In February, the Russian director Dmitri Tcherniakov produced a new staging of Alexander Borodin’s Prince Igor at the Metropolitan Opera. It was last performed there in 1917, sung in Italian. The opera, which was left unfinished at the time of the composer’s sudden death while attending a military ball in 1887, is based on a twelfth-century Slavic epic, The Song of Igor’s Campaign, best known by the English-speaking world in a translation by Vladimir Nabokov. Borodin had left several plot points unresolved: What happens to the young lovers, for example? Other important events, including a major battle in which Igor suffers defeat and is taken captive, were left to the audience’s imagination. Because Borodin died before finishing most of the orchestrations, they were eventually provided by two of his friends, the composers Nikolai Rimsky-Korsakov and Alexander Glazunov. As the musicologists Elena and Tatiana Vereschagina note in the Met program, there is no “authorized, definitive final version” of the opera. It is a jumble of unfinished ideas and marvelous music.

For his reconstruction, Tcherniakov rearranged the order of scenes, cut several passages—including Glazunov’s overture—and inserted music originally sketched by Borodin but never before integrated into the opera, with orchestrations by Pavel Smelkov, including a major new monologue for Igor. He also imagined a new, ambivalent ending, set to a passage from the opera-ballet Mlada, a significant departure from the rousing chorus that had previously closed the opera. It is a bold reappraisal of the work: what has traditionally been considered an “earnestly nationalist” opera whose overriding message was the “endorsement of Russia’s militaristic expansion,” according to the musicologist Richard Taruskin, has been recast as a more complicated piece about the internal struggles of an impotent and tormented hero, played here by the Russian bass-baritone Ildar Abdrazakov. The first image we see in this production is a projection of Igor’s face wearing an anxious expression, followed by the aphorism “To unleash a war is the surest way to escape from one’s self.” Perhaps Tcherniakov was thinking of George W. Bush on the eve of the Iraq War; Vladimir Putin’s incursion into Crimea also comes to mind. Tcherniakov’s Igor is an antihero for our time.

Borodin’s opera is the story of a Slavic prince who goes off to fight a battle against a “barbaric” non-Christian neighbor, a Turkic tribe known as the Polovtsians (or, more often, as the Cumans), led by the cheerfully truculent Khan Konchak. By the second act—in Tcherniakov’s staging, it’s the second half of the first—Igor has suffered a brutal defeat and is being held prisoner in a Polovtsian encampment. His captor turns out to be a gracious host, offering friendship, rich cuisine and the company of wondrous, dark-eyed beauties in return for an alliance. “Like two panthers, we would prowl together [and] feed on the blood of our enemies,” he promises. Igor declines the offer, but the Khan seems unperturbed, good-naturedly inviting him to enjoy a brilliant spectacle by his side: “Bring the captive girls! Let them entertain us with their songs and dances!”

So begins one of the most famous dance suites—or divertissements—in opera: the “Polovtsian Dances,” introduced by a lilting, descending scale on the woodwinds, followed by the floating treble voices of women singing of a land beyond the Caspian Sea, where “the air is filled with languor,” the roses bloom, and the nightingales sing. Borodin’s exotic musical idiom in this scene, heavy with haunting melismas, was partly inspired by musicological sources like Alexandre Christianowitsch’s Esquisse historique de la musique arabe; but mostly, it was the product of his fanciful imagination. Taruskin has described the scene’s sinuous melodies as “the supreme musical expression of nega,” or “tender lassitude,” a characteristic of the exotic East that “emasculates, enslaves, renders passive,” the embodiment of “S-E-X à la russe.” In short, nineteenth-century Orientalism in all its shimmering, titillating glory.

For the opera’s premiere in 1890 at the Mariinsky, the ten or so minutes of choreography that close the act were created by Lev Ivanov, deputy ballet master of the Imperial Theater School. Ivanov was famously musical—among other things, he was responsible for the exquisite lakeside acts of Swan Lake. The suite had considerable success with audiences and critics, although Ivanov was hardly mentioned, an omission that was not unusual at the time. According to one of the company’s régisseurs, “after the act it was not applause, but a veritable ‘roar.’”

