Pitch is not celestially ordained; it is relative. Today we usually set concert a between 440 and 444 vibrations per second, but some orchestras persist in pushing the pitch still higher to obtain a brighter sound. During the nineteenth century, there were enormous pitch differences from one country to another, even from one city to another. When an effort to regulate pitch throughout Europe took shape during the 1880s, Verdi was a staunch supporter of a uniform pitch level, and would have preferred that it be set at a = 432, despite a faction of Roman musicians who wanted it to be as high as a = 450. (Verdi contemptuously commented that such an a in Rome would be a b anywhere else.)6 Finally an international commission (with the participation of Arrigo Boito) brought a modicum of uniformity to European practice by setting standard pitch at a = 435.7 Despite their formal agreements, in some countries and theaters the pitch level continued to be higher, in others lower. A soprano’s high c today is generally higher than a similar note during the first half of the nineteenth century. Asking a prima donna to sing “Casta Diva” in Norma in its original key of G major, rather than transposing it down to F major, or the mad scene in Lucia di Lammermoor in its original key (beginning in D minor and concluding in F major), rather than transposing the whole scene down a full tone, is not the same when a was set at 430 vibrations per second, as it might have been heard in Milan in 1831 or Naples in 1835, rather than in the higher tunings of today.
Only rarely do we find composer-generated transposition in Rossini’s operas (although that does not mean he would have disapproved of singers making transpositions), but the problem of transposition has broader importance when we consider the operas of Verdi, and it becomes fundamental for those of Bellini and Donizetti. As we gain access to more of Verdi’s compositional sketches, for example, we see how often he drafted compositions in higher keys, then brought them down when he prepared his autograph manuscripts.8 Before examining this problem from a practical viewpoint, however, it is worth considering a more general question: to what extent do choices of tonality in Italian opera depend upon a composer’s vision of a coherent role for tonality in a stage work?
Over the past thirty years, the use of tonality to carry dramatic and musical meaning has been one of the more bitterly contested and politically charged terrains in the study of Italian opera. Keys and key relationships are central to German and Austrian instrumental music, after all, and scholars have demonstrated how these instrumental principles also influence the structure of operas by Mozart, Beethoven, or Wagner. Some of these demonstrations, in their insistence on the organic cohesion of entire operas, push the evidence far beyond what it seems able to bear.9 It is true, for example, that Wagner, in Opera and Drama, his 1850 –51 treatise setting forth some of the precepts that would inform the composition of Der Ring der Nibelungen, developed a theory of the “musical-dramatic period” that emphasizes the unifying role tonality could play in his evolving concept of music drama. The gulf between that important insight and the systematic theory set forth by Alfred Lorenz in his monumental The Secret of Form in Richard Wagner, however, is immense.10 It is no surprise that over the past twenty years scholars and critics have tended to set aside Lorenz’s massive symmetrical designs and to seek other ways of thinking about the structure of Wagner’s operas. Sometimes, however, their rejection has gone too far. To ask, “What do we gain by saying that Tristan Act III is ‘in’ B major?” in order to promulgate other critical approaches is an arrogant denial of ordering principles that generations of composers and thinkers about music found of fundamental importance.11 No one has ever claimed that the assertion of a tonal plan for all or part of an opera exhausts that opera’s meaning.
