Antonio Vivaldi: Tito Manlio RV 738

Edited by Alessandro Borin

Two-volume set: score pp. XXII, 316 / Introduction and critical commentary pp. 277 [Italian and English texts]
PR 1411
Piano vocal score available
CP 141077

[Excerpt from the Introduction]

The score of Antonio Vivaldi’s Tito Manlio is preserved, in two copies, in the Biblioteca Nazionale universitaria of Turin. The first copy (Giordano 39, fols 171–365) is a composition manuscript, while the second (Foà 37, fols 119–306) is a calligraphic copy of the foregoing. The text transmitted by Giordano 39 is almost entirely in Vivaldi’s own hand, with the sole exception of fol. 287r and the first two systems on fol. 331r, where the handwriting of “Scribe 10” is in evidence. Further, the last bar of the last system on fol. 253v has been deleted and rewritten by a third hand matching that of none of the Italian copyists of Vivaldi’s music known to us. The text transmitted by Foà 37 was the work of five of the latter (“4”, “10”, “11”, “12” and “48”), working under the direction of, and with some interventions from, the composer himself. In keeping with the practice of the time, the preparation and performance of an opera did not, however, require a complete ‘master copy’ of the score. Thus the instrumental and vocal parts were extracted directly from the autograph manuscript under the direct supervision of the composer, and even during the periods of the rehearsals and the production this manuscript, in the custody of the first harpsichordist, continued to serve as the central authority. Hence until now it has generally been believed that the text transmitted by Giordano 39 represents the composer’s “working score” and that in Foà 37 the “fair copy” used in performance.

In fact, Vivaldi continued to make changes to the autograph manuscript even after the copy had been completed. Many of these alterations were made at a rather late stage and were carried over to Foà 37 as well; others, introduced still later (perhaps after the opening performance), appear in only one or other of the two scores. If, from a given point onwards, only the score in Foà 37 were actually in use, what would be the possible reason for making such changes in the composition draft? And, above all: why were some later alterations introduced only in Giordano 39 and not in the other score?

To these and other questions a response will be attempted in the sections dedicated to the description of the processes and typologies of revision of the autograph manuscript. For the moment, it is enough to observe that both scores remained in the possession of the composer, who took them out of circulation immediately after the close of the operatic season of 1718–1719 in Mantua. This circumstance may also explain why no separately copied aria serving as a travel souvenir or destined for the thriving market of collectors and amateur performers has come down to us.

The only other authoritative source for Tito Manlio is the libretto published by the archducal printer, Alberto Pazzoni. The text contains neither virgulated passages nor cancels (fogli volanti or carticini) inserted to indicate possible additional or replacement texts. Even though the libretto and the score took shape, and were used, simultaneously, they nevertheless preserve a degree of autonomy, especially in connection with their differing function and respective end-users. The presence of “autonomous variants” in both sourcecategories excludes from the start the possibility that either was derived exclusively from the other, while the occasional discrepancies between the words printed in the libretto and those underlaid to the notes in the autograph score do not reflect a more advanced stage of revision in either, but are simply the result of imperfect communication between the composer and/or librettist and the printer.

In many instances, there is an evident polarization of the variants in Foà 37 and the libretto vis-à-vis those in Giordano 39. In the final recitative of I.01, for example, the libretto and the copy transmit the reading “seco non porte”, which in the autograph manuscript was hypercorrected to “seco non porta”. So the copyists, in contrast to the composer, may have been able from the start to consult an example of the printed libretto, or at least a printer’s copy. This is a factor that also leads one to reconsider the relative chronology of the three sources: in other words, it is possible that between the printing of the libretto and the occasion of the first performance there elapsed a period of time considerably longer than is generally thought to have been customary for that period. We know, indeed, that in order to minimize the discrepancies between the text of the drama set to music by the composer and the words actually heard by the theatre audience librettos were sent to press only just before the first performance. In this particular instance, however, the state of the sources suggests that the interval of time was somewhat longer—at least as long as that needed by the copyists to complete their own work.

