Divas and Scholars

Home > Other > Divas and Scholars > Page 43
Divas and Scholars Page 43

by Philip Gossett


  Thus, in applying appoggiaturas in modern performances of Rossini or Donizetti, we must look beyond the mere opportunity to introduce an appoggiatura (two notes at the end of a phrase on the same pitch) and evaluate how the phrase or sub-phrase fits into the broader musical and dramatic discourse. A composer, for example, might set a lengthy speech as a group of shorter phrases, each of which concludes with an appoggiatura opportunity (example 9.6). If the singer were to add an appoggiatura at the end of all sub-phrases ( f; a; c), they would emerge as weighty and separate. If the sub-phrases are declaimed more rapidly, with intervening rests practically ignored, music and drama fly across these breaks and land only at the end of the passage, with an obligatory appoggiatura (b) on the penultimate note of the passage. Modern performers and modern audiences want recitative to move along, and nineteenth-century practice can appropriately be adapted to meet these expectations.

  EXAMPLE 9.6. NOT EVERY APPOGGIATURA OPPORTUNITY OUGHT TO BE EMBRACED.

  I was responsible for preparing and coaching the ornamentation employed in the first production of Rossini’s Otello using the critical edition of the opera, prepared by Michael Collins, at the Rossini Opera Festival during the summer of 1988. With a cast consisting of June Anderson as Desdemona, Chris Merritt as Otello, and Rockwell Blake as Iago, we had available some of the finest Rossini voices of the moment, singers who had no fear of early nineteenth-century vocal technique. My ornamentation and suggested appoggiaturas tried to recapture the style of 1816, as best I could reproduce it. While the production was a great success for all concerned, I gradually became convinced that I had exaggerated, that taking every appropriate opportunity for ornamentation slows things down too much for modern taste.

  I particularly regretted introducing extra cadenzas into the recitative. I did so following the hint of Rossini’s own manuscripts of ornamentation and of the practice of Manuel García, as described in his son’s treatise on singing. García, after all, in addition to having been the first Almaviva in Il barbiere di Siviglia, was Rossini’s original Otello. So on making his entrance in the opening scene, Otello sings first in recitative (no Verdian “Esultate” here), while laying before the Doge of Venice the arms and flags of the naval forces he has conquered. Example 9.7 reproduces Rossini’s setting of this phrase and one of the three ornamented versions that the younger García prints in his treatise.35 Not only did I adopt something akin to García’s cadenza here, but I suggested similar interventions throughout the opera at points of dramatic emphasis. Each one sounded plausible individually, but the overall effect was leaden. When this stunning Pier Luigi Pizzi production was brought to the Lyric Opera of Chicago in 1992, I tried to remove most of these added recitative cadenzas, but Chris Merritt rebelled: he rather liked the vocal outbursts. Having learned that such ornamentation could be historically and stylistically justified, he only wanted more of the same!

  EXAMPLE 9.7. GIOACHINO ROSSINI, OTELLO, RECITATIVO DOPO L’INTRODUZIONE (N. 1), OTELLO PHRASE (MM. 13–15) AND AN ORNAMENTED VERSION FROM THE GARCÍA TREATISE.

  VOCAL ORNAMENTATION:

  CADENZAS AND VARIATIONS IN ROSSINI

  While the tasteful use of appoggiaturas in recitative is obligatory and simple to learn, other interventions in recitative are dependent upon individual preference and taste. The music of a recitative, after all, is usually constructed to follow details of the text (in versi sciolti), and rarely are words or musical phrases repeated. In arias, duets, or ensembles employing versi lirici, on the other hand, where the musical structure is more regular, the situation changes radically. Indeed, we must bear in mind the structural precepts that underlie most fully developed arias by Rossini, Bellini, and Donizetti (and many ensembles), for these precepts were cultivated in part to permit the introduction of cadenzas and variations. One of the major problems with Verdi’s early and middle operas is that he continued to use these forms while attitudes toward ornamentation were changing, hence depriving the forms of the interpretive dimension that justified their shape.

