To one little change, however, Verdi appears to have assented. When the Duke enters the inn of Sparafucile in the last act, he says (in all secondary sources of the opera, manuscripts and printed editions alike):
These verses are different from what Verdi had written both in his sketch and in his autograph:
We know, of course, that the Duke has been enticed to the inn by Maddalena, so that these lines offer no new information. Yet the bald statement, taken directly from Hugo, was too much for the Venetian censors. In Verdi’s autograph “tua sorella” is dutifully crossed out and “una stanza” added, not in Verdi’s hand. So it had been performed ever since, even when the Duke all but raped the girl on stage, until the critical edition made the original text available in the early 1980s.55
Transforming L’Italiana: Structural Changes
During the first half of the nineteenth century composers were often asked to direct one of their older operas for a new city, at a new theater, in a new production, and with new singers. In the process a composer sometimes modified his score. To make sense of nineteenth-century sources and to make intelligent decisions for modern performances, one must understand these alterations and the motives behind them. Sometimes composers felt that sections of their original were weak, musically or dramaturgically; occasionally there were sections that had been prepared by other musicians. Given the opportunity to intervene, composers did so.56 Thus, for a revival of L’Italiana in Algeri at the Teatro Re of Milan in the spring of 1814, Rossini replaced a solo for Lindoro in the second act (“Ah come il cor di giubilo”), probably prepared by an associate, with an aria of his own.
Other revisions favored particular singers, who, because of their vocal range or other characteristics of their voices, could not negotiate the original versions. More insidious vanities were not infrequent: singers felt their parts were too small or that their solos did not adequately display their gifts. In a world where the prima donna was paid considerably more than the composer (a world that has changed little today), operas were at the mercy of such caprice. To suit the talents of Maria Marcolini, an Isabella who may have been uncomfortable with the sexual innuendos of “Cruda sorte” discussed above, Rossini prepared the more heroic “Cimentando i venti e l’onde” for a performance in Vicenza later in 1813.
Important revisions were frequently tied to performances in a different cultural environment. In Restoration Naples, Isabella’s “Pensa alla patria” was impossible, and Rossini replaced it with an aria that avoids all mention of patriotism, “Sullo stil de’ viaggiatori.” Adaptations often took place when an Italian opera was performed in France or Vienna; similar changes were made when a French opera was performed in Italy. Audiences, even in major European capitals, were notoriously provincial, and a composer was expected to bend his talents to local custom. Coming to grips with multiple versions of an opera, all prepared by its original composer, is one of the most difficult aspects of performing Italian opera.57
None of these problems disappeared when the composer himself was absent: weaknesses in the score were equally palpable; singers did not become more timid in their demands or more flexible in their throats; and if a theater felt that alterations in a score were necessary for its local audience, it had no compunction about intervening. Unauthorized rearrangements of opera were commonplace in the nineteenth century, although some theaters began to develop a sense of responsibility, identifying in printed librettos those compositions added to a score. There are many documents, letters, and reviews that testify to a growing appreciation of the integrity of a composer’s work, but this idea made its way slowly during the first half of the century.
The Martorelli manuscript of L’Italiana in Algeri, “Rossini 22–1–61,” has only a small problem of this kind. In its main body one piece from the Venice 1813 score is missing, Haly’s aria, “Le femmine d’Italia.” While a lovely piece, it is not by Rossini: this is one of the compositions he assigned to an associate. Was it originally missing from the Martorelli manuscript and did Haly object to a part that had no solo? Or was it originally present and did Haly find “Le femmine d’Italia” not right for his voice? In any case, a new aria was prepared, “Ad amare un vago oggetto,” and it was included as an appendix to the manuscript. While there is no reason to assume the music is by Rossini, in at least one series of performances it became part of his score. Its significance in the textual history of L’Italiana in Algeri, however, was marginal.