Comments on the Editorial Choices

The present critical edition sets itself the aim of giving the reader the text of the opera in as close a form as possible to that which it had reached by the time of its opening performance in Mantua in 1719. This choice, which might be considered a merely pragmatic one, conforms in fact to the concept of “opera as a social phenomenon” advanced by Jerome J. Mcgann, on the relevance of which to the musicological sphere Jeffrey Kallberg, Susan Lewis Hammond and especially James Grier were the first to draw attention. According to Mcgann, every artefact is the product of a complex process of interaction between author(s), performer(s) and consumer(s) that influences not only the act of creation but also that of reception. Hence, still according to Mcgann, opera is “a social event […] embedded in dynamic, multilayered historical and ideological context”; conversely, the text appears to us as “a deanimated linguistic structure”, a document that fixes a particular stage in its process of socialization.

This is particularly evident in the case of the musical theatre, “a genre that compounds the difficulty of conceptualizing in terms of work and text the two arts, music and theatre, that by their nature manifest themselves principally in the form of an Event”. Rather than depending exclusively on the (presumed) will of an individual author, the process of socialization in opera involved a plurality of different ‘players’ (librettist, composer, performers, impresario, public and patrons), and in order for the event to take place it was equally necessary to reach an agreement, in a more or less explicit form, between them all. Indubitably, the moment of the first performance coincided with an initial stabilization of this relationship, since for an opera to reach the stage, the composer and librettist had to be satisfied with their respective work, all the singers had to have accepted their roles and been accorded an appropriate amount of rehearsal time with the orchestra, and the public and the possible patrons had to be willing to (co)finance the spectacle through a financial contribution, renting a box or buying admission tickets. Even if this point of consensus was fragile and could be renegotiated several times in the course of the same season (for instance, on the basis of the public’s reactions or the demands of the singers), from a methodological point of view an approach that takes account of the social dimension of the text is better suited to the purposes of philological enquiry than one based on the traditional concept of “authorial will”. The establishment of a critical text thus ceases to be the result of a psychological analysis and becomes instead the fulcrum of a historical investigation.

As regards the choice of a primary source for the critical edition, the autograph manuscript (source A) and the copy (source B) are on a substantially equal footing: the first transmits an authorial redaction; the second, a derived version overseen by the author and therefore of equal value. Nevertheless, since B retains all the errors in A, plus some peculiar to it introduced by the copyists and overlooked by the composer, the decision has been made to use as the ‘base’ text the score transmitted by Giordano 39. In addition to being more accurate, source A is distinguished by showing a whole series of changes made prior to the premiere that, despite their irrelevance to the aim of a achieving a critical restoration of the text, are able to shed light on the various stages of the process of its compilation and, to a certain degree, of its ‘socialization’.

Another of the more controversial aspects of making modern critical editions of operas of the early eighteenth century is that concerning the treatment of the literary text (the words underlaid to the notes and the stage directions). The divergences between the texts furnished respectively by Vivaldi in source A and by Alberto Pazzoni in source L are, as already explained, the result of a lack of communication between the composer and the printer, as well as of the different practical functions of sources A and L (with regard to both their uses and their end-users). In no instance is it legitimate to suppose that A depended on L. Since the text written into the score has a much closer relationship to the music than that printed in the libretto, in all instances where sources A and L transmit equally acceptable variants, the edition opts for the reading in the score, recording in the Critical Commentary that found in the libretto. Thus the textual underlay does not adopt the variants present in L when A transmits an acceptable version, and accepts those of L (if correct) only when A is clearly in error.

The absence from source A of stage directions (except for a precise mention of the characters who are singing) and of many particulars concerning punctuation is perfectly consistent with the practice of the age, since the composer included in the score only information strictly necessary for the performance. The libretto, in contrast, was intended to be read by the theatre-going public and possibly preserved: as such, it needed to contain a precise description of the stage sets and a poetic text presented in an orthographically correct fashion. unlike sources A and L, the modern published edition is aimed both at readers and at performers, and it is the accepted custom to include all the stage directions present in the libretto. It is the editor’s belief that this fusion of two sources has to be accepted for what it effectively is: the autograph text augmented by information omitted by the composer only because writing it into the score served no practical purpose.

Regarding the normalization of Vivaldi’s text, the underlaid words follow the modern conventions described analytically in the New Editorial Norms of the Istituto Italiano Antonio Vivaldi.

Specific forms found in source A have been retained in all instances where the process of normalization would have deprived the original text of some aspect of its pronunciation given expression in the musical setting. Hence the poetic text published in the score is in most instances an amalgam of the text transmitted by source A and (in normalized form) the spelling and punctuation of source L, with the purpose of offering the present-day reader and performer as much information as possible, but without compromising the correct understanding of information important from a musical standpoint.

List of Arias