  In a standard scena and aria, as we have seen, accompanied recitative prepares a slow cantabile, which may include a partial or complete repetition of the principal thematic material.36 The cantabile is usually followed by a short tempo di mezzo (middle section), whose musical shape and contents depend on its dramatic purpose, on the presence or absence of other characters or chorus, on whether it motivates action or simply provides an interlude. This tempo di mezzo prepares the cabaletta, with its principal melodic period (the cabaletta theme), a short transition, a repetition (whole or partial) of the theme, and a series of cadential phrases of decreasing length, generally with internal repetitions.

  Cadenzas

  Embedded in Rossini’s formal language are opportunities for singers to insert cadenzas. On some occasions they are included in the composer’s autograph manuscripts (hence in vocal scores), but mostly they are left to the discretion of the singer. And nowhere is discretion more important. In these moments of soloistic abandon, the orchestra, the dramatic action, the harmonic motion come to a halt as the singer demonstrates his or her artistry. If the cadenza is poorly shaped musically, if its expressive qualities do not draw sustenance from the drama, if it seems unduly prolonged or too short, or if the singer fails to exploit successfully his or her vocal means or reaches beyond them, what should be a transcendental moment crashes to earth. Then again, how do we judge the success or failure of a cadenza? The same notes may seem magical when emerging from one throat and disappointing from another. Some members of the audience may be so outraged by not hearing a cadenza that a recorded performance has imprinted in their ear as “the text” that they are incapable of judging the artistry of a new one. And the “heavenly length” of a cadenza to one critic may seem boringly protracted to another.37

  These are not problems exclusive to Italian opera. Cadenzas, whether written by the composer or interpolated, play a fundamental role in eighteenth-century concertos, chamber music, and solo sonatas, as well as operas. Instrumentalists, however, tend to know both what is notated in the score and the origin of cadenzas traditionally introduced into a concerto movement. Some proudly write new cadenzas; others, like Robert Levin, vaunt their ability to improvise. In Italian opera, singers too often reproduce what they have learned from teachers or heard on favorite recordings, and only strong intervention can modify their habits. Yet one of the most interesting productions with which I have worked in recent years was a Barbiere di Siviglia at Santa Fe Opera during the summer of 1994, conducted by Evelino Pidò and directed by Francesca Zambello. At the beginning of the rehearsal period, singers, conductor, and musicologist agreed that all cadenzas and variations would be newly devised. While this approach would hardly be viable on every occasion, its freshness communicated itself to the audience assembled in the Sangre de Cristo mountains, and the performances were an unqualified success. Dwayne Croft, debuting in the role of Figaro, was loath to give up a standard interpolated high g in “Largo al factotum,” but we found compensating fireworks elsewhere: no one hearing this production could have thought for a moment that Croft didn’t have high notes to burn.

  Much contemporary evidence pertains to the use of cadenzas. As with appoggiaturas, though, taste has changed markedly. I have yet to find a conductor willing to permit singers to interpolate cadenzas of the length considered normal in the nineteenth century: they are more likely to shorten cadenzas already in the score. As we have seen, Verdi wrote a wonderful cadenza a due for Gilda and the Duke in their first-act duet in Rigoletto, but its heart is frequently ripped away. When Muti opened the La Scala season on 7 December 2000, anticipating the centenary of Verdi’s death and using the critical edition of Il trovatore, his Count di Luna, Leo Nucci, boasted that he was going to sing Verdi’s entire cadenza for “Il balen del suo sorriso,” as if he deserved some kind of medal. Few admit that they have become unaccustomed to hearing long cadenzas. Instead they invent excuses about singers going flat, as if singing Verdi in tune were
harder than singing Alban Berg, Benjamin Britten, or Luigi Nono. Only when a character is not fully lucid do we seem to countenance a long cadenza: hence our willingness to accept Lucia’s interpolated fantasy with flute, an accretion to the opera dating from more than forty years after Donizetti’s death.38

