An excerpt from

Divas and Scholars

Performing Italian Opera

Philip Gossett


Every summer Italians find themselves engaged in delicate negotiations on which the happiness of a family depends: Should they spend their vacation at the seaside or in the mountains, mare o monti? Some believe in the virtues of clean air and brisk walks on carefully marked paths far above the heat and humidity of an Italian August. Some prefer sea breezes, swimming in the Mediterranean (less polluted than a decade ago), and quiet rest under an umbrella in one of the symmetrically arranged beach chairs that line Italy’s shores. If papà loves the mountain scenery, mammà looks forward to joining her friends at the sea; if thirteen-year old Emma expects to hit the trail before daybreak, eighteen-year old Massimo wants only to ogle the procession of teen-age beauties in ever briefer bathing attire on their endless walks for his benefit, up and down the hot Adriatic sands. There is no hope of resolving this dispute, only various degrees of compromise. A mountain refuge with an all-night discothèque will help control Massimo’s hormones, while a beach resort with tennis courts will allow Emma to keep in shape. And a range of cultural activities can provide sufficient distraction for all concerned.

Tourism supports the economy of large parts of Italy, not to mention the United States. Since cultural tourism broadens the appeal of a vacation destination, many summer festivals are located in places that compete for tourist dollars. The resulting transformations in local institutions do not win universal approval. Ask long-time residents of Santa Fe, New Mexico, at the foot of the Sangre de Cristo mountains, what they think about the hordes of visitors pouring into their town for the summer Opera Festival, about brand-name chains replacing local stores, about restaurants with international cuisine transforming a dining community once known for regional specialties. The increased cost of real estate in Santa Fe has forced many locals to seek homes in Albuquerque, some sixty miles south. The situation isn’t much different in Pesaro, an Italian beach resort where a three-to-four-month summer season provides resources that sustain the region and its workers for an entire year. But the international clientele of the Rossini Opera Festival in Pesaro each August has also brought shops devoted to designer clothes by Versace and Max Mara, expensive leather goods, and restaurants earning Michelin stars. Local businesses can no longer sustain themselves on the main street, where rents keep rising, and trattorias that used to serve the local population find it harder and harder to survive.

Each festival has a particular repertory niche. Since 1957, Santa Fe, while offering a broad range of opera, has emphasized the works of Richard Strauss and contemporary music. Since 1980, the younger festival in Pesaro has celebrated the works of a native son, Gioachino Rossini, born in the Adriatic city in 1792. Young artist programs in both festivals help ensure the liveliness of the community, and stars of future seasons often begin their careers as apprentices. The early success of each festival was due to the presence of significant artistic figures. Igor Stravinsky became a central participant in the making of the Santa Fe Opera, where The Rake’s Progress was featured in the first festival in 1957. Maurizio Pollini and Claudio Abbado played defining roles in Pesaro. Pollini conducted La donna del lago in 1981, a landmark production in the revival of interest in the Rossini serious operas. Abbado, in what is widely viewed as one of the great musical events in Italy of the past half century, unveiled in 1984 the modern premiere of the reconstructed Il viaggio a Reims, written for the coronation of Charles X in 1825 and long believed to be lost. Both festivals, dogged by their past successes and frequently accused of having lost their way, are urged to reinvent themselves year after year.

The renovated Santa Fe opera house sits on a hill, high above the surrounding landscape, open to the skies on both sides and—when the set permits—through the back of the stage. As with all outdoor theaters, acoustics require careful attention, but the sound is usually glorious and stage designers have access to the latest technology. There is ample room for operas to be performed in repertory. In contrast, although for several years Pesaro tried to alternate two or three operas at the thousand-seat Teatro Rossini—a traditional Italian theater built in the mid-1810s and inaugurated by Rossini himself with a revival of La gazza ladra in 1818—the experience was nerve-wracking. Parts of the set often slept under the stars; fortunately there is little rain in Pesaro during August. The Rossini Opera Festival soon added the Sala Pedrotti, a concert hall in the Conservatory with excellent acoustics and a stage that can accommodate a simple set. In the absence of an orchestra pit, players sit at the level of the audience, exactly the way Italian orchestras performed during the first half of the nineteenth century. Finally, the festival commandeered an indoor sports arena, cleverly transformed. While particularly appropriate for monumental operas, it served equally well for comedies in the hands of imaginative directors and designers, until what is widely regarded as real-estate speculation forced its closing after the season of 2005.