Transforming L’Italiana: The Orchestration
The worst transformations visited upon L’Italiana in Algeri, however, infected the very texture of the score, and the Martorelli manuscript, “Rossini 22–1–61,” gives eloquent witness to the process by which the disease took root. Rossini’s use of the orchestra was both highly personal and characteristic of his time, a period of significant changes in orchestral technique.58 In his first operas, even those for major opera houses, Rossini made no use of trombones. In earlier Italian opera these instruments were generally reserved for special effects: oracles, scenes of hell and damnation, stone guests. Only gradually during the 1810s did their use become widespread in Italian opera, particularly in opera seria. Rossini employed a single trombone in Il Turco in Italia (Milan, La Scala, 14 August 1814), but only in Naples did he introduce three trombones, which soon became the norm. Occasionally he added a fourth low brass instrument, the “serpentone,” more akin to the Verdian “cimbasso” than to the modern tuba, which is not the sound Rossini (or Verdi) envisioned.59 For no work did he ever employ more than two bassoons. And he had a particular inclination for using the piccolo in soloistic contexts and in special orchestral combinations. The overture to L’Italiana in Algeri, where a theme is played by bassoon and piccolo at a distance of three octaves, is a lovely example (see mm. 184–192).
By midcentury, feelings about orchestration were quite different. Every opera employed three trombones and cimbasso (the standard scoring in Verdi). No one would have dreamed of writing a serious opera with only two horns, as in the orchestration of Rossini’s Otello (except for its sinfonia). Four bassoons were sometimes available (Meyerbeer had employed two bassoons and a contrabassoon in his Italian opera of 1824, Il crociato in Egitto.)60 Piccolos were no longer used soloistically. As a result, the few Rossini operas regularly performed during the second half of the century were heavily revised by musicians of the time anxious to bring them into line with orchestral values of that period. These late revisions were among the very few elements of performing practice transmitted through printed scores.
In the Martorelli manuscript, Rossini’s original orchestration is heavily edited by later hands: three trombones are supplied for most ensembles; to compositions without trumpets, two trumpet parts are added; passages written for piccolo are assigned to flute; additional flute and bassoon lines are invented. These modifications were at first made directly on the original score or provided as separate fascicles, so that one could see where Rossini’s notation ended and where the reviser’s work began. When the modifications became too numerous, however, entire numbers or sections were copied anew with the revised orchestration. For anyone picking up the newly prepared score, then, these passages were Rossini’s opera, and there was no way to differentiate between what he had written and what was imposed on his score more than fifty years later.
A score of L’Italiana in Algeri manipulated in a similar way was to be the basis of the rental score and parts prepared by Ricordi late in the nineteenth century and provided to opera houses until about 1973, when Azio Corghi’s critical edition of the opera, later published by the Fondazione Rossini, finally restored the opera to its correct orchestral format. The delicacy, lightness, and precision of Rossini’s orchestration was sacrificed to a late nineteenth-century vision of orchestral sonority and was then sanctified by ignorant twentieth-century musicians as belonging to the “tradition,” a “tradition” invented by musicians with no sense of Rossini’s orchestration and totally extraneous to the opera that d
elighted all Europe during Rossini’s lifetime. A similar fate greeted La Cenerentola, where the trombones, instrumental doublings, and thumping bass drum and cymbals that pepper the score in many twentieth-century performances are totally inauthentic. But some modern opera houses, concerned about being charged rental fees to use performing materials for the new editions, continue to opt for these deficient scores, as we saw in chapter 1, thereby compelling themselves to hire extra orchestral musicians to play instrumental parts that Rossini never wrote and that denature his orchestration.