  Rossini either wrote cadenzas directly in the text of a piece or provided cadenza opportunities by notating progressions such as the one given in example 9.8 from the end of Tancredi’s opening period in his second-act duet with Amenaide, while adding either a fermata or the indication a piacere. Reproducing a cadenza opportunity in performance come scritto reveals ignorance of Rossini’s style. One can disagree about how to fill in the cadence, not about whether or not it is meant to be filled in. Positioned at the end of musical sections or at points of musical division that prepare new sections, cadenzas usually elaborate a small number of harmonic functions. In one of the most typical functions, as in this example, the harmonic bass moves from the dominant (or fifth) degree of the scale to the tonic degree. The dominant degree supports first a tonic triad in second inversion (what musicians call a six-four chord), then a dominant sonority (V). At the end of the cadenza, voice and orchestra resolve together on the tonic (I). Depending on the length of the cadenza, the singer can muse vocally over the six-four chord or over both. Rossini himself suggested the cadenza given in Example 9.9 for Giuditta Pasta to use in the Tancredi duet.39 It carefully follows the shape of the cadenza opportunity. It begins in mid-register on b (supported by the six-four chord), leaps an octave higher, and then employs a repeated pattern of three notes to descend to a low b, with the dominant harmony. Now the voice leaps to high a where an arpeggio on the dominant seventh chord leads again from high to low, then back to midregister. Finally, the voice concludes one , the tonic, as in the original score. Pasta gets to display her leaps to registral extremes, her control of short, repeated patterns, her arpeggios, her staccato articulation, and her ability to color the entire cadenza with appropriate dynamic levels, all to express her rage at what she believes to be Amenaide’s infidelity.

  EXAMPLE 9.8. GIOACHINO ROSSINI, TANCREDI, DUETTO (N. 14), CADENTIAL FORMULA (MM. 25–26).

  EXAMPLE 9.9. GIOACHINO ROSSINI, TANCREDI, DUETTO (N. 14), CADENTIAL FORMULA AS FILLED IN BY THE COMPOSER FOR GIUDITTA PASTA (MM. 25–26).

  Even when Rossini wrote a cadenza in full, a singer need not feel restricted to those pitches, although the composer’s own cadenzas are usually of superlative quality. In Guillaume Tell, for example, Rossini provided a sober and elegant cadenza, built entirely on the dominant harmony, for Mathilde to conclude her romance, “Sombre fôret.” Although there is no particular reason why a singer would not want to use this cadenza, Cinti-Damoreau, an artist deeply knowledgeable about Rossini’s style, offers five alternatives in her treatise on singing, employing a wide range of different techniques: simple or chromatic scales (up and down), arpeggios, complex harmonic patterns, dissonant lower and upper neighbors, turns and trills, vocal leaps, in a dizzying array of alternatives. While Rossini’s original expresses only a single harmony, several of Cinti-Damoreau’s cadenzas complicate the harmonic progression considerably. Most performers will probably prefer the more sober cadenza by Rossini, but experimenting with alternatives is perfectly justified.40

  As we were preparing the critical edition of Rossini’s La gazza ladra in the late 1970s, I recall glancing distractedly one day at a volume of the Strenna marchigiana of 1891, one of many periodicals perched casually at that time on the shelves of the Fondazione Rossini. This one caught my fancy, because its cover sported the picture of a pretty girl in peasant dress. Absentmindedly I leafed through its pages, when to my astonishment I found the facsimile of a full set of variations and cadenzas in Rossini’s hand for the cavatina of Ninetta, the heroine of La gazza ladra, “Di piacer mi balza il cor.” The manuscript had belonged to Giuseppina Vitali, a young singer who performed this cavatina at one of Rossini’s soirées musicales in 1866. Although we were in final proofs, there was a blank page at the end of the first appendix, and with a little pushing and prodding we incorporated this material into the critical edition without renumbering hundreds of pages. The variations are fascinating, and the cadenza Rossini wrote in 1866 to prepare the reprise of the main theme in the cantabile is harmonically audacious with respect to his original.41 But, of course, his taste and the taste of his singers and audience had changed considerably between 1817 and 1866. And our collective modern taste has changed between 1967 and 2006, and will certainly change again before 2037. While there are basic principles so deeply imbedded in the musical style of Rossini’s operas that we ignore them at the risk of destroying the character of his music, there is great latitude about how to interpret those principles.