During the summer of 2000 I worked in both Santa Fe and Pesaro. Five nineteenth-century Italian operas were on the boards: in Santa Fe Rossini’s Ermione and Verdi’s Rigoletto; in Pesaro three operas by Rossini: Le Siège de Corinthe, La scala di seta, and La Cenerentola. I participated directly in some productions, advised informally for others, and watched from the sidelines for the rest. This introductory chapter suggests some of the problems we faced in bringing these works before the public. The issues and questions that arose in these particular productions are representative of those that recur in opera houses throughout the world when facing this repertory. No one should take my remarks as being particularly critical of a specific institution: in every opera house or festival highly successful productions rub shoulders with questionable ones, but even within productions that succeed, many issues require further reflection. My examples, in short, are intended to make clear why all of us—scholars, performers, and audiences—need to think harder about what it means to perform Italian opera.


An expectant and knowledgeable public gathered at the Teatro Rossini of Pesaro during the summer of 1987 to witness the first staged performance of Ermione since the spring of 1819. This was one of nine serious operas written by Rossini for Naples between 1815 and 1822, but unlike other Neapolitan works (Otello, Armida, or Zelmira), Ermione was almost entirely unknown. After its unsuccessful premiere on 27 March 1819, and six additional performances that season, the opera disappeared. Except for a fleeting reference to its failure to please, no written records document this inaugural season. Withdrawing the score from the Neapolitan impresario Domenico Barbaja, Rossini is alleged to have said, “You’ll see it again sooner or later, and perhaps then the Neapolitan public will recognize its mistake.” Although he tried to resurrect individual numbers in other operatic contexts, he basically put Ermione away. Asked whether he would allow a French translation, he responded, “No, it is my little Italian Guillaume Tell, and it will not see the light of day until after my death.” We can’t be sure, of course, that he actually said any of these things, but none of them is implausible.

Scholars who had performed Ermione over and over in their imaginations were convinced that it was one of Rossini’s most important serious operas. The long-standing claim that it was “all recitative and declamation,” words Ferdinand Hiller attributed to Rossini in 1855, seemed absurd. There are important scenes of dramatic recitative, accompanied by the orchestra, and intense moments of impassioned declamation, to be sure, but the score abounds in beautiful melody and artfully devised florid passages. Furthermore, the librettist, Andrea Leone Tottola, treats the opera’s literary source, Racine’s Andromaque, with both respect and appropriate freedom. The principal characters are the unhappy children of the Greek heroes of the Trojan War: Pirro, son of Achilles; Oreste, son of Agamemnon; and Ermione, daughter of Menelaus and Helen of Troy; as well as Andromaca, the widow of the Trojan leader, Hector. All four protagonists are locked into an impossible chain of love and hate that leads inevitably to the death of Pirro, the destruction of both Ermione and Andromaca, and the despair of Oreste. In Ermione herself, furthermore, Rossini created one of the most complex characters in the bel canto repertory.

Nonetheless, the 1987 performances of Ermione in Pesaro received a chorus of boos, the worst reception of any opera ever performed at the festival. I had to restrain myself from not joining the chorus. There were many fine things about this Ermione. Marilyn Horne was a superb Andromaca, a relatively small but significant part. As always she turned up for the first rehearsal with the entire role memorized, ornamentation in place, and a fine sense of her character and its relationship to the dramaturgy of the whole. The tenors performing Oreste and Pirro were stalwarts of the Rossini renaissance of the 1980s, Rockwell Blake (known affectionately throughout the operatic world as “Rocky”) and Chris Merritt, both of whom were well received.

