pedro of Portugal that

Infimos son aquellos que syn ningund orden, regla nin cuento fazen estos romances e cantares de que las gentes de baxa e servil condición se alegran16.


Santillana's contempt for popular poetry is, as Mrs Florence Street has shown17, an illusion, but it is clear that he too regards irregularity as one of the distinguishing features of this kind of verse. Another point to be remembered is that French epic texts do not exhibit the consistent irregularity that we find in Spain. Nevertheless, both solutions are open to serious objections, and having stated the problem I must at once confess that I do not know the answer.

Nor is this the only difficulty involved. Menéndez Pidal argues for written transmission of the text of CMC (3a ed., Madrid 1954, I, 28-33). At least one of his arguments was always of doubtful validity, and has been made untenable by Lord's discoveries: this is that the apparent prosification of a different and longer version in the Primera Crónica General shows that the version in Per Abbat's manuscript was no longer being sung by the juglares, and must by the end of the thirteenth century have survived in writing only. The protean nature of oral verse makes it impossible to say with certainty that one version was replaced by another and died out. Other points made by Menéndez Pidal in favour of written transmission are much harder to refute, though they can perhaps be reconciled with the idea of a dictated oral text: these include the omission and displacement of single lines18. Finally, some features can, as Menéndez Pidal points out (I, 29) come only from a written textual source: the misreadings deyna for Denya (1161) and atineça for Atiença (2691) could hardly occur except through faulty resolution of a scribal contraction. Thus Per Abbat's manuscript seems, because of metrical irregularity and perhaps apocopation of structure, to derive from dictation of an oral poem, while some textual features strongly suggest that at least its immediate ancestor was a written text. There is nothing improbable in such a compromise solution: once a poem, whatever its origin, is orally diffused, it will begin to take on the features of oral poetry; similarly, once an oral poem is transmitted through a series of MSS., it will increasingly acquire the characteristic marks of MS. transmission. This second point is illustrated by Lord's study of sample passages from the manuscripts of Digenis Akritas (212-17), which leads him to the conclusion that the Escorialensis MS. is either a dictated oral text or very close to one, that the Oxford MS. is a fully literary production of the late seventeenth century, and that the other three poetic texts fall in between, with the Athens (Andros) and Trebizond MSS. nearer to a dictated text, and the Grottaferrata MS. further away19. It will not, of course, be possible to decide how far Per Abbat's text is from its dictated source until Lordian tests of formulas, themes and enjambement have been applied to it; and before that is done, it will probably be necessary to try out the tests themselves in a Spanish context20. The longer the period of written transmission intervening between the juglar's dictation and Per Abbat, the greater the number of scribal errors is likely to be, and the less conservative becomes the editing required (cf. Harvey, 140-41).

Yet another complication -and here MR points in the same direction as CMC- is the question of learned influence. While the mere mention of documents in these poems cannot be regarded as evidence of learned authorship now that Lord has published Yugoslav oral poems by illiterate singers that nevertheless refer to letters and seals21, the detailed and accurate knowledge of legal and chancery practice shown in CMC is another matter, and must inevitably point to some kind of learned influence22. Further, it has recently been shown that we cannot entirely rule out a connexion between CMC and the cult of the Cid's tomb at Cárdena23. MR does not show detailed legal knowledge, but it attaches great importance to fueros and privilegios (lines 63, 75, 168, 173, 186-87, 193, 203, 281-82, 292, 778), and here, as in CMC 23-24, documents are sometimes referred to in a way reminiscent of mediaeval notarial practice24. Moreover, the poet of MR shows a close interest in the history of the diocese of Palencia, and a strong desire to associate the Cid with the diocese. Study of the text and investigations in Palencia suggest that, in its present form, MR was composed in order to bolster up the waning prestige of Palencia at a critical moment in the history of the diocese.

Thus in the two poems that survive in reasonably lengthy, and apparently dictated, texts, there are clear signs of learned influence or ecclesiastical propaganda (there are no such signs in the Roncesvalles fragment), and one of the poems seems, after it was dictated, to have been transmitted through an unknown number of manuscripts. This almost inextricable mixture of «popular» and «learned» elements is startling to anyone accustomed to Spanish discussions of the epic, but not to readers of Lord, who says (20) that the oral poet can belong to any social group -he neither excludes nor specifically includes clerics-, and who says at one point that the oral poet can even be literate25. Lord also tells us that the oral poet can start from a printed text and treat it as though it were another singer's performance, though such use of a text is uncommon26. It is thus not inconceivable that in twelfth- to fourteenth-century Spain a poem, learned or partly learned in origin, could have been handed to juglares for diffusion and re-creation at each performance. If, indeed, such a poem were aimed at a wide audience, diffusion by juglares would be the obvious method. This is, of course, mere speculation, but it can at least be claimed that such a suggestion fits the ascertainable facts about CMC and MR, and is also consistent with the facts of modern Yugoslav oral composition. Whether it has any basis in reality cannot be known until the Spanish poems have been analysed in the ways suggested by Lord 130-31, if indeed it can be known even then.





 
Indice