* * *

A subsequent version of the dances, based on Ivanov’s design, has survived. This staging, created in 1909 by the Russian choreographer Michel Fokine, was first performed as part of an evening of opera and dance presented by Serge Diaghilev’s Ballets Russes at the Théâtre du Châtelet in Paris. Fokine’s “Polovtsian Dances” became one of the most popular ballets in Diaghilev’s repertory, performed throughout its twenty-year history. (After the demise of the company, it was revived by its various successors, including the Ballet Russe de Monte Carlo, which performed in the United States throughout the 1940s.) It was the model for a whole series of exotic ballets that Diaghilev periodically served up to Parisian audiences hungry for the frisson of the magical East. These included Rimsky-Korsakov’s Schehérazade and Le Coq d’Or as well as Stravinsky’s Firebird and The Rite of Spring.

In fact, Fokine’s “Polovtsian Dances” still exist today (an anomaly, since most ballets disappear); as recently as 2005, the ballet was revived for American Ballet Theatre (ABT) by Frederic Franklin, who had danced it with the Ballet Russe de Monte Carlo. The ballet goes more or less like this: a veiled slave girl in pantaloons and soft slippers enters, followed by several comely companions. The women’s mincing steps are accompanied by the lament of a women’s chorus. Their torsos bend, the better to show off their luxuriant curves, and their arms move in limp arcs, their motions imbued with lassitude and longing. The music quickens and a new melody, filled with skittering chromatic passages for the clarinet, rings out. Clutching a bow, a Polovtsian warrior launches into a series of spinning jumps, circling the women. A troop of men advances in waves, like soldiers in an assault. The orchestral texture thickens as more instruments join in and the action becomes frenzied, with the men enacting a kind of war dance, kicking their legs and launching into barrel turns. Quite suddenly, the music becomes bombastic, bolstered by the percussion and full chorus. “Our Khan is glorious, Khan Konchak!” the singers cry out, as the men smack the floor with their bows and raise their arms in praise. The women provide a sinuous counterpoint to this onslaught of raw masculinity. There it is, the essence of Orientalism: sex and violence.

“Of course, the music is a little bit silly,” the dance historian Lynn Garafola, an expert on the Ballets Russes, quipped recently. What was once titillating and exciting can now look decidedly kitsch, like something out of a musical extravaganza or historical epic from Hollywood’s golden age. (Indeed, among people of a certain age, the opening melody is perhaps best known as the basis for “Stranger in Paradise,” a pseudo-tropical love duet from the musical Kismet, set in an exotic garden filled with parrots, peacocks and tacky Moorish follies à la Metro-Goldwyn-Mayer.) One can see how a serious contemporary opera director, particularly one as brainy as Dmitri Tcherniakov, might want to avoid such connotations. “His whole experience growing up was that everyone left after the ‘Polovtsian Dances,’ after the high point,” the musicologist Simon Morrison, a specialist in Russian music, recently told me. So “he wanted to tone it down, to de-energize it.” Tcherniakov has done exactly that in his production by setting the entire “Polovtsian” act in an enormous field of thigh-high poppies—designed, like all the operas he directs, by him—with no visible open spaces. The scene occurs behind a scrim, further distancing the audience from the action. Unseen pathways allow the characters to cross the stage with care—I saw someone stumble at the dress rehearsal—but the set makes a ballet in the traditional sense impossible.

* * *

Many operas have built-in ballets—a dance interlude was a requisite for a work to be performed at the Paris Opera in the nineteenth century—but directors have seldom taken much care with them. Shoved into a shallow space on the stage or sprinkled like filler into the action, the dancing almost always fails to leave much of an afterimage. But when taken seriously, as in the 2006 Met revival of Ponchielli’s La Gioconda, these little jewels can make quite an impact. On that occasion, the Met invited a rising star, Christopher Wheeldon, to create a fantasy ballet set to the “Dance of the Hours,” made famous by a hippo ballet in the animated movie Fantasia. The dancing takes place, as is often the case, during a party scene. Wheeldon made a delicious little ballet full of brilliant classical steps, small, quick jumps and spins, and sophisticated counterpoint between the ensemble and soloists (“Like the popping of a cork,” he described it). It brought down the house, and Wheeldon’s own dance company, Morphoses, performed it with success as a stand-alone piece.