But even those three stars could not prevail within a production that worked against the opera and the singers, a conductor who seemed utterly lost, and a prima donna who didn’t belong there. Roberto De Simone, famous for his reinterpretations of the popular dramatic traditions of Naples, was well regarded as an operatic stage director, and his 1985 Pesaro production of Rossini’s farsa, Il signor Bruschino, had been a delight, even if the Gallic wit of the original was transformed into Neapolitan slapstick. For Ermione, however, he invented a stage that severely restricted the space available to singers for entrances, exits, and movement. It hardly mattered, since the prima donna was basically motionless all evening, with two principal gestures: lifting her left arm and pointing a menacing finger to the left, and lifting her right arm and pointing a menacing finger to the right. One can sympathize with the reluctance of stage directors to impose Greek togas on a modern audience, but De Simone’s setting in the Naples of 1819, the post-Napoleonic Bourbon reign in the Kingdom of the Two Sicilies, had little resonance with the story of Ermione.

Still, De Simone’s inoffensive staging would not itself have aroused public ire. That was reserved for the conductor and the prima donna. I knew the production was in trouble from the first orchestral reading, which I attended with the provisional critical edition in hand, in case errors had slipped into our score that required immediate attention. Errors, after all, translate into wasted rehearsal time, a serious matter, given the expense involved. But it soon became apparent that Gustav Kuhn had arrived at this reading without knowing the music. A man of considerable talent and instinct, Kuhn must have assumed he would learn the opera during rehearsals. He began the reading with Ermione’s unusual sinfonia, beating time mechanically, while its series of atypical tempo and meter changes, its use of an off-stage chorus, its complex orchestration, and its structural abnormalities began to unfold. Perhaps realizing how badly he had miscalculated, he appeared uncomfortable, even terrified. And so the rehearsal period went. Rather than providing musical leadership, Kuhn strove to catch up with the rehearsal accompanists and singers, most of whom had been hard at work for months. He never succeeded.

Even this, however painful, could have been forgiven were it not for Monserrat Caballé’s difficulties with the title role. Blame must be shared. How could the management of the festival not have known that Caballé no longer had the vocal skills or the histrionic ability to perform this challenging part? How could Caballé have accepted the role at that point in her career? In her prime she was a singer of great gifts, breathtaking pianissimi, elegant coloratura, fine musicianship. By the late 1980s she had become a caricature of herself. Often she caricatured the operas in which she appeared: during Rossini’s Il viaggio a Reims in London in 1992, she compensated for her inability to sing the role of Madama Cortese by playing the buffoon, mugging and throwing apples at the conductor, Carlo Rizzi. But what could she do with Ermione? Physical ills had restricted her range of movement. Most of all, she could not sing the music and did not seem to understand that it mattered.

It is normal for singers to make small adjustments to help them negotiate passages they find particularly difficult (nineteenth-century musicians called these adjustments puntature). It is quite another thing for a singer to omit or rewrite to the point of unintelligibility large portions of a score. At the climax of Ermione’s “gran scena,” she dispatches the love-sick Oreste to kill Pirro at the altar, where he is exchanging marriage vows with Andromaca. Caballé mortally weakened Rossini’s melodies by radically simplifying them, then by altering the notes at the climactic moment of the melodic line. Why? Simply, as she told me directly, because she found it harder to sing the phrase Rossini had written. In Ermione’s final duet with Oreste, where she berates the confused son of Agamemnon for not understanding that she continues to love the man he has murdered, Ermione is supposed to repeat obsessively, six times, a pattern of four sixteenth notes (with the text “hide yourself from the eyes of living beings, murderer, traitor”); then she must leap fortissimo to a high b♭, one of the highest notes she is asked to sing anywhere in the opera, precisely on the powerful syllable “[tradi]-tor” [traitor]. Caballé reduced the passage to one fleeting four-note pattern, then let loose with a resounding high b♭, as if that were all her fans had come to hear. Having decimated the musical content of the part, she presented herself before the booing public with a copy of the score, pointing to it as if to assert, “I have sung what Rossini wrote.”