Attempts to move discussion of Italian opera into the mainstream of scholarly discourse have invariably spawned efforts to show that keys and key relationships are also crucial to this repertory, as if such evidence of compositional planning might help ensure musicological respectability to the sunny South. Thus, while analysis and criticism of oltremontani composers were moving away from organicist tonal explanations, students of Italian opera began to embrace them.12 Controversy still surrounds many of these efforts, but a number of highly successful studies have centered on Verdi’s Rigoletto, an opera with a complex but convincing tonal underpinning.13
The prelude of Rigoletto begins with a phrase that recurs many times throughout the opera, referring to the curse that Monterone launches against Rigoletto and the Duke, the curse that works itself out with the seduction and death of Rigoletto’s own daughter, Gilda. While centered on the pitch c and leading to a cadence in C minor, the phrase moves toward and away from a complex chord that can be heard in two ways: as a dark, dissonant sonority coloring the c, a momentary detail lending dramatic weight to the theme, or (respelling the f as the enharmonically equivalent g and putting the a in the bass) as a chord functioning as the dominant of D major (or minor), which forces the tonality up a half tone to a resolution in a new key (example 10.1). These tonal poles recur throughout the opera: c seems to represent the curse or threat, d its terrible consequences. Not only does the opera begin in C minor and conclude in D, but this progression and related ones become implanted in our hearing. Early in his soliloquy (“Pari siamo”) comparing himself to the assassin, Sparafucile, Rigoletto recalls that the old man cursed him, invoking the opening motive on c, then immediately moving to d for “O uomini!... o natura!...” The jester’s furious attack on the courtiers, “Cortigiani, vil razza dannata,” begins in C minor, but the aria concludes with his plea for mercy, “Miei Signori, perdono, pietate,” in D major. Composers do not begin an aria in one key and conclude a half step higher without strong motivation. When Rigoletto weeps with his daughter at the end of the second act, they do so in D; but immediately after, when the doomed Monterone passes escorted by guards, the music is in C minor.
EXAMPLE 10.1. GIUSEPPE VERDI, RIGOLETTO, PRELUDIO (N. 1), MM. 1–3, WITH AN ALTERNATIVE INTERPRETATION OF THE HARMONY IN M. 2.
Many scholars are content to concentrate on those tonal relationships in an opera that seem central to the drama, without insisting on a scheme in which each key traversed must be thought to have profound symbolic meaning. Rossini’s Maometto II, for example, is a tragic opera that sets the fate of four characters against a background of historical events—the wars between the Turks and the Venetians, culminating in the fall of Negroponte.14 Paolo Erisso is in charge of the Venetian forces; Maometto II leads the assault of the Turks. It is a fourfold tragedy: of Anna, Erisso’s daughter, who is in love with Maometto but cannot permit herself to surrender to feelings that betray her people; of the conqueror Maometto, who cannot have the only joy he truly desires; of Erisso, who loves his daughter above all else but feels honor-bound to supply her with the dagger she will ultimately use to take her own life; and of Calbo, the young Venetian warrior, who wishes to wed Anna but will not impose his affections on a woman who loves another.
The introduction, with which the opera begins (there is no overture), is in E major; the entire opera will conclude in E major with the suicide of Anna. A casual glance at the score might suggest that Rossini was indifferent to tonal structure, since he ends the opera a semitone higher than its point of departure; but there is strong evidence that this is not the case. A central image of the opera is the tomb in which the ashes of Anna’s mother lie. When Erisso asks his daughter to wed Calbo, he invokes his wife’s tomb. Accompanied by tremolo strings alone, the phrase begins in E major, then lurches upward a semitone to E major. The last scene of the opera takes place in the church, amid the tombs. Erisso pauses before the tomb of his wife. Within a recitative passage he breaks into a lyrical phrase, which begins not in E major but in the closely related A major. As Erisso invokes the heaven where his wife now resides, the music sits on the dominant (E major) and then tortuously modulates up a semitone to the familiar E major. After Calbo’s E-major aria, Anna enters. Before her mother’s tomb, she vows to follow her father’s wishes and accept the hand of Calbo. Her
music is identical to the phrase Erisso sang earlier in the scene.
Ultimately, her vow of fidelity before the tomb of her mother and her vow to kill herself rather than be captured by the Turks are musically integrated by their similar settings and tonality. During the course of the first-act terzettone, Erisso hands her a dagger, her “inheritance” on this fatal day: she is to use it rather than fall captive to the Turks. It is a phrase declaimed first over a tremolando accompaniment on the pitch e. At the very end of the opera, Anna turns on Maometto and tells him that, in front of her mother’s tomb, she has sworn her faith to Calbo; then she stabs herself, singing:
Sul cenere materno
Io porsi a lui la mano;
Il cenere materno
Coglie il mio sange ancor
Over my mother’s ashes
I gave him my hand;
Let my mother’s ashes
Now receive my blood.