But even less substantial dances can create an enchanting interlude. Mark Dendy’s dance for Monostatos’s minions in Julie Taymor’s 2004 Magic Flute is delightfully silly, full of gleeful twirling, rope-jumping and acrobatic stunts. Dances have frequently served as a momentary diversion (hence the name, divertissement) or moment of pure pleasure. In seventeenth-century court entertainments, they provided spectacular intermezzi and festive finales, meant to arouse amazement and admiration. “The province of ballets was the more inchoate world of le merveilleux,” explains the dance historian Jennifer Homans in her book Apollo’s Angels, an “expansive arena, with its pagan and Christian resonances and fascination with miracles, magical, and supernatural events.” In addition to arias and choruses, a spectacle like Antonio Cesti’s Il Pomo d’Oro (1668) included ballets depicting the spirits of the air and the denizens of the sea.

But as opera developed, the dances were expected to have a deeper connection to the drama’s themes or action. The final scene in Jean-Baptiste Lully’s Atys (1676), a “tragédie en musique,” is a dance of mourning for the death of its central character. And in Gluck’s Orfeo ed Euridice (1762, revised with additional dances in 1774), the dances are part of the plot, depicting the world Orpheus encounters as he enters the realm of the dead in search of Eurydice. First the Furies block his path, then the spirits of the departed dance to the sound of his lyre. The music for both is sublime. A contemporary of Gluck commented that the choreography—by Gasparo Angiolini, then ballet master of the Vienna court opera house—brought “the dances together with the choruses and the action in such a way as to provide a magnificent model for the future.”

A future that would not materialize: by the nineteenth century, the two art forms were beginning to pull apart. Italian opera tended to dispense with ballets, and though the conventions of French grand opera required the inclusion of one, often in the third act, its main purpose was to please the wealthy patrons of the Jockey Club, who usually arrived just in time to ogle the ballerinas. These ballets had only the most tenuous of connections to the plot. In Verdi’s Paris version of Don Carlos, for example, a scene of treachery that leads to the death of a major character was followed by a bubbly underwater ballet about the search for a perfect pearl. The music is marvelous—George Balanchine later used it for his plotless ballet Ballo della Regina (1978)—but written in a much lighter style than the rest of the opera. Like most of the nineteenth-century opera ballets, it is almost always cut. Last summer, when Verdi’s Les Vêpres Siciliennes was performed in its full French version, including the ballet music, at the Bel Canto at Caramoor festival, I overheard many patrons grousing during intermission about the inclusion of such lightweight, frothy fare.

Yet opera has influenced the evolution of dance. The Giacomo Meyerbeer opera Robert le Diable (1831) included a ballet in the third act for a group of ghostly, depraved nuns who rise up out of their graves in the night to tempt the hero. The ballet created a sensation: its eerie atmosphere, enhanced by gas lighting that emulated “strange and lurid moonlight” (in Homans’s words), led to the vogue for dances about ethereal female spirits, eventually explored in Romantic ballets like La Sylphide and Giselle. The choreographer Filippo Taglioni, who created the nuns’ dance, went on to create the choreography for La Sylphide; his daughter, Marie, was to become one of the great Romantic-era ballerinas, as well as one of the first to dance en pointe, a major development in the history of the form. Balanchine also noted his debt to opera, to which he was exposed in his years as a yeoman opera-ballet choreographer with the Ballets Russes: “From Verdi’s way of dealing with the chorus,” he said, “I learned how to handle the corps de ballet, the ensemble, the soloists—how to make the soloists stand out against the corps de ballet and when to give them time to rest.”