Subsequent productions of Ermione have been more successful, including notable revivals in Rome in 1991 and Glyndebourne in 1995 (a fine staging by Graham Vick, with Sir Andrew Davis conducting). The decision by Santa Fe Opera to produce Ermione during the summer of 2000 was a new direction for the theater, which had never been particularly interested in this repertory. As rehearsals proceeded, however, buzz developed around the project, and more and more people (even the founding director of Santa Fe Opera and Strauss champion, John Crosby, no admirer of bel canto opera) made their way into the Tusuque public school, where a mock-up of the stage had been constructed. As in Glyndebourne, Ermione was well received by public and critics (excluding a New York Times observer, whose vocabulary to describe Rossini’s style was limited to “chirpy”).

Yet even in a production where all the pieces fall into place, many controversial decisions need to be taken, affecting music, drama, stage setting, and costumes. The public sees a finished product, but every staging of an Italian opera embodies a series of responses to difficult questions. The opera was performed from the critical edition prepared by Patricia Brauner and myself. But the conductor, Evelino Pidò, who had led an earlier set of performances in Rome, was perplexed by one change between the provisional critical edition he had used in 1991 (before the edition was actually published) and the published critical edition he now had before him. In the chorus that introduces Ermione in the second scene of the opera, an orchestral figure recurs several times. It had been played in the provisional edition by a flute, two oboes, and two clarinets, in octaves, fortissimo, then echoed by a single clarinet, piano; in the printed score the same figure was assigned to four horns in unison, echoed by one solo horn (example 1.1). Not only did Pidò want to know why this had been changed, he was quite reasonably concerned about the ability of four horns to play this figure in unison.

 Gioachino Rossini, Ermione, Coro (N. 2), mm. 1-11
 Example 1.1 Gioachino Rossini, Ermione, Coro (N. 2), mm. 1-11.

The question was fully justified. In Rossini’s own manuscript for this chorus, which we will refer to as his “autograph manuscript” or simply as his “autograph,” the version for four horns was physically altered by the composer himself to that for flute, oboes, and clarinets. If Rossini himself modified the passage, why did we want to return to the version he seems to have canceled? The truth is thatBrauner and I had at first misunderstood the history of this passage. Seeing the composer’s correction, we imagined that he made the change because the passage was too challenging for four horns. But we were wrong. The autograph of this chorus is not with the remainder of Ermione in the Bibliothèque de l’Opéra in Paris, but in Pesaro in the collection of the Fondazione Rossini, with autograph materials for Le Siège de Corinthe and the Italian opera on which it is based, Maometto II. Convinced that Ermione would never be revived, Rossini inserted this chorus into Le Siège de Corinthe in 1826, providing a text in French, writing a new orchestral introduction, and modifying the orchestration of this figure. He made this modification not because the original passage was too difficult to play but because the dramatic situation had changed. In Le Siège de Corinthe, “L’hymen lui donne” is sung by Ismène and the chorus of Turkish women at the beginning of a divertissement (with choruses and ballet), inviting the Greek Pamyra to enjoy the fruits of love by entering into marriage with the Turkish sultan Mahomet. “Dall’Oriente l’astro del giorno,” the same chorus in Ermione, is sung by Cleone and a chorus of Spartan women, armed with bows and arrows, who invite the sorrowing Ermione to join the hunt. That Rossini would use four horns for a hunting chorus, and replace them when the Greek huntresses became Turkish maidens in a divertissement, is perfectly comprehensible. When completing our research into Ermione in order to publish the critical edition, furthermore, we examined all surviving manuscript copies of the opera: every one of them had the version with four horns.

Pidò was right: playing this unison passage at a rapid pace is no easy feat for four horns (not even for nineteenth-century natural horns, without valves), but as rehearsals continued, the performers gained confidence and the sound was splendid. Did they ever make mistakes? Sure, but who has ever heard a performance of an opera (even Siegfried under James Levine) in which a horn has not emitted a few croaks? You write for horn, you take your lumps.

Ermione is a relatively short opera and does not require cutting to reduce its length. Pirro’s aria (which includes sections for the chorus and all major soloists) is long, but so much action and character development takes place that it is difficult to shorten. A lovely “duettino” for Pilade (friend of Oreste) and Fenicio (adviser to Pirro) may not be essential to the drama, but this scene allows some respite for the prima donna between her gran scena and the explosive scena and duet that brings the opera to a close. Just imagine the reaction of the fine Santa Fe Ermione, Alexandrina Pendatchanska, who again sang the role with distinction at New York City Opera in the spring of 2004, had we proposed to cut the duettino! Composers were sensitive to such matters, and many seemingly superfluous passages in Italian opera serve intensely practical needs.