ENTER GIOVANNI RICORDI
Rossini’s operas suffered much worse from this system of transmission, leading—as it invariably did—to freewheeling and unprincipled interventions, than did those of Bellini and Donizetti; their operas in turn suffered worse than those of Verdi. The key figure in this cultural transformation was a music copyist turned publisher, Giovanni Ricordi, a genius and positive force in the history of Italian opera, despite occasional misdeeds. The company he founded, still Italy’s largest music publisher and one of the most significant in the world, although now linked to the multinational BMG, supported and fostered much of the development of Italian opera in the nineteenth century and into our own time. Giovanni Ricordi and his sons and their sons were friends and associates of Rossini, Bellini, Donizetti, Verdi, and Puccini, creating a vast publishing empire. Their approach blended commercial acuity with a recognition of the rights—artistic and pecuniary—that should accompany a composer’s creative work.61
The history of copyright legislation, particularly international copyright, is complex. The early nineteenth century was a period of confused and contradictory laws. Some countries, such as England and France, had strict legislation, but it applied only to works first published within their borders. In the geographical reality and political fiction that was Italy, it was more difficult to rationalize the system.62 Nor should we imagine that effective copyright legislation solves all problems. It is easy to favor laws that guarantee artists the right to appropriate financial recompense for their labor and a way to maintain the integrity of their work. Yet copyright has its insidious side. When Stravinsky decided to reorchestrate Petrushka in the 1940s, he was seeking in part “to safeguard his copyright position.”63 Whether there were always artistic gains in this reorchestration is a matter for debate: the original work has a coherence perhaps compromised in the stylistically more eclectic revision.
But we have already considered the alternative. The failures of the system of distribution and the inadequacy of protection given to composers in early nineteenth-century Italy were evident, and Giovanni Ricordi shrewdly turned these problems to his advantage. In a series of contracts Ricordi signed with Milanese theaters one can follow the birth of an idea, an inspiration. He parlayed a copisteria into an archive, an archive into a publishing house, a publishing house into a quasi monopoly on the theatrical production of Italian composers.64
There are strokes of fortune that come only to those prepared to recognize them. The Teatro alla Scala had an archive in which were deposited scores prepared for the theater (autograph manuscripts or manuscript copies) and derivative materials. As the copyist attached to the theater, Ricordi was responsible for preparing parts for performances and providing complete manuscripts where needed. Although his position allowed him the right to profit from scores commissioned by the theater, precise limits to his activities were defined. Gradually he expanded those limits. At first he could make manuscript copies of extracts alone (arias, duets, but not concerted numbers—ensembles, introductions, and finales), which he could sell to dilettantes who sought to play and sing the most successful numbers from the latest opera in the privacy of their homes. The orchestral parts and manuscripts he was asked to produce, on the other hand, were jealously guarded by the theater. Then Ricordi found a quicker, more satisfactory means of providing copies of morceaux favoris: to engrave the music on copper plates and print multiple copies. Here too his rights were limited: he could print single numbers but not complete scores, for the prohibition against concerted numbers remained in force. By withholding these pieces, after all, the theater sought to maintain control over subsequent performances of the work, and to make it more difficult for an unscrupulous musician to reorchestrate an opera from a vocal score and sell it to theaters elsewhere on the peninsula.
No such restrictions prevented French or German publishers from making complete vocal scores of favorite operas, however, and Ricordi was painfully aware that beautiful editions began crossing the Alps in the late 1810s and early 1820s. No copyright problems existed, since the works had been first performed in Italy and hence were unprotected under French and German law. Publishers unmercifully pirated each other’s work. Shortly after the premiere of Rossini’s La donna del lago, for example, Ricordi published a few extracts. These made their way to Leipzig, where the German publisher Breitkopf & Härtel found them so attractive that the firm decided to bring out a complete score of the opera. For this purpose they used all available Ricordi extracts and a complete manuscript copy. They simply followed the Ricordi piano reductions for numbers the Milanese publisher had issued as extracts; for the other numbers and recitatives not previously published by Ricordi they had new reductions made. In this way Breitkopf & Härtel cobbled together one of the earliest editions of La donna del lago. Ricordi gained a modest revenge several years later. Finally relieved of contractual impediments to publishing complete operas, the company not only made use of the new Breitkopf reductions for its own complete edition but even borrowed ornamental designs that had first appeared on the title page of the Breitkopf score.65
We do not know just how Giovanni Ricordi succeeded in wringing ever-greater concessions from the Milanese theaters. His copisteria was undoubtedly efficient, of course, and theaters turned to him increasingly to receive the best possible materials. At a certain point, however, he stopped being an employee of the theaters and became a private entrepreneur, from whom theaters rented materials. Most extraordinarily, the Teatro alla Scala appears to have been so pleased with this arrangement that they ceded to Ricordi the entire archive of operatic manuscripts, autographs, copies, and performing materials, related to works that had been presented at or commissioned by the theater. For a while Ricordi needed to make a new contract for each new opera performed at La Scala, but soon the passage of rights to Ricordi became automatic. By the mid 1820s, he controlled a vast archive and wielded immense power.