A tension between the two forms remains. “There is a utilitarian, condescending attitude toward dance in the opera/music world—or outright ignorance—and you can find analogous feelings on the other side,” the music critic Alex Ross recently wrote to me in an e-mail exchange on the subject. Partly, this is due to the specialization of the two forms over the course of the nineteenth century, and the resulting theatrical requirements of each. Singing demands extreme focus and a certain amount of stillness; dancers need space and freedom of movement; and music (and text) can convey emotion without any movement at all. Opera directors usually relegate the dancers to small areas on the stage so they won’t get in the way of the story—who hasn’t seen a version of Eugene Onegin or La Traviata in which the waltzing couples are crowded into a shallow area next to a piece of scenery? The opposite can also be true when choreographers are given free rein. In 1914, when the Ballets Russes produced Rimsky-Korsakov’s fantastical opera Le Coq d’Or, the choreographer Fokine had the dancers mime the story while the singers sat offstage. This was not how the composer had imagined it. Mark Morris also pushed the singers to the sidelines in his 1989 version of Purcell’s Dido and Aeneas, using dancers as doubles for their voices. But Morris has also successfully placed singers and dancers side by side, as in his delightful 2008 setting of Purcell’s King Arthur for the New York City Opera. He is planning to do the same in his upcoming Acis and Galatea, which will premiere later this week: “It’s going to be danced throughout,” he told me last year. “The principal singers are onstage with the dancers…everyone occupies the same world.” In Pina Bausch’s rendition of Gluck’s Orpheus and Eurydice (1975), the singers shadow the dancers onstage, with results that are sometimes touching but other times stilted, both for the singers and the dancers. It is difficult to find an ideal balance.

* * *

When the Prince Igor production was first announced, in 2012, the choreographer attached to the project was Alexei Ratmansky, also Russian and a connoisseur and admirer of nineteenth-century spectacle [see Harss, “Ratmansky Takes Manhattan,” September 23, 2009]. As the former director of the Bolshoi Ballet, and later as a freelance choreographer and with ABT, he has created clever, subtly tongue-in-cheek restagings of such over-the-top ballet warhorses as Le Corsaire and Don Quixote, as well as a witty faux-Egyptian divertissement for the Met’s Aida. Ratmansky seemed to be the ideal man for the job: a truly great choreographer with a contemporary perspective on Russian theatrical history. Dance lovers quivered with anticipation—just imagine what he would do with the ersatz exoticism of the Polovtsians! But according to Peter Gelb, the general manager of the Met, after Ratmansky took one look at Tcherniakov’s poppy field, “amicably it was agreed it just wasn’t right for him.” Understandably, Ratmansky could not imagine setting a ballet in narrow passageways that limited the dancers’ trajectories and blocked the audience’s view of their lower bodies.

Ratmansky isn’t the only choreographer to have encountered practical difficulties working in opera. For Crystal Pite, who created the spectacular acrobatic movement for Robert Lepage’s production of Thomas Adès’s The Tempest at the Met, the medium feels “like another planet,” with different priorities and limitations of time and space. Increasingly, in contemporary stagings, directors expect singers to move, so there is the difficulty of working with untrained bodies that are simultaneously engaged in the extremely taxing job of producing beautiful, unconstricted sounds that will be audible in large theaters. “I’m not drawn to it the same way I am to working in theater,” says Pite, “because by their very nature, actors are much more likely to be able to hurl themselves at things…and to not be precious in the way a singer is or needs to be.” Wheeldon faced similar difficulties when working with two different Carmens at the Met. The younger, svelter Elina Garanca was eager to move, but Olga Borodina, who sang some of the performances—gorgeously, I might add—was less willing. “She has a fabulous, creamy voice,” he recalled, but she “did not look good in my choreography…. Is that my fault in the end, or is it the nature of these operas?… The interesting question here is what is more important in opera, the entire production or one starry voice? As an audience member, I prefer artistic embodiment in a performance. I find it hard to believe in opera if it isn’t complete this way. When it’s right, it’s the perfect art form for me. When it’s wrong, it’s just plain tedious and sometimes laughable.”