We did make some cuts in the recitative. In his libretto, Tottola frequently provided short lines for handmaidens, lieutenants, and friends, briefly commenting on the action. He may have thought he was imitating an element of French classical tragedy, but when set to music these asides acquire more importance than they can sustain. After Oreste challenges Pirro to turn over Astianatte, the son of Hector and Andromaca, so that the Greeks can put the boy to death, Andromaca and Ermione comment, “How unhappy I am!” and “How will the ingrate respond?” But before Pirro is allowed to erupt into his aria, Tottola provided further text for two aides, Attalo and Fenicio: “How boldly he expresses himself!” and “Heavens! I anticipate Pirro’s anger, I turn cold and am confused!” We eliminated or abbreviated similar passages, carefully adjusting the sequence of chords so that harmonic continuity was preserved.

Working with the stage director Jonathan Miller on Ermione was stimulating and challenging. But why, many asked, did he and his designer, Isabella Bywater, not only dress the characters in mid-nineteenth century costumes but specifically clothe the Trojans as defeated Confederate soldiers and the Greeks as victorious Yankees? What was the point of turning a story of the aftermath of the Trojan War into one visually invoking the American Civil War, while not changing a word of the libretto? For dramas set in classical Greece, Miller believed that modern audiences are too easily distanced from events on stage. The mythic qualities and formal austerity of the Oresteia or Oedipus Rex, however, are different from the dramatic intensity of Ermione. For Miller, the setting allowed an American audience to get closer to these tormented characters. Gone with the Wind, The Little Foxes, Ermione: it is not an unlikely combination. Oreste’s troubled entrance as an ambassador from one Northern general to another, for example, with his aide-de-camp Pilade trying to keep him focused on his task, gives these characters immediacy. Is the transposition necessary? Hardly. Would the setting have been equally relevant to a Czech audience in Prague? Surely not, although the same could be said about most theatrical productions. Did it create moments that were historically implausible? Only if a viewer insisted on interpreting every detail in terms of the Civil War.

Rather than insisting upon the historical moment, Miller successfully offered suggestion and understatement. Every personal interchange among characters and most details of the dramaturgy followed precisely Rossini’s indications (although the Southern belles, not traditionally known for their prowess as hunters, appeared without bows and arrows). This was in no sense a “radical staging,” alienating us from the work to comment critically on it. It was a conventional staging, attentive to the dramatic values of Rossini and aware of the underlying traditions of French classical tragedy. Indeed, Miller boasted that he was surely the only person in the world who had staged not only Ermione but also Andromaque (with the Old Vic). Still, approaches to staging remain among the thorniest issues in performing Italian opera, and another summer 2000 production, Le Siège de Corinthe, as we shall see, posed this problem in a far more outrageous manner.


Butter or ice? It was over twenty years ago, in March 1983, that Riccardo Muti conducted the first performances of the critical edition of Rigoletto—the first volume to appear in The Works of Giuseppe Verdi—at the Vienna Staatsoper. Much of that week is impressed in my memory, although I remain unable to recall whether the sculpture of the hunchback that served as a centerpiece for the after-theater reception was carved from butter or ice.

It is not the only imponderable about those performances. More perplexing is why this event took place in Vienna at all, a city that treats Italian opera with the studied scorn due a former colonial culture. Verdi was all right, as long as his tunes were passed from one hurdy-gurdy to another, but in the temple of great art only Otello and Falstaff were admissible. The common public, with its debased taste, might need to be humored by allowing childish melodramas to be produced, but the less energy put into the process, the better. A critical edition of Rigoletto? Only an intellectual half-wit would devote himself to such a project, unless he were the dupe of voracious music publishers eager to make money on works long out of copyright. In fact, I was accused publicly of being both a half-wit and a dupe at a conference in which the new edition of Rigoletto was presented. Over a coffee mit Schlag in a Viennese café, Rudolf Stephan, who was to become director of the Arnold Schönberg edition, chided me for wasting my time on Italian opera.