In a passage that refers explicitly to the earlier phrase, the first two lines are declaimed on the pitch e over a tremolando accompaniment in E major. Hence, the E major to E major progression heard in Erisso’s very first invocation of his wife’s tomb takes on a wider musical and dramatic meaning throughout the opera and duplicates on a small scale the large-scale E major to E major tonality of the entire work.
Some scholars construct schemes in which every tonal choice throughout an opera is assigned deeper meaning. As their arguments become subtler, they attract dissenters, who refuse excessive claims for tonal significance and insist on the difference between more autonomous instrumental music and the complex musical and dramatic web that constitutes opera. Dissenters also point out how often composers modified keys during the composition of an opera or during later revisions, to which believers retort by insisting on the complex functioning of tonality, in which new relationships are brought out when keys are modified.15 But if we cannot imagine a series of keys for which it would be impossible to find an ostensible explanation, if every tonal choice is meaningful, we are in the world of Lake Wobegon, where all the children are above average, or of The Gondoliers: “When every one is somebodee, then no one’s anybody!” It is the principle of non-negation: if there is no way to demonstrate that a particular tonal progression is without dramaturgical meaning, then the dramaturgical significance of tonality is tautological, and why do we waste our breath?
Tonal choices for a composer of Italian opera were not without significance. Sometimes they are embedded in a web of meaning that underlies an entire work, sometimes they have local significance (ensuring a musically viable sequence of keys), sometimes they highlight particular instrumental colors. But tonal choices represent only one set of decisions: many other factors in the composition or performance of an opera can assume greater importance, leading a composer to modify the tonality of a passage or to accept the desire of a singer to do the same.16 In practical terms, however, these modifications often create problems that are insufficiently understood and, in some cases, imperfectly resolved in the opera house.
Two of the worst transitional passages in all of Italian opera, one in Verdi’s La traviata (the progression that links the end of the duet for Violetta and Alfredo in the last act with the “finale ultimo”) and one in Bellini’s Norma (the orchestral introduction to “Casta Diva”), are the result of transpositions gone awry, interventions made after the operas were completed and performed. Some critics have gone so far as to praise these infelicities, as if the composers had turned necessities into masterstrokes. Poppycock: no similar passages exist anywhere else in the works of Verdi or Bellini, precisely because Verdi and whoever made the modification in Norma (possibly Bellini) invented awkward, last-minute solutions to accommodate transpositions made to suit the needs of their singers.
VERDI TRANSPOSES VERDI: LA TRAVIATA
The version of La traviata normally performed today was prepared for a revival of the opera on 6 May 1854 at the Teatro San Benedetto of Venice, slightly more than a year after the first version of the work was launched at the Teatro La Fenice of the same city on 6 March 1853. In the revised opera, the cabaletta of the duet for Violetta and Alfredo, “Gran Dio!... morir sì giovine,” concludes in C major, but the ensuing finale begins in A major, with a c in the orchestral melody. In order to move from C major to A major, Verdi jumps from a one-measure rhythmic figure on c to the same figure on e. That, in turn, functions as the dominant of A major (example 10.2). Julian Budden attempts to explain Verdi’s modifications to the end of the duet by asserting that “in his revision Verdi tightened up the coda giving it additional urgency and harmonic strength and cutting off any possibility of applause by a purely rhythmic transition to the key of the Finale ultimo.”17 But anyone who has ever watched the behavior of an audience bent on cheering at the conclusion of such a piece—and Verdi was always attentive to his audiences—knows that the public pays little attention to niceties of the kind suggested by Budden. Even if making applause more difficult was a byproduct of Verdi’s revision, it could hardly have been his motive. Why, then, did Verdi introduce this unattractive and atypical progression into his score?