After Ratmansky bowed out of Prince Igor, Tcherniakov proposed Itzik Galili, an Israeli-born contemporary dancemaker who has worked extensively in Great Britain and the Netherlands, though never in opera. “I’m not an opera guy,” he told me in late January, when he was still developing the choreography in a large studio at the Met. “I’ve never worked on a project before where I didn’t know in advance what the set would look like.” Galili’s choreographic options were, inevitably, circumscribed. Literal allusions to the libretto—maidens, warriors, songs of praise—were eliminated. Instead, thirty or so dancers in pale body paint and white pajama pants (slips for the women) appeared from beneath the poppies, where they had crouched hidden during one of the production’s film projections. With the first note of the women’s chorus—which Tcherniakov had placed in twelve boxes within the theater itself—the dancers popped up and began undulating and twisting their torsos, tracing calligraphic shapes with their arms in the air or touching their hearts, their heads. In the faster passages, the dancers jumped over the hedges and ran in zig-zagging paths across the stage. When the chorus sang “Khan! Khan!”, a few of the dancers formed couples, with the men pushing and pulling the women by their necks and long, loose hair. During the finale, the men began jumping more frantically, reaching and kicking until, with the last ringing note, the dancers fell to the ground, once again concealed beneath the flowers.

“It is an image of feeling,” Galili explains, of “love and hate, caressing and pushing,” in “a paradise full of joy and sublime optimism.” Says Simon Morrison: “He seemed to be riffing on erotic, exotic stereotypes, making the [scene’s] Garden of Eden quality…a little dirtier, more like soft porn than burlesque.” Pragmatic, functional and entirely unspectacular, Galili’s gestural choreography had the advantage of not distracting from the director’s concept of the Polovtsian act as a kind of hallucination conjured by Prince Igor, injured in battle. “The exotic land of the Polovtsians is not a hostile den of barbarians but rather a quasi-utopian alternate reality, where Igor is able to come to terms with his mistakes and examine his life in a new way,” says the program note. In other words, it’s not about the Polovtsians; it’s about Igor. The dancers’ near-flesh-tone white coloring makes them appear like part of the décor, a kind of tableau vivant. On the two occasions I attended, the response to the dances was subdued, despite the stirring music. More than anything, it seemed like a missed opportunity to me, a moment in which the powerful dance value of the music was sacrificed in favor of an overarching idea. The Fokine choreography may have been kitschy, but it was thrilling as movement, thereby reflecting the ecstatic singing of the chorus. Here, the sequence was little more than a curiosity.

In some ways, despite such missed opportunities, the situation of dance in opera is improving. Several recent productions at the Met have integrated intriguing passages of nondance movement, infusing the stage action with a greater sense of poetry and life. In the second movement of Anthony Minghella’s Madama Butterfly, choreographed by Carolyn Choa, a flock of birds—puppets handled by invisible mimes—give physical expression to Cio-Cio-San’s fluttering thoughts as she awaits the return of her American husband, Lieutenant Pinkerton. Another example is the slowly moving clusters, stylized gestures, and hypnotic swaying of the dancers and chorus members—indistinguishable from one another—in last year’s Parsifal, also choreographed by Choa. Here, bodies were used almost as moving sculptures to complement the music and the mood, creating a sense of meter and counterpoint in an otherwise static, desolate staging.

Prince Igor includes a similar moment. In Tcherniakov’s antiheroic conception of the opera, the triumphant ending is replaced by a tentative return to life. The prince’s beleaguered countrymen, played by members of the chorus, begin to pick up the fragments of their destroyed civilization, dragging and carrying rubble across the stage. The scene is reminiscent of Roberto Rossellini’s images of devastation in Germany Year Zero. Their movements are accompanied by a glistening orchestral passage from Mlada titled “The River Don Floods.” No one sings, but there is something deeply stirring about the sight of these bent bodies moving about the stage, heavy with sadness and history. It’s rather like a ballet.