The key to Ricordi’s fortune was its insistence on renting full orchestral scores and instrumental and vocal materials to theaters for performance, rather than publishing these full scores. In Italy, after all, with no central authority and only regional laws, nothing could stop local theaters from making their own parts and proceeding without a thought for the composer or his publisher. Indeed, among Italian publishers only Leopoldo Ratti and Giovanni Battista Cencetti in Rome behaved differently: during the 1820s and 1830s they issued full scores of eight Rossini operas. Though typographically attractive (and therefore much in demand by collectors today), the Ratti and Cencetti publications resemble hackwork manuscripts, with articulation poorly marked, notes and rhythms inaccurate. In no case do they derive from a Rossini autograph: the publishers worked from faulty secondary sources.
Ricordi, on the other hand, knew that this route would not offer him financial or artistic success. To a growing number of composers he stressed the advantages of his not only publishing vocal scores of their operas but also representing their interests in dealing with theaters. He would assure them proper compensation for their artistic labors both when they wrote an opera (through fees from the opera house and the sale of proprietary rights to Ricordi) and for a certain period of time after the first performances (through royalties on sales of vocal scores and a percentage of fees paid by opera houses to perform the work). The interests of the composer and the publisher, in short, appeared to be the same. They were, indeed, the same in principle, but different elements of the relat
ionship had a different relative importance for the two parties: that was the zone in which serious problems were eventually to emerge.
In the meanwhile, the system was gradually put into place, and Ricordi brought into his orbit Bellini, Donizetti, Pacini, Mercadante, Verdi, and a host of lesser-known composers. These composers still sometimes made agreements with opera houses that gave the commissioning theaters complete possession and all rights to their new scores, including the right to control and profit from all future performances. More and more Italian composers, however, accepted smaller payments from the commissioning theaters in order to reserve for themselves all subsequent rights to the commercial exploitation of their own music. These, in turn, they ceded to Ricordi along with the original autograph manuscripts of the works; Ricordi, then, would represent them in all further commercial dealings. This was a crucial element of Ricordi’s success, since these autograph manuscripts guaranteed that the editor had access to a reliable source for each opera he acquired. Ricordi thus accumulated an enormous collection of autograph manuscripts, including almost every opera by Verdi and Puccini, and an impressive array of scores by earlier composers. Even when composers sold their rights to a theater, the impresarios in turn frequently sold them to Ricordi, either directly or through the mediation of a local copisteria. La Fenice, for example, wanted to purchase all rights to Rigoletto, but Verdi—always a shrewd negotiator—asked an exorbitant price. A compromise was achieved: Verdi retained the rights (which he immediately ceded to Ricordi), but in recognition of La Fenice’s special status as commissioning theater, they were permitted to have their copisteria prepare a manuscript copy to keep in the theater archives. The copy could not be made available to any other theater, but could be used as the basis for future performances at La Fenice, without additional payment. As a result of this arrangement, which proved satisfactory to Verdi, La Fenice, and Ricordi, Rigoletto joined the Ricordi stable.66
Divas and Scholars Page 15