Ensuring a comfortable tessitura for the singers who would most likely be engaged to perform La traviata had everything to do with motivating Verdi’s 1854 revisions. As Fabrizio Della Seta has shown in his critical edition of the opera, which presents both the 1853 and 1854 versions, many 1854 modifications affected the part of the father, Germont, whose original tessitura was quite high. While Verdi knew Varesi’s voice well (the singer had been his first Macbeth and Rigoletto), he sought to characterize the father of Alfredo in 1853 through an insistent, imperious declamation, in a register often centered above e, already a high note for the baritone. Before his duet with Violetta, Germont originally introduced the idea of his two children by declaiming “[domanda or] qui de’ suoi due figli!” on high f, descending on the last syllable to e; in the revised version the part descends and the final two notes are an octave lower (the lower f – e). When Violetta at the beginning of the finale expresses her gratitude to die in the arms of those who love her, Germont responds “Che mai dite?” In the original version he leaped to high f for “di-[te]”; in the revision he remained more docilely on middle c.
EXAMPLE 10.2. GIUSEPPE VERDI, LA TRAVIATA, THE TRANSITION FROM THE DUETTO (N. 10), MM. 318–320, TO THE FINALE ULTIMO (N. 11), M. 1, IN THE REVISION OF 1854.
In the often-eliminated cabaletta of his aria, Germont originally sang a cadential phrase that kept him in a high baritonal register, and then returned to the phrase a third still higher, insistently and aggressively repeating f. In the revision, down came the voice. Lowering the register of Germont was by no means the only effect of Verdi’s modifications. In this same cabaletta, for example, he introduced an improved cadential phrase before the final cadenza, avoiding the more obvious symmetry of the original. Yet practical necessity certainly motivated the composer’s intervention. The revised version of the cabaletta is more anodyne in comparison, one of the reasons it has proved so easy to cut: Verdi sacrificed in his revision the highly charged aggression of the original, which was in part dependent on the very high register.
Matters of register are also fundamental to Verdi’s revision of the last-act duet for Violetta and Alfredo. In 1853 their cabaletta (“Gran Dio!... morir sì giovine”) was in D major, whereas in 1854 Verdi brought it down a half tone to C major. In the original key, the transition to the finale functioned smoothly: the duet ended on the tonic chord of D major, while the first melodic note of the finale transformed this d to its enharmonic equivalent, c. In D major, however, the part of Alfredo in the duet lies rather high (by comparison with the tessitura of his aria, for example), and it would have been awkward to change a melody he sings in alternation with Violetta. Yet the revision and transposition also gave Verdi the opportunity to replace a more traditional repeated cadential phrase (as he did in the Germont cabaletta), to eliminate a full stop after the cabaletta for the audience to applau
d (as suggested by Budden), and to avoid insisting on the tonic D (major or minor), a key that returns immediately for the final ensemble (“Prendi, quest’è l’immagine”). Ultimately he may have felt that all these apparent “goods” were sufficient to justify the poor transition that was needed to make them possible.
Verdi’s autograph manuscripts and sketches are filled with examples of transpositions decided upon during the course of composition. In Ernani, the cantabile of Elvira’s cavatina “Ernani, Ernani involami” was originally drafted in A major within the composer’s autograph manuscript, but was transposed up (a rare occurrence) by a half tone to B major before Verdi orchestrated the piece.18 The final ensemble in act 2 of Un ballo in maschera, with its ironic laughter for the conspirators as they realize that the veiled woman Renato is escorting is his own wife, was sketched in B major; by the time Verdi entered the music into the skeleton score that would become his autograph manuscript, it was in B major. The same fate befell Oscar’s invitation to the masked ball, “Di che fulgor, che musiche”: sketched in B major, it ended up in B major. In Stiffelio many pieces were sketched in one key, but were placed in a different key when Verdi entered them into his autograph manuscript.19 Thus, the cabaletta in the scena and aria for Stiffelio was sketched in B major, but ultimately realized in A major; the scene and prayer for Lina was sketched in both E major (the definitive key) and E major;20 the cabaletta of Stankar’s aria was sketched in A major, a tone higher than its definitive collocation, whereas the intensely dramatic solo for Lina in her duet with Stiffelio “Non allo sposo volgermi” is also sketched a tone higher than in the definitive version. In almost every case, the tonality for solo numbers in the completed Stiffelio is lower by a half or a whole step than the tonality in the sketch. The pattern certainly suggests that vocal range played an important, perhaps decisive, role in Verdi’s decisions about keys in Stiffelio.
Divas and Scholars Page 46