Muti prepared the performance with care, and the Rigoletto (Renato Bruson) and the Gilda (Edita Gruberova) were excellent. The orchestra played with elegance, and many details corrected in the new edition emerged with convincing clarity. The production, while not particularly interesting, was free from scandal. Many individuals inside and outside the theater, however, wanted the project to fail, for it was widely considered to be the brainchild of Lorin Maazel, then director of the Staatsoper, who had political problems in Vienna’s complex society. Maazel’s own reaction to the political turmoil was philosophical: during intermission on opening night he compared himself to Mahler, whose shabby treatment by the Viennese is legendary.

Unfortunately the performance was severely marred by the tenor, or rather by a succession of tenors. The originally scheduled Duke of Mantua withdrew several weeks into the rehearsal period. His replacement fell ill and had to bow out altogether after the dress rehearsal, leaving this new production without a Duke. Rather than postponing the opening, Muti chose a tenor then in Vienna who he thought would respond to intensive coaching and whom he might teach the most important elements of the new edition overnight.

He was wrong. The late Franco Bonisolli was one of those Italian tenors with good lungs, a strong sound, but little musical intelligence. He knew just fine how to sing the role of the Duke, thank you, and he was perfectly willing to perform it—on his own terms. I cannot imagine the nature of those coachings between Muti and Bonisolli, to which no one had access, but in the performance Bonisolli was appalling. His Duke was vocally coarse, dramatically vulgar, and indifferent to this production’s efforts to rediscover and interpret Verdi’s original text and music. Add to that his outrage that Muti wanted him to sing the music as written, which meant in particular that he was not permitted to sing and sustain a high b at the conclusion of “La donna è mobile” in the third act. Muti wanted the aria treated as a popular tune, a feather in the breeze, not an excuse for tenor high jinks. That particular battle Muti won, but to no avail, for Bonisolli found many ways to express his distaste for the proceedings. His outrageous behavior with Maddalena conveyed nothing of the eroticism of Verdi’s quartet, but invoked the spirit of the burlesque hall. When the Duke is finally allowed a pianissimo high b as he makes his exit from the opera intoning “La donna è mobile” offstage, Bonisolli bellowed the note at full volume and sat on it. And sat. And sat. He couldn’t see either Bruson’s anxiety as he stared impatiently at the sack that is supposed to contain the Duke’s body or Muti’s furious glares. Many in the audience understood this gesture of defiance and there were pockets of nervous laughter. During the curtain calls, Bonisolli received a solid round of boos, at which point he ostentatiously thrust his rear end at the public. The Viennese press scoffed at what they took to be Maazel’s discomfiture.

In all the turmoil, who could remember that serious artistic aims were at stake, that scholars and performers alike, convinced that Rigoletto was a work of art deserving a better fate than it had been accorded, were attempting to revisit this staple of the repertory? As the first series of performances based on the first critical edition published in The Works of Giuseppe Verdi, it was anything but an auspicious occasion. But it woke an entire generation of Verdi scholars—all of whom in one way or another were to collaborate over the next decades with this major editorial project of the University of Chicago Press and Casa Ricordi—to the realities of the operatic world. And as the hunchback in butter was consumed (or the hunchback in ice melted), we vowed to continue our efforts.

Copyright notice: Excerpt from pages 3-13 of Divas and Scholars: Performing Italian Opera by Philip Gossett, published by the University of Chicago Press. ©2006 by The University of Chicago. All rights reserved. This text may be used and shared in accordance with the fair-use provisions of U.S. copyright law, and it may be archived and redistributed in electronic form, provided that this entire notice, including copyright information, is carried and provided that the University of Chicago Press is notified and no fee is charged for access. Archiving, redistribution, or republication of this text on other terms, in any medium, requires the consent of the University of Chicago Press. (Footnotes and other references included in the book may have been removed from this online version of the text.)

Philip Gossett
Divas and Scholars: Performing Italian Opera
©2006, 704 pages, 92 musical examples
Cloth $35.00 ISBN: 0-226-30482-5

For information on purchasing the book—from bookstores or here online—please go to the webpage for Divas and Scholars.

See also: