Divas and Scholars

by Philip Gossett

  By mid-November, Verdi reluctantly informed Somma that the Neapolitan censors would not accept Gustavo III: there could be no Swedish court, no King, no firearms, no political conspiracy. Instead, he and Somma moved the action from Sweden to the Swedish province of Pomerania, from Stockholm to Stettin. the King of Sweden became the Duke of Pomerania, the Duke of Ankastrom became Count Renato, and so on. Composer and versifier worked together during Somma’s visit to Verdi’s home over Christmas; by Somma’s departure, around 27 December, the libretto of the now-retitled Una vendetta in dominò was finished.

  What about the music? Behind schedule, the composer decided to remain at Sant’Agata until he had completed “everything that requires imagination.” 19 On 9 January he wrote to Naples, “The opera is done and even here I am working on the full score; I hope to depart on board the Sicilian Courier, and as soon as I land in Naples I will be able to deliver music to the copyist.”20 It is uncertain when Verdi began sketching again, since only fragmentary sketches survive for Amelia’s aria at the beginning of the second act and the following duet and trio, but once the continuity draft resumes with the finale of act 2, Verdi is setting Una vendetta in dominò. He makes no mistakes. There is no hint of a “King” or a Swedish royal court. Gustavo is always “Duke” and Renato “Count.” Verdi’s work could not have begun before Somma arrived at Sant’Agata a few days before Christmas and was surely completed by the time the composer left Busseto on 5 January.21

  A composer’s continuity drafts are private documents. From them no one, not even singers, could rehearse or perform. That is the significance of Verdi’s words, “even here I am working on the full score [...] and as soon as I land in Naples I will be able to deliver music to the copyist.” As we saw in chapter 2, Verdi prepared a full score in two stages. First, working from the continuity draft, he prepared a skeleton score, entering all the vocal lines, with important instrumental solos and the instrumental bass. In a second stage, he completed the orchestration.

  Preparation of a skeleton score was urgent: only from this document could copyists derive the vocal parts. While rehearsals with singers proceeded, Verdi would then complete the orchestration. It is unlikely that he waited until the entire continuity draft was finished before beginning his skeleton score. We know from other operas that these stages overlapped. Creative artists divide their time between moments that depend on inspiration, when a work takes shape, and the arduous task of taming that inspiration into a practical document. However, Verdi did not begin preparing the skeleton score for Una vendetta in dominò before composer and versifier decided on the Pomeranian setting, and the work must have been far advanced when Verdi left for Naples.

  Although the complete autograph manuscript of what we know as Un ballo in maschera has been in the Ricordi archives since 1859 and has been consulted by many, no one until the beginning of the twenty-first century had fully appreciated that 75 percent of that manuscript is, in fact, the skeleton score of Una vendetta in dominò, with modifications and orchestration for Ballo introduced directly on the same pages. When he began work on Ballo in September 1858, nine months after having “completed” the earlier opera, Verdi modified the names, scraped away some words and music, crossed out superseded passages, manipulated tonalities he had decided to change, and completed the orchestration. By reading beneath these superimposed alterations, one can discover what was there in January 1858. Here are two examples from the second scene of the first act, all but two folios of which were prepared for Una vendetta in dominò and transformed on the same pages into Un ballo in maschera.

  1. Ulrica originally sang the words “le chiavi del futuro” (the keys to the future), as in all sources for Gustavo III and Una vendetta in dominò; for Ballo, Verdi replaced “the keys” with “la face” (the torch). As Somma later wrote, “And they don’t want Satan to have in hand the keys to the future, but the books of the future and who knows what” (CVS 90, p. 278). In Rome, only St. Peter is permitted to have keys.

  2. The final choral hymn, for which there is no text in the continuity draft, was originally entered into the skeleton score with the text of Una vendetta in dominò: “Gloria a te, grande e pio, / Che come volle Iddio / Le nostre sorte moderi: / Gloria e salute a te.” (Glory to you, great and pious one, who guides our fortune, as God wills: glory and long life to you). For Ballo, Verdi subsequently crossed out this text celebrating the divine right of kings, substituting “American” verses, which address the tenor: “O figlio d’Inghilterra / Amor di questa terra” (O son of England, beloved of this land).

  Almost 80 percent of the manuscript pages of the concluding festa da ballo, which fills forty-six folios of the composer’s manuscript, show the same kind of modifications as in in these two examples from the second scene of act 1. Verdi, in short, completed a skeleton score of Una vendetta in dominò in preparation for what he expected to be its Neapolitan premiere.

  Although Orsini’s bombs failed to assassinate Napoleon III, the explosions resonated in Naples. As a result, instead of rehearsing Una vendetta in dominò and orchestrating the score, Verdi engaged in complex negotiations about the text of the opera, about his legal responsibilities, about compromises that would salvage honor and purses. With his “completed” (but not fully orchestrated) opera intact, the composer sent a libretto of Gustavo III to the Roman impresario at the beginning of March 1858. This libretto was essentially identical to the one he had set to music, that is Una vendetta in dominò, but with the original setting and names restored. There could be no clearer indication that he considered this text to be the proper one for his opera. When the Roman censors, too, insisted on elaborate modifications, Verdi set aside Gustavo III, returning to it only after he acknowledged receipt from Somma of the libretto for Un ballo in maschera, on 11 September 1858.

  And what had the composer been doing while negotiations with the Roman theaters and with Somma proceeded? As far as we can tell, nothing. It made no sense to return to the score until matters pertaining to its libretto were resolved. As Giuseppina Strepponi said in a letter of 12 July 1858, written from the baths at Tabiano, to their Neapolitan friend, Cesarino De Sanctis, “Verdi is very busy... doing nothing. His most demanding activity at present is to get dressed and undressed, to eat and sleep, so that his stomach is beginning to get round, to his great satisfaction.”22 That idleness would soon be coming to an end.

  From the time he received the libretto of Un ballo in maschera in Busseto until his arrival in Rome on 10 January 1859, Verdi was completing and orchestrating the opera. There is relatively little information about the chronology of his work. Most was accomplished in Naples, where he remained after mounting Simon Boccanegra on 30 November. With the skeleton score of Una vendetta in dominò before him, Verdi had several tasks to accomplish:

  He had to transform the setting and character names, substituting “Riccardo,” “Silvano,” “Samuel,” and “Tom” for “Gustavo,” “Cristiano,” “Ermanno,” and “Manuel.” He had to emend titles: Riccardo is a “Count,” not a “Duke”; Renato loses his title of “Count.”

  He had to modify text throughout the score to agree with the libretto of Ballo, not that of Una vendetta in dominò. This meant eliminating Somma’s earlier text, normally by scraping away or crossing out earlier words and inserting new ones. In some cases Verdi also made changes in the melodic lines.

  He had to complete the orchestration. It is impossible to tell how much of the orchestration was written down when Verdi worked out the skeleton score of Una vendetta in dominò during the winter of 1857–58, but it is unlikely to have been extensive.

  He took the opportunity, having thought further about the opera, to make aesthetically motivated musical changes.

  For several important parts of the opera Verdi did away with the earlier skeleton score of Una vendetta in dominò and started anew. This is particularly evident when he entered names and vocal text that did not exist before September 1858. Here are two examples:

  After
receiving the libretto of Ballo in September 1858, Verdi removed the first part of the original introduction, which includes the instrumental prelude and introductory choral scene, and completely rewrote it on new paper. Riccardo’s name is entered directly (rather than replacing “Gustavo”), as are the new words. But Verdi would not have rewritten the gathering just to fix the text. A comparison of the continuity draft and the autograph manuscript of Ballo shows that he made significant musical changes in the melodic lines and in the structure of the prelude.

  The very last gathering of the opera was written anew for Ballo. The names and American verses are entered without a previous layer. But there were other differences between the continuity draft and this manuscript. Most striking, the entire passage after the choral outburst “Ah! morte... infamia” (Ah! death... infamy) and the beginning of Riccardo’s final Andante, “Ella è pura” (She is pure), was originally drafted as a continuation of the 2/4 meter of the choral music. The idea of a final reprise of the offstage dance music, after the assassination, came to Verdi only as he worked on Un ballo in maschera, that is, after 11 September 1858.

  It is easiest to identify the new pages when they involve American verses or modified names, but there are other possibilities. Verdi originally wrote a continuity draft for what would become Renato’s scena ed aria in act 3, for example, using a largely different melody (its first words are “E sei tu,” And it is you). From that draft he must have prepared a skeleton score by January 1858, one that no longer survives. Later, after 11 September 1858, Verdi made a second, independent sketch for this scena ed aria, corresponding to the famous “Eri tu” (It was you) from Ballo. We know this because two phrases from “E sei tu” associated with the Swedish court had become problematic by the fall of 1858. In the scena, Renato wants revenge against “altro e più nobil sangue” (other, more noble blood), while in the aria he refers to himself as “vassallo tuo primo” (your principal vassal). For the New World setting of Ballo, both phrases were modified: the first became “altro, ben altro sangue” (other, very different blood) and the second “l’amico tuo primo” (your closest friend). The revised continuity draft has the revised text, as do the new pages in the autograph manuscript. Thus, “Eri tu” must have been composed between September 1858 and January 1859.

  While working on Un ballo in maschera, then, Verdi introduced such extensive changes that he felt obliged to replace 25 percent of the original skeleton score pages of Una vendetta in dominò (some 80 to 85 folios of a manuscript of 329 folios). On these added pages, Riccardo, Renato, Sam, and Tom are the only names found; no original Swedish verses are peeking out from under the “American” ones; and references to the political situation in Europe are gone. As far as we know, the rejected 25 percent of the skeleton score of Una vendetta in dominò no longer exists.23

  Reconstructing Gustavo III for Gothenburg

  Anders Wiklund, a genial Swedish musicologist and conductor, has collaborated closely in both the critical editions of the works of Rossini (for which he edited La scala di seta and, in collaboration with Patricia Brauner, La pietra del paragone) and those of Donizetti (Maria Stuarda and, in collaboration with Roger Parker, Le convenienze ed inconvenienze teatrali). After a trip I made to Copenhagen for the Verdi Festival in that city at the end of May 2001—where I admired the Calixto Bieto staging of Un ballo in maschera in Fascist Spain24—I visited Wiklund at his home in nearby Gothenburg and saw a fascinating new opera by Jan Sandström, Macbeth2 (Macbeth squared), at the Gothenburg opera, where Wiklund works as an adviser. The production showed off beautifully the remarkable scenic capabilities of the new theater and its first-rate orchestra.

  While taking a long walk through one of the parks on the outskirts of the city, we began to talk about Ballo. I related the history to Wiklund and shared my thought that it might be possible to reconstruct Gustavo III. “And we can perform it in Gothenburg,” he immediately added. From a hypothetical “might” to a performance less than sixteen months later was hardly what I had in mind, but he was absolutely serious. In September, Wiklund and the director of the theater, Kjell Ingebretsen, himself a conductor, flew from Sweden to Parma to see the production of Verdi’s revised Macbeth (1865) that concluded the 2001 Verdi Festival, and was the first to use the new critical edition prepared by David Lawton. Before the evening was over, Ilaria Narici, editor of Ballo, and I had agreed to undertake the reconstruction of Gustavo III for the September opening of the 2002–3 season of the Gothenburg Opera.

  Three-quarters of an opera is not an opera, but we had available to us not only three-quarters of Verdi’s skeleton score for Una vendetta in dominò, but also the continuity draft from which he had prepared that skeleton score. For the 75 percent of the skeleton score of Vendetta that still exists, embedded in the autograph of Un ballo in maschera, the versions of the continuity draft and the skeleton score are amazingly similar: the differences are in details or in the addition of brief orchestral conclusions to closed numbers (Verdi often stopped sketching a piece with the last notes of the vocal line). For the 25 percent of the skeleton score that no longer exists, the problem often involved a transposition, so that there were only minimal musical changes. One can differentiate two situations:

  Passages where neither an original continuity draft nor an original skeleton score exists. This involves primarily the duet for Amelia and Gustavo/ Riccardo in the second act.25 All 20 folios of this duet in the autograph manuscript were added after 11 September 1858. While it is possible to modify the words of Ballo in accordance with the libretto of Gustavo III, we have no certain knowledge about the original music.26 Still, the changes that can be introduced in the text are dramaturgically significant. Whereas Riccardo promises Amelia the universe, “La mia vita... l’universo, l’universo per un detto...” (My life... the universe, the universe for a word...), Gustavo offers her something more tangible, “Il mio nome e la corona, la corona per un detto...” (My name and the crown, the crown for a word...).