  Ornamental Variations

  Late nineteenth-century musicians (and the twentieth-century progeny that perpetuated their attitudes) approached Rossini’s scores with scissors in hand. Insensitive to the stylistic context of his music, they complained about repetition, without being aware that for Rossini repetition—of short phrases, entire sections, or even cadential formulas—provided singers with an opportunity for introducing variations. Rossini and his singers inherited techniques that underscored eighteenth-century operatic tradition, where the da capo aria, the heart of Metastasian and post-Metastasian practice, was constructed to allow the singer to offer an ornamental variation of the first section (A) when it was repeated, da capo (from the top), after a brief contrasting section (B).

  Da capo forms pertaining to an entire musical number almost never occur in nineteenth-century works, however, because composers and their librettists sought to incorporate greater dramatic action within each musical number, a structural approach to musical dramaturgy that remained fundamental to Italian opera well into the second half of the century. Musical numbers are therefore constructed of contrasting sections, with different tempos and employing different poetic or musical meters. Repetition, along with the ornamental variations it engenders, occurs within individual sections of a multipartite musical number. In order to employ variations appropriately, one must be sensitive to the structural principles underlying the art of Rossini and his followers. Contemporary sources are helpful in suggesting where to introduce variations and where to withhold them. While there may be internal repetitions within a lyrical period (since these periods are often constructed in four phrases, AA'BA''), for example, the entire melody should be heard before ornamentation is applied. That is what the overwhelming majority of contemporary models demonstrate, and these should serve as our basic frame of reference.42 The sin of overornamenting Italian opera is just as bad as the sin of not ornamenting at all.

  To treat all circumstances in which ornamental variations are appropriate in Italian opera from the first four decades of the nineteenth century would require a treatise devoted to that question alone. Some general principles can be articulated for solo arias, principles that can be applied to similar situations in ensembles. First, we need to distinguish principal lyrical sections (the cantabile and concluding cabaletta) from sections that link or introduce these lyrical sections (a tempo di mezzo or other subsidiary passages). Then, within the cantabile and cabaletta, we need to focus on principal melodic periods, without neglecting transitional material or cadential phrases.

  While cantabile designs are so varied that attempts to embody them in a few ideal types are necessarily reductive, two structural approaches in Rossini’s arias particularly favor ornamental variations. In the first, the principal melody is lyrical throughout. After it is heard in its entirety, a transition (perhaps with a cadenza) leads to a full or partial repetition. It is this repetition that requires an ornamental variation. The Ninetta cavatina from La gazza ladra is a perfect example. Example 9.10 gives the principal melody of the cantabile and Rossini’s suggestion for ornaments from the Japanese autograph manuscript.43 The theme is tuneful, an eight-measure period (four measures leading to the dominant at m. 45, four similar mea
sures remaining in the tonic); it is followed by a repeated cadential figure of two measures, featuring a horn solo, and two concluding orchestral measures (these cadences are not included in the example). Rossini ornaments only the main theme, with its simple chordal accompaniment, not the cadences, where active orchestral figurations discourage intervention in the vocal line. His variation respects the basic shape of the melody but elaborates it rhythmically and melodically. The original melody sat between a low e and a high f , while the variation extends the range down to b below middle c and up to the high b, Ninetta’s actual range throughout the opera. The variation, in short, does not exaggerate her tessitura; it simply exploits it more fully in the context of her cavatina. Of course, this variation should be attempted only by a singer confident of handling with precision the octave-and-a-half leap from an e in the upper register to a low b and the two-octave leap back up to the high b, all of it, most appropriately, to the word “balza” (“My heart jumps for joy!”). The singer also needs to produce both a clean arpeggio down at the end of the first full measure of the theme and that lovely combination of triplets and an upward scale (from d to a) for “ge-[nitor].” The penultimate measure, in which the original figurations on the second and third beats (an eighth note and two sixteenths) are made more florid (a sixteenth note and six thirty-seconds), looks worse than it is: in the manuscript for Vitali the composer actually wrote col canto, instructing the orchestra (whose part is strictly accompanimental here) to follow what should be a rhythmically free vocal rendition. Attempting to perform this ornamental variation in strict time would be profoundly antimusical. Indeed, the figuration in m. 42 is grammatically wrong: there are too many notes in the measure. And yet Rossini’s notation is perfectly appropriate to the musical situation.

 

‹ Prev