  Passages where only a continuity draft exists, but where the music underwent minimal changes for “Un ballo in maschera.” The canzone for Oscar in the first scene began with the words “Pallida, pallida, / Volta alle stelle” (Pallid, pallid, turned toward the stars) in Gustavo III, but with “Volta la terrea / Fronte alle stelle” (Her earth-colored face turned toward the stars) in Un ballo in maschera.27 This piece was originally entered into the skeleton score of Una vendetta in dominò / Gustavo III in B major; for Un ballo in maschera Verdi transposed it down to B major. Musical differences between the two versions are minimal. Thus, there is every reason to believe that the original skeleton score (nine missing folios) was essentially the same as the music we know from Ballo, but written a half tone lower.

  For a good half of the missing 25 percent of the original skeleton score, then, the music of Gustavo III was probably the same as Un ballo in maschera, but in a different key.

  In several places, though, the versions of the continuity draft for Una vendetta in dominò / Gustavo III, on the one hand, and Ballo, on the other, are strikingly different. There are important changes in significant melodies (such as the opening chorus), structural differences (the entire scena for Amelia that opens the second act, through the beginning of her aria), heavily modified passages (the end of Gustavo/Riccardo’s romanza in the last act), or even largely new pieces (Ankastrom/Renato’s aria in the last act). But wherever there is a hint remaining in the autograph manuscript of Ballo of what was once there for Vendetta, that hint is very close to the continuity draft. Here are three examples:

  1. THE SCENA PRECEDING AMELIA’S ARIA AT THE BEGINNING OF ACT II. The last measures of the original version of Amelia’s scena are still visible in the autograph manuscript of Ballo: never orchestrated, Verdi crossed them out as he worked on Ballo.28 These surviving measures, however, are basically the same as those in the continuity draft. All preceding pages for Vendetta were removed by Verdi as he prepared Ballo. Given the way Verdi constructed his manuscripts phys
ically, however, we know that the passage he removed was about half as long as the music he added. Indeed, the version found in Verdi’s continuity draft is half as long as the music he introduced for Ballo.29 The text of this draft is basically what Somma sent to Verdi on 5 November 1857 (CVS 56, p. 211), and this same text occurs in the libretto of Una vendetta in dominò that Verdi prepared in February 1858 to press his case in the legal proceedings, as well as in the libretto of Gustavo III he sent to Rome in March 1858. There is every reason to assume that the music of the scena currently found in the autograph manuscript of Un ballo in maschera, with its partially different text, was prepared for Ballo, after 11 September 1858.

  The original passage is fascinating, because the music explicitly refers back to the orchestral introduction to the second scene of the opera, in Ulrica’s den. The opening sonority and the ensuing orchestral figuration that play such an important part in that scene also dominate Amelia’s scena. But the continuity draft of Amelia’s scena (and presumably its transformation into a skeleton score) lacks what is the most striking melodic reference here in Ballo. the return of the lyrical theme that Amelia sang during her trio with Ulrica and Gustavo, to the text “Concedimi, o Signore, virtù ch’io lavi il core” (Grant me, Lord, virtue to cleanse my heart). This is the passage which, in his autograph manuscript of Ballo, Verdi associates with the stage directions “Appare Amelia dalle eminenze” (Amelia appears from the hills) and “A questa ripresa Amelia s’inginocchia e prega” (At this reprise Amelia kneels and prays). There is no such respite for Amelia in Una vendetta in dominò until her aria: the scena itself is stark and terrifying from the